Consumer Buying Behaviour On Glue Stick

Copyright is owned by the Author of the thesis.
Permission is given for

a copy to be downloaded by an individual for the purpose of research and
private study only. The thesis may not be reproduced elsewhere without
the permission of the Author.
CONSUMER RESEARCH AND PRODUCT OPTIMIZATION

FOR GLUE STICK DEVELOPMENT IN THAILAND
A thesis
presented in partial fulfilment of
the requirements for the degree of
Doctor of Philosophy in Product Development
at Massey University
HATHAIRAT UAPHITHAK
1994
ii
ABSTRACT
This study, on the development of a new glue stick product based on tapioca starch for
Thai consumers, had a major emphasis on the use of consumer input in the product
optimization. At the beginning of the development process consumers identified the
problems of existing products and generated the important a ttributes of glue stick
products. Then product prototypes were made using mixture experimental designs and
quantitative relationships between the ingredients and the product attributes were
determined. For evaluating the sensory attributes of prototypes, a trained panel was
employed during the development of the first prototypes and then a consumer panel
in the second prototype development. Consumers not only evaluated the product
attributes of the prototype products using line scales but also indicated their ideal
product attribute levels. Physical attributes of the products were also measured.
Mul tiple regression was used to generate the empirical equations showing the
relationships between the ingredients and the product attributes.
These linear relationships were then used to develop the constraints for a linear
programming model. The consumer ideal product profile as well as the sensory profiles
of the commercial products were employed to create upper and lower acceptable limi ts
of the a ttributes' constraints. The raw materials and the physical properties were also
included in the linear programming model. Acceptability maximization and cost
minimization were used to generate the optimum formulations. The prototype products
from these formulations were tested by a small consumer panel to select the one with
h ighest acceptability. A pilot scale plant was designed and built and then a small
quantity of the final formulation was produced.
The final product, from the successful pilot scale production, was tested in a horne-use
test by 1 08 students and 64 office workers in Bangkok. The consumers evaluated the
performance of the developed product in comparison with their 'usual brand'. The
results from the consumer testing showed that the developed glue stick was generally
accepted by the target consumers. However, some improvements of the product in
terms of colour, aroma and packaging are still necessary.
1Il
The product could be made conunercially in Thailand in a simple plant using a closed
stainless steel vessel with steam jacket, condenser and central anchor type mixer.
Suitable packaging equipment is needed in order to maintain the high temperature of
the mix while d ischarging the glue mixture into moulds. The product should be able
to compete with the glue sticks already in the Thai market since it has the distinctive
feature of adjustability and could be sold at a lower price.
iv
PREFACE
Tapioca (cassava) has been regarded as one of the world's most important crops that
can be used for human consumption, for animal feed
and in various industries.
Thailand is the largest exporter of tapioca products with the total export about 86
percent of the world's exports in 1 989 (TTTA, 1990). The exports are made in two main
forms, tapioca chips and pellets for use as animal feed and tapioca flour for human and
industry consumption. The European Community is the dominant tapioca importer
taking about 59 percent of the total export. Tapioca chips and pellets have been used
extensively as one of in the most important feed ingredients in the animal feed industry
in Europe for over 20 years: Starch another important product from tapioca is used in
both food and non-food industries. Tapioca flour is employed as a raw material by
industries making such products as soup, candy, pudding, sausages, bread, ice-cream,
noodles and vermicelli. It is used as a binder by the pharmaceu tical industry in making
pills. Moreover due to its saccharification property, tapioca flour is used for
manufacturing food seasonings, glucose, fructose, soft drinks and canned food. Tapioca
flour is also used in the production of adhesives, paper, textiles, plywood and alcohol.
In the adhesive industries, tapioca flour qm also be used in the form of dextrin or
modified starch. Adhesives from tapioca are mainly used in the corrugated board
industry which manufactures vast amounts of board to be used for cartons, boxes and
containers. Tapioca adhesive have also been used in laminated paper board,
remoistening gums, wall paper and horne use. Although tapioca starch has permanent
use in some starch using industries, there is still the need to expand its use in various
ways to be able to compete with other starches. Most research has been done on
improving formulation and techniques in processing of adhesives used in paper and
board industries. Nevertheless there is the demand for the development of an adhesive
product for consumers' use particularly in Thailand.
Glue stick was considered to be a suitable product to be developed for Thai consumers.
Glue stick is a consumer product which has become popular recently owing to its ease
of use, convenience and good performance compared with other glue products. It was
found that all the glue sticks commercially available in the Thai market are imported
v
from other countries: Germany, Japan, Korea and China. These glue sticks are based on
synthetic polymers, mainly polyvinyl pyrrolidone which is one of the factors that
contribute to the h igh cost of the product. It was decided that if tapioca starch could be
used to replace polyvinyl pyrrolidone in glue stick formulation the cost of product
could be reduced. This would also increase the use of tapioca starch in the non-food
product"
area.
This project was done partly in New Zealand where the product was developed in the
laboratory scale and partly in Thailand where the product prototypes were tested with
Thai consumers, a pilot plant built and an optimum product was developed and tested
with Thai consumers.
vi
ACKNOWLEDGEMENTS
I would like to express my deepest gratitude to my supervisor, Professor
Mary D. Earle,
for her valuable guidance, assistance and encouragement provided throughout the
project.
This research was financially sponsored by the Royal Thai Government Scholarship, I
express my sincere appreciation for this assistance.
I wish also to thank my co-supervisors, Professor Vichai Haruthaithanasan, Product

Development Department, Kasetsart UniverSity, Bangkok, who helped with the pilot
plant production and consumer research in Thailand; and Dr. Lai Pai Wan for her
useful suggestions.
The research has been successfully completed by either the direct or the indirect
cooperation of many people in New Zealand and in Thailand. In particular I would like
to thank -
"'The staff of the Consumer Technology Department, Massey

University, especially Mr. John O'Connor fo.r his technical
assistance.
'" The staff of the Food Technology Department Massey University.
'" The staff and students of the Product Development Department,
Kasetsart University, Bangkok, for their help and cooperation
during the research in Thailand, especially, Mr Chavalit.
'" The students of the Consumer Technology Department who took
part in the trained sensory panel.
'" The Thai consumers who participated in the product testing.
'" Mrs. K.B.L. McDonald for her kindness during the time
I stayed
with her.
Finally, I would like to thank my parents, Mr. Virat and Mrs. Kosum Uaphithak, my
sister, Thaveerat, and my friend, Burin, for their continual moral support during the
course of my study.
vii
TABLE OF CONTENTS
ABSTRACT
PREFACE
ACKNOWLEDGEMENTS
TABLE OF CONTENTS
LIST OF TABLES
LIST OF FIGURES
LIST OF APPENDICES
ii
iv
vi
vii
xiv
xviii
xix
CHAPTERS
1
INTRODUCTION
PRODUCT OPTIMIZATION SYSTEM
1.1
1.1.1 What is Product Optimization
1.1.2 Product Optimization Procedures
1.2
CONSUMER E V ALUATION D U R I N G P RODUCT
OPTIMIZATION
1.2.1 Initial Development of Product
1.2.2 Product Formulation Development
1.2.3 Formal Optimization Study
1.2.4 Final Consumer Testing of the Product
1.3
DEVELOPMENT OF A GLUE STICK
1.3.1 Why Glue Stick was Chosen as the Model Product
1.3.2 Development of Glue Sticks in the Past
AIM AND OBJECTIVES
1.4
LITERATURE REVIEW
2.1
PRODUCT OPTIMIZATION TECHNIQUES
2.1.1 Linear Programming in Product Optimization
2.1.2 Use of Sensory Attribute Constraints in Linear
Programming Models
2.2
USE OF CONSUMER SURVEY AND CONSUMER TESTING IN
PRODUCT OPTIMIZATION
2.2.1 Consumer Survey
2.2.2 Consumer Testing
2.2.3 Identification of Important Product Attributes
2.2.4 Testing of Product Acceptability and Important Attributes
of Product with Consumers
2.2.5 Consumer Research and Consumer Testing in Glue Stick
Optimization
2.3
PRODUCT PROFILE
2.3.1 Use of Line Scale in Product Profile
2.3.2 Ideal Product Profile
2.3.3 Use of Product Profile in Product Optimization
DEVELOPMENT OF MODEL USING PRODUCT PROFILE
2.4
FlNA L PRODUCT TESTING
2.5
2.6
GLUE STICK
2.6.1
Raw Materials in Glue Stick Formulation
2.6.2 Adhesive
2.6.3 Gel Forming or Shape-Giving Substance
1
2
2
2
3
3
4
5
5
6
7
7
9
11
11
12
12
13
13
15
16
18
20
20
22
22
23
24
25
26
27
28
29
viii
2.7
2.8
2.9
3
2.6.4 Solvent
2.6.5 Plasticizers and Emulsifiers
2.6.6 Fillers, Antifoaming Agents, Colours, Perfumes
2.6.7 Formu lation, Processing and Packaging
USE OF STARCHES IN GLUE STICK
2.7.1 Uses of Starch and its Derivatives in Adhesive
2.7.2 Use of Tapioca Starch in Thailand
2.7.3 Use of Starches in Glue Stick Formulation
TESTING OF GLUE STICK PROPERTIES
2.8.1 Stick characteristics
2.8.2 Amount of glue applied on paper
2.8.3 Initial adhesion
2.8.4 Adhesion a fter drying
CONCLUSIONS
PROJECT METHODS
3.1
CONSUMER STUDY
3.1.1 Consumer Survey
3.1.2 Consumer Testing
SENSORY TESTING OF GLUE STICKS BY TRAINED SENSORY
3.2
PANEL
3.2.1 Panel Selection
3.2.2 Development of Descriptive Terms
3.2.3 Training of the Panel
3.2.4 Use of Trained Panel
3.3
PHYSICAL TESTING
3.3.1 Choosing the Physical Testing Methods for the Product
3.3.2 Hardness
3.3.3 Melting Point
3.3.5 Water activity
3.3.6 Amount of Glue Applied per Area
3.3.7 Open time
3.3.8 Peel Strength of the Adhesive Bond
INGREDIENTS AND THEIR PROPERTIES
3.4
3.4.1 Starches
3.4.2 Polyvinyl Pyrrolidone
3.4.3 Glycerol
3.4.4 Stearic Acid
3.4.5 Sodium Hydroxide
3.4.6 Glyceryl Monostearate
3.4.7 Brij 35
3.4.8 Dextrin
3.4.9 Casein
3.5
PROCESSING OF GLUE STICK
3.5.1 Equipment
3.5.2 Method
EXPERIMENTAL PLANS
3.6
DATA PROCESSING METHOD
3.7
29
30
30
31
32
32
34
35
35
36
36
37
39
40
41
42
42
42
42
43
44
44
47
49
49
50
51
51
52
53
54
56
56
56
57
57
57
57
57
58
58
58
58
60
61
63
ix
4
CONSUMER STUDY
4.1
AIM AND OBJECTIVES
4.2
METHOD OF CONSUMER STUDY
4.2.1 Products of Interest
4.2.2 Stages of Survey in New Zealand and Thailand
4.2.3 Method Used in Generation of Important Attributes
4.2.4 Method of Measuring Importance of Attributes
4.2.5 Questionnaires for Consumer Survey in New Zealand and
Thailand
4.2.6 Survey Method
DA TA PROCESSING
4.3
PATTERN OF GLUE USAGE
4.4
ATTRIBUTES OF GLUE STICKS WHICH SHOULD BE
4.5
IMPROVED
IMPORTANT ATTRIBUTES OF GLUE STICK AND SELF4.6
ADHERING NOTE PAD
4.7
IMPORTANCE MEASURE OF GLUE PRODUCT ATTRIBUTES
4.7.1 Importance of Attributes Measured by Open-ended
Elicitation Method with Panel in New Zealand
4.7.2 Importance of Attributes Measured by Direct-rating
Method with Panels in New Zealand and Thailand
4.7.3 Importance of Attributes of Glue Products with Panel
4.8
ACCEPTABILITY OF ATTRIBUTES OF GLUE PRODUCTS
4.9
BUYING INTENTION, SIZE AND PRICE FROM PANEL IN
THAILAND
4.10
DISCUSSION
4.10.1 Pattern of Glue Usage
4.10.2 Importance of Attributes
4.10.3 Differences between Cultures in Glue Usage
4.10.4 Comparison between Glue Products
4.10.5 Consumers Intention of Buying between the Two
Products - Permanent and Temporary Bonding Glue
Sticks
4.10.6 Development of Product Concept
4.10.7 Using the Consumer Study for Development of Glue Stick
Products
4.11
CONCLUSIONS
64
64
65
65
65
66
66
67
68
68
68
71
73
75
.
76
77
79
80
81
83
83
84
85
85
86
86
87
88
I NIT IA L DE VELO P ME NT ST U DY O N GL UE STIC K FOR M U LAT IO N
5.1
5.2
AI M A N D OBJECTI VE S
SELECTIO N OF P RELI MI NA RY FO R M ULATIO N
5.2.1
5.2.2
5.2.3
5.2.4
5.3
5.4
5.5
5.6
Criteria for Selection of Preliminary Formulation

Preliminary Experimentation
Dextrin Experimentation
Casein Experimentation
A TT RlBUTE S OF CO M ME RCIAL GL UE STIC K S

DE VELO P ME NT O F GL UE ST IC K PROTOTY PE S
P RO PE RTIE S O F GL UE STIC K PROTOTY PE S
CO M PA RI SO N
BETWEE N
PROTOTY PE
P RO D UCT S
AND
103
107
107
108
CO M ME RCIAL P ROD UCT S
5.7
RELATIO N S HI P S BETWEE N GL UE STIC K ATTR IB UTE S
5.7.1
5.7.2
5.7.3
Correlations bet ween Glue Stick Physical Attributes

Correlations bet ween Glue Stick Sensory Attributes
Relationships bet ween Physical Attributes and Sensory
110
Attributes
5.8
RELATIO N S HI P S BETWEE N I N GREDIE NT S A N D GL UE ST IC K
5.9
ATT RIBUTE S
DI SC U S SIO N
5.9.1
5.9.2
111
115
115
Effect of Ingredients on Attributes of Glue Sticks

The Suitable Levels of Basic Ingredients for Starch Based
115
116
116
117
Glue Stick
5.9.3
5.9.4
5.10
6
Selection of Physical Attribute Measures

Selection of Sensory Attributes for Consumers Testing
CO NCL U SIO N S
117
117
118
118
119
120
PRO DUCT FO R M ULATIO N DE VELO P ME NT WIT H CO N S U MER
6.1
6.2
6.3
6.4
AIM AND OBJECTIVES
EX PE RI ME NTAL METHOD S
6.2.1 Basic formulation
Experimental Design
PHY SICAL TE STI NG
6.2.2
SE N SO RY
E VA L UATIO N
BY
CO N S U MER
PA NE L
IN
120
121
121
122
122
123
125
125
THAILA N D
6.4.1
6.4.2
6.4.3
6.4.4
6.4.5
6.5
6.6
6.7
Consumer Panel in Thailand

Sample Preparation
Questionnaire
Preparation for the Consumer Panel Test
Consumer Testing Procedure
DATA P ROCE S S I NG A N D A NA LY S I S O F RE S U LT S
P HY SICAL ATT RIB UTE S OF T HE PROTOTY PE S
SE N SO RY ATT RIB UTE S OF T HE P ROTOTY PE S E VAL UATE D
127
127
130
BY CO N S U ME R
6.7.1
6.7.2
6.8
Affective Attributes
Sensory Attributes
RE SPO N SE OF CO N SU MER S
IN
DIFFE RE NT
MAR KET
131
SEG ME NT S
6.9
89
89
89
90
91
92
94
99
100
102
IDEAL RATIO SCO RE S A ND LO G OF I DEAL RATIO SCORE S

OF THE SE N SO RY ATT RIB UTE S O F T HE P ROTOTY PE S
6.9.1
Jdeal
Ratio
Prototypes
Scores of the
133
Sensory Attributes of the
133
xi
Log of Ideal Ratio Scores of the Attributes of the
Prototypes
RELATIONSHIPS BETWEEN ATTRIBUTES OF THE
PROTOTYPES
6.10.1 Physical Attributes
6.10.2 Sensory Attributes
6.10.3 Acceptability and Sensory Attributes
6.10.4 Sensory Attributes and Physical Attributes
RELATIONSHIPS BETWEEN THE COMPONENTS AND
ATTRIBUTES OF THE PROTOTYPES
DISCUSSION
CONCLUSIONS
6.9.2
6.10
6.11
6.12
6.13
7
AIM AND OBJECTIVES
7.1
OPTIMIZA TION PLANNING
7.2
DEVELOPMENT OF LINEAR PROGRAMMING MODEL
7.3
7.3.1 Objective Functions
7.3.2 Decision Variables
7.3.3 Constraints
7.4
DEVELOPMENT OF OPTIMUM FORMULATIONS FROM
LINEAR PROGRAMMING
7.4.1 Cost Minimization
7.4.2 Acceptability Maximization
7.5
SELECTION OF FORMULATIONS OBTAINED FROM LINEAR
PROGRAMMING
7.6
PHYSICAL ATIRIBUTES OF GLUE STICKS DEVELOPED FROM
LINEAR PROGRAMMING MODELS
7.7
SENSORY TESTING
7.7.1 Selection of Samples for Sensory Testing
7.7.2 Sample Preparation
7.7.3 Method of Sensory Testing
7.7.5 Optimum Formulation
DISCUSSION
7.8
7.8.1 Comparison of Optimum Formulation with Commercial
Products and Consumers' Ideal Product
7.8.2 Comparison between Cost minimization and Acceptability
Maximization
7.8.3 Use of Linear Programming in Glue Stick Optimization
7.9
CONCLUSIONS
PRODUCTION TRIAL
8.1
AIM AND OBJECTIVES
8.2
PILOT SCALE PRODUCTION
8.2.1 Equipment
8.2.2 Formulation
8.2.3 Processing
8.2.4 Testing of the Finished Product
YIELD AND COSTS IN GLUE STICK PRODUCTION
8.3
8.4
CHANGES OF PRODUCT ATTRIBUTES DURING 30 DAY
133
134
134
135
136
136
137
140
143
145
145
146
147
147
148
149
151
152
154
156
158
159
159
159
159
160
161
163
163
164
165
166
168
168
170
170
170
171
172
172
xu
8.5
8.6
8.7
8.8
8.9
8.10
9
10
STORAGE
PRODUCT VARIAl3IUTY BETWEEN BATCHES
PRODUCT VARIABILITY BETWEEN STICKS FROM DIFFERENT
LOCATIONS
PRODUCT VARIABILITY BETWEEN LABORATORY SCALE
AND PILOT SCALE SAMPLES
COMPARISON OF NEW GLUE STICK WITH A COMMERCIAL
GLUE STICK
DISCUSSION
CONCLUSIONS
FINAL CONSUMER TESTING OF PRODUCT

9.1
AIM AND OBJECTIVES
9.2
METHOD OF CONSUMER TESTING
9.2.1 Selection of Consumers
9.2.2 Sample Preparation
9.2.3 Questionnaire
9.2.4 Testing of the Product
9.3
DATA PROCESSING AND ANALYSIS OF RESULTS
9.4
PATIERN OF GLUE STICK USAGE OF THE CONSUMERS
9.4.1 Frequency of Glue Stick Usage
9.4.2 Source of the Glue Stick Consumers Used
9.4.3 Brand Name of Glue Stick Consumers Used
9.4.4 Size of Glue Stick Consumers Used
9.5
CONSUMER ACCEPTABILITY OF THE NEW PRODUCT
9.6
CONSUMER BUYING INTENTION
9.7
PRICE OF THE NEW GLUE STICK
9.8
THE CHARACTERISTICS OF THE DEVELOPED GLUE STICK
WHICH ATIRACTED CONSUMERS
9.9
DISCUSSION
9.9.1 Reaction of the Target Consumers towards the Developed
Glue Stick
9.9.2 Selection of Market Segment
9.9.3 Improvement of the Developed Product
9.9.4 Future Research
9.4.5 Position of the Developed Product in the Market
CONCLUSIONS
9.10
DISCUSSION AND CONCLUSIONS
INTRODUCTION
10.1
10.2
OPTIMIZATION PROCESS
10.3
THE USE OF CONSUMER INPUT IN THE OPTIMIZATION
PROCESS
10.3.1 Benefits of Using Consumers throughout the Optimization
Process
10.3.2 Differences between Thai and New Zealand Consumers
10.3.3 Selection of the Important Attributes
10.3.4 Relationship of Acceptability to Purchase Intention and
Price to Buy
10.3.5 Comparison between Consumer Panel and Trained
Panel
174
176
177
178
179
180
181
182
182
183
183
183
183
184
184
184
184
185
186
186
187
188
189
190
191
191
192
192
193
193
193
195
195
195
198
198
200
201
201
202
xiii
10.4
10.5
10.6
10.7
10.8
USE OF EXPERIMENTAL DESIGNS IN PRODUCT

OPTIMIZATION
10.4.1 Selection of the Preliminary Formulation
10.4.2 Use of Ing redient/ Product Attribute Relationship Models
in Final Product Optimization
GLUE STICK FORMULATION AND ATIRIBUTES
10.5.1 Use of Tapioca Starch in Glue Stick
10.5.2 Sensory Properties of Glue Stick
10.5.3 Relationships of Sensory Properties and Physical Tests
10.5.4 Properties of Developed Glue Stick Compared with
Commercial Prod ucts
RELA TIONSHI P BETWEEN MARKET POSITIONING AND
PRODUCT ATIRIBUTES
RECOMMENDATIONS FOR FUTURE WORK
CONCLUSIONS
203
203
204
205
205
206
206
206
207
208
209
REFERENCES
211
APPENDICES
220
xiv
LIST OF TABLES
3.1
Six components extreme vertices
62
4.1
Users of each glue product
69
4.2
Frequency of glue product usage
70
4.3
Purposes of glue product usage
71
4.4
Attributes of glue stick product needing improvement
72
4.5
Attributes of self-adhering note pad needing improvement
73
4.6
The attributes of a glue stick product
74
4.7
Important attributes
consumers
of self-adhering
note
pads
given
by
the
75
4.8
Importance of glue stick attributes from open-ended elicitation method
76
4.9
Importance of self-adhering note pad attributes
77
4.10
I mportance of glue stick product attributes from panel in New

Zealand
78
I mportance of self-adhering note pads a ttributes from panel in

New Zealand
78
4.12
Importance of glue product attributes from Thailand panel
79
4.13
Importance of permanent bonding glue stick attributes
80
4.14
Importance of temporary bonding glue stick attributes
80
4.15
Acceptability of glue products by panel in Thailand
81
4.16
Buying intention of consumers towards the new product
82
4.17
Usage time for one stick of glue
82
4.18
Price of.glue stick consumer prepared to pay
83
5.1
Formulations used in preliminary experimentation
91
5.2
Properties of glue stick samples from preliminary experimentation
92
5.3
Experimentation plan for dextrin experiment
93
5.4
Physical attributes of dextrin glue sticks
93
4.11
xv
Empirical equa tions showing relationships between ingredients and
attributes of glue sticks from dextrin experiment
94
5.6
Experimentation plan for casein experiment
95
5.7
Empirical equations showing relationships between ingredients and

physical attributes from casein experiment
97
Empi rical equations showing relationships between ingredients and

sensory attributes of glue stick from casein experiment
98
5.9
Physical attribu tes of the commercial glue sticks
99
5.10
Sensory attribu tes of the commercial glue sticks
100
5.11
Experimental plan for prototype development
101
5.12
Physical attributes of the glue stick samples from constrained mixture

design experiment
102
Sensory a ttributes of the glue stick samples from mixture design

experiment
103
Comparison between physical attributes of prototype products and

commercial products
104
Comparison of sensory attributes of prototype products and commercial

produc
105
Correlations between physical attributes of glue stick samples from

mixture design experiment
107
Correlations between sensory attributes of glue stick samples from

108
Correlations between physical and sensory attributes of glue stick

samples from mixture design experiment
111
5.19
Correlations between ingredients and output variables
112
5.20
'
Empirical equations showing relationship between ingredients and
physical attributes
113

sensory attributes
114
6.1
Experimental plan for mixture design with three components
120
6.2
Physical attributes of the prototypes
126
6.3
Comparison between physical attribu tes of the prototypes and

commercial products
5.5
5.8
5.13
5.14
5.15
5.16
5.17
5.18
5.21
126
xvi
6.4
Acceptability, purchase intention and price to buy of the prototypes
128
6.5
Sensory attribu tes of the prototypes
130
6.6
Ideal scores of the sensory attributes of the prototypes
131
6.7
Correlations between sensory scores obtained from three groups of

consumers
132
Ideal ratio scores of the sensory attributes of the prototypes (Original

data)
133
Ideal ratio scores of the sensory a ttributes of the prototypes

(Transformed data)
134
6.10
Correlations between physical attributes of glue stick samples
135
6.11
Correlations between sensory attributes of glue stick samples
135
6.12
Correlation between acceptability and sensory attributes
136
6.13
Correlations between sensory attributes and physical attributes
137
6.14
Correlations between the components and the attribu tes
138
6.15
Empirical equations showing relationships between glue stick a ttributes

and glue stick components
139
7.1
Costs of the ingredients used in glue stick formulation
147
7.2
Costs of the components used in glue stick formulation
148
7.3
Decision variables in glue stick formulation
149
7.4
Glue stick optimum formulations and their expected attribute levels

obtained from cost minimization
154

obtained from acceptability maximization
156
7.6
Selected formulations obtained from linear programming
157
7.7
Physical attributes of the glue sticks developed from linear programming

models
158
Sensory attributes of the glue sticks developed from linear programming

model
160
Sensory attribute ideal ratio scores of the glue sticks developed from
linear programming
161
6.8
6.9
7.5
7.8
7.9
xvii
Attributes of optimum glue stick compared with commercial products
and ideal product
163
Comparison between using cost minimization and acceptability

maximization
164
8.1
Levels of glue stick ingredients used in the pilot scale production
171
8.2
Costs incurred in producing an 8 g glue stick
174
8.3
Physical attributes of glue stick at di fferent storage times
175
8.4
Sensory attribu tes of glue sticks at different storage time
175
8.5
Physical attribu tes of glue sticks from different batches
176
8.6
Sensory attributes of glue sticks from different batches
176
8.7
Physical attributes of glue sticks from d ifferent locations on the tray
177
8.8
Sensory attribu tes of glue sticks from different locations on the tray
177
8.9
Physical attributes of pilot scale and laboratory scale samples
178
8.10
Sensory attributes of pilot scale and laboratory scale samples
179
8.11
Sensory attribu tes of new glue stick and a commercial glue stick
180
9.1
Frequency of Glue Stick Usage
185
9.2
Source of Glue Stick Consumers Used
185
9.3
Brand Name of Glue Stick Consumer Normally Used
186
9.4
Size of Glue Stick Consumers Used
186
9.5
Comparison between the New Product and the Product Consumer

C urrently Used
187
Comparison between the developed glue stick and the glue stick
consumers normally used
189
Price that consumers who said they would buy the product
recommended for the developed glue stick compared with their glue
stick
189
Characteristics which could attract consumers to buy or try the

developed gl ue stick product
190
Environmental friendly ranking by consumers
191
7.10
7.11
9.6
9.7
9.8
9.9
xviii
LIST OF FIGURES
3.1
Stages of glue stick optimization
41
3.2
Descriptive terms and their definitions used by the trained panel
48
3.3
Hardness testing diagram
50
3.4
Peel strength testing diagram
55
3.5
Laboratory scale equipment
59
5.1
Product profiles of glue stick prototypes compared with commercial

product attributes
106
5.2
Diagram Showing Relationships between Physical Attributes
107
5.3
Diagram Showing Relationships between Sensory Attributes
109
6.1
Complete mixture space showing feasible area for experimentation
119
6.2
Acceptability of the prototypes
129
6.3
Product profiles of the prototypes compared with ideal product profile
141
7.1
Constraints for glue stick linear programming model
150
7.2
Model (1) for cost minimization
152
7.3
Model (2) for acceptability maximization
155
7.4
Positions of the optimum formulations on the mixture space
158
7.5
Product profile of optimum formulation compared with the ideal

profile
162
8.1
Pilot scale equipment
169
10.1
Steps in Glue Stick Optimization
196
10.2
Comparison of prototypes from different steps during product

optimization
197
--
--
------
xix
LIST OF APPENDICES
3.1
Prescreening Questionnaire for a Glue Stick Panel
220
3.2
Descriptive Terms Developed by the Trained Panel
221
3.3
Questionnaire for Glue Stick Testing by Trained Panel
222
4.1
Questionnaire for Consumer Survey in New Zealand
224
4.2
Questionnaire for Consumer Survey in Thailand
226
5.1
Correlations between Ingredients and Physical Attributes of Glue Stick
from Dextrin Experiment
236
5.2
Physical Attributes of Glue Sticks from Casein Experiment
236
5.3
Correlations between Ingredients and Physical and Sensory Attributes of

Glue Stick from Casein Experiment
236
5.4
Commercial Glue Sticks
237
6.1
Questionnaire for Glue Stick Testing by Consumer Panel
238
6.2
Sensory Attributes Scores of the Prototypes Obtained from Consu mer

Testing
241
Idea l Ratio Scores of the Prototypes' Sensory Attributes (Original

Data )
248
Log of Ideal Ratio Scores of the Prototypes' Sensory Attributes

(Transformed data )
253
9.1
Questionnaire for Final Product Testing
258
9.2
Summary of Results from Fina l Product Testing
262
9.3
Cross-ta bulation of 'Reason Consumers Though that the Developed Glue

Stick was Better or Worse than their Glue Stick' by Acceptability
267
6.3
6.4
9.4
Cross-tabula tion of 'Comparison of Developed Glue Stick with Glue Stick

Consumers Normally Used' by , Purchase Intention'
267
1
CHAPTER 1
INTRODUCTION
The product development process involves the application of different techniques:

consumer research, processing development, product formulation, engineering and
marketing, to develop new products or to improve the existing products already in the
market. The characteristics of the developed product depend on several factors: target
consumers, the company policy and strategy, introduction of new raw materials,
competitors, government regu lations, culture of the society and availability of
technology. Consumers' needs always change with time, products have to be improved
to meet these new needs. The company may want to reduce the formulation cost or to
introduce new raw materials. A new government regulation may specify the level of
some ingredients in the formulation or ban ingredients currently used. All these factors
have to be taken into account in order to obtain a successful improved or new product.
In general, the main steps of the systematic product development process as described
by Earle (1989) are:
* Development of project aim
* Setting of project constraints
* Product idea genera tion
* Product idea screening
* Development of the product concept and product design specification

* Development of the prototype product
* Development of a suitable process or manufacturing method
* Testing of the product
* Decision on and development of a market plan and type of product launch

* Evalu a tion of the probable product success
* Launching the product
However, this study did not involve the whole process of product development. It
emphaSized mainly the development of prototype products in order to obtain an
optimum product prototype through product optimization.
2
1.1
PRODUCT OPTIMIZATION SYSTEM
The optimization of all aspects of a product is the goal in product development. In

product optimization, the consumer study, the systematic experimentation, and the
optimization technique are combined to generate an optimum product.
1.1.1
What is Product Optimization
Product optimization is defined as a comprehensive and efficient approach which can

reduce the time used for developing the product and simultaneously develop a highly
acceptable product for consumers while minimizing unit cost. The ultimate goal of
consumer oriented product optimization is to maximize consumer acceptance of a
product, given a fixed set of ingredients. The outcome of the process should support
key decisions evolving from company policy and strategy (Gordon and Norback, 1985).
Hence, a product optimization method considers all the constraints and objectives of the
project, and will help the product developer to obtain an optimum formulation.
1.1.2
Product Optimization Procedures
Optimization requires a well-defined procedure. Several techniques have been

developed for product optimization (Fishken, 1983; Sidel and Stone, 1983; Giovanni,
1983; Schutz, 1983; Moskowitz, 1983 and 1987). Al though these methods differ, they all
include the following steps (Lagrange and Norback, 1987):
,.. An initial development study in which prototypes are developed and critical
input and output variables are identified.
,.. Product formulation development which includes the determination of the

levels (or ranges) of the ingredients and processing variables for use in the
subsequent optimization study. The evaluation of product attribu tes is
conducted and the model showing relationships between inputs and output
variables are generated at this stage.
,.. Formal optimization study, with or without constraints, that includes consumer
testing, data analysis, reformulation and implementation.
,.. Final consumer testing of the product. In final product testing, the reaction of
consumers toward the new product is measured. The product may be tested
by i tself or against the competitive products.
1.2
CONSUMER EVALUATION DURING PRODUCT OPTIMIZATION
Consumers can provide useful information to be used during product optimization.

Consumers can be involved in selecting product ideas and product concepts. The
information regarding the product to be developed and the reaction of consumers
toward the existing products are used at the initial stage of product optimization to
guide the direction of further study. Evaluation of the prototype products by consumers
has been employed extensively in optimization of mainly food products. Moskowitz
(1 987) described product evaluation by a consumer panel during development of a
cereal product. Consumer panels also took part in opti mization of a fish cake product
(Moskowitz and Rabino, 1 983). However, the consumer panel has not been used in a
great extent in optimization of non-food products.
1 .2.1
Initial Development of Product
In the early stages of product optimization, product developers need some guidance on
how to develop the product to gain consumers' acceptance. Strengths and weaknesses
of competitors' products have to be identified by generating profiles of competing
products and comparing them with an ideal product profile. Attributes that play an
important role in consumers' acceptance and buying intentions have to be identified.
Several techniques have been used to elicit importance of attributes from consumers
(Alpert, 1971; Moskowitz and Chandler, 1978; Heeler et al., 1979; Jaccard et al., 1986).
Consumer panels have also taken part in generating profiles of existing products
( Rabino and Moskowitz, 1980). Normally a number of products available in the market
are purchased and tested with consumers. Products are usually tested on a blind basis
4
so there is no brand name effect on product rating. I f prototypes have been developed
from some basic formulations in the laboratory, they can be tested together with the
products on the market. This will help the researcher to select the suitable basic
formulation to be used in further study. Sometimes, consumers may be asked to
indicate their ideal product on the same scale they evaluate each attribute of the
samples, so the direction that the product should be improved can be obtained
(Moskowitz, 1 984 and Cooper et al., 1989).
Moskowitz and Jacob (1988) repor ted using a consumer panel to evaluate in-market
products at the beginning of a Mexican entree development. A consumer panel was also
used to generate profiles of competitive products at the early stages of development of
a hand lotion (Rabino and Moskowitz, 1980). In optimization of a rice bran oil based
hand cream, Uaphithak (1990) also used consumers to develop an ideal product profile
which could be used as a guidance to optimize the product.
1 .2.2
Product Formulation Development
From the initial development stage, the direction in which the product should be
developed is identified; in the development of product formulation, the product
developer has to decide which basic formulation should be used to generate the product
prototypes. Since consumers determine the success or failure of a product, it is logical
to collect consumers' reactions to various formulation as early as possible in the product
optimization process.
When new products are developed, another goal of the optimization procedure is to
provide the product developer with information regarding the effect of input variables:
ingredients, process conditions, and storage conditions, on various attributes of the
prototypes: physical properties, acceptability, sensory attributes, liking of attributes, and
shelf life. Sensory attributes of each prototype are measured by a sensory panel, either
a trained panel or a consumer panel. Consumer panels have been used in evaluation
of both food product and non-food product prototypes: a dry-powdered soft drink
(Griffin and Stauffer, 1990), skin lotion (Moskowitz, 1 982). Relationships between
product attributes and input variables are investigated and empirical models are
generated for further use in the optimization stage. Moskowitz (1982) generated model
-------
--
relationships between purchase intent and sensory attributes of lotion which were later
used to develop an optimum product.
1 .2 .3
Formal Optimization Study
In the product optimization stage, the researcher has to identify the appropriate levels
of the i mportant input variables which produce a product that achieves optimum
acceptance from consumers. The relationships between input variables and consumer
perceptions are applied with optimization techniques such as linear programming,
mul tiple regression, response surface methodology, or gradient search, to generate an
optimum product.
After the optimum formulation is obtained, the product has to be made on the pilot
scale level in order to make sure that the product can be made on a larger production
scale and that a suitable processing method and processing equipment are developed.
Scaling up products from laboratory scale to pilot plant and from pilot plant to large
scale production will often result in product changes. Sensory panels can be used to
detect and identify the differences between the desired product and the product from
the large-scale production.
1 .2.4
Final Consumer Testing of the Product
The product obtained from the optimization process is tested with consumers in order
to measure product acceptance in terms of acceptabili ty, purchase intention and price
to buy. Penny et al. (1 972) reported on the interrelationship between product rating on
an intention-to-buy scale and the company's assessed product success or failure in the
marketplace. They found that actual purchasing increased with scale position - there
.
were higher percentage of consumers who subsequently bought the test product from
the group who rated the product high on purchase intention compared with the group
who rated the product low on purchase intention.
In final product testing, researchers may wish to compare their new product with a
current formula or with the market leader. By comparison of the developed product
with the commercial products it is possible to identify unique advertising points of view
6
about the product. Marketing information such as price, brand name and packaging can
be included in product testing. Product acceptance and purchase intention is normally
evaluated at this stage.
Even though consumer purchase intention is influenced by factors other than the
product itself, product optimization is still necessary because if a product contains the
attributes which sa tisfy consumers' needs i t will have more chance of competing
successfully with other products on the market.
Sensory testing has been applied to non-food products in particular personal care
products, for example for more than 20 years in the form of deodorancy testing.
Recently descriptive analysis techniques have been applied to lotion (Gibson, 1973;
Moskowitz, 1982), soap, antiperspirants, shampoos and other personal care products.
Dethmers and Boomsma (1989) described how traditional sensory methods had been
modified and expanded to accommodate the multidimensional aspects of home care
products such as: air care, floor care, furniture care, laundry care and insect control
product. However, the science of sensory evaluation and the use of standard methods
and procedures for conducting controlled sensory evaluations of products and materials
are relatively new to the glue product i ndustry especially to glue sticks.
This research was conducted in order to demonstrate the use of optimization techniques
together with consumer panel evaluation in the optimization of a non-food product.
Clue stick was selected as a product model to be studied.
1 .3
DEVELOPMENT OF A GLUE STICK
Although the product optimization procedure has been used for optimization of food
products and skin care products (Rabino and Moskowi tz, 1980; and Uaphithak, 1 990),
there has been no evidence of using consumer evaluation in glue product optimization.
This present research was conducted to ascertain if the consumer panel can be
considered as an important tool in consumer-oriented product optimization, of glue
sticks.
7
1 .3 . 1
Why Glue Stick was Chosen as the Model Product
Product optimization techniques have been used mainly for food products (Norback
and Evans, 1 983; Lagrange and Norback, 1987; Moskowitz, 1987 and 1 988). Linear
programming one of the methods for optimization was used by Rust ( 1976) in
formulating preblended meats, Dano (1972) in formulating ice cream. Bender et al.
(1982) used this method in selecting a formulation for a least cost mayonnaise. Although
Kavanagh ( 1978) used linear programming in paint and resin development, there is little
evidence of the use of consumer data in the optimization of non-food products in the
area other than skin care products. Glue stick is a product which is new to the Thai
market. The price is high compared with other glue products with similar performance.
The high cost of raw material is one of the factors which contributes to this high price.
Replacing of some synthetic raw materials with natural raw materials not only could
produce a cheaper product but also create an environmental friendly and a safe
product.
No publications were found on using
op timi z a tion
techniques and consumer panels i n
develo p men t o f a g l u e s l ick. I-f ence, a g l ue s t i ck w a s considered
suitable
as
a product
model in this project.

1 .3.2
Development of Glue Sticks in the Past
Glue stick was invented in the late 1 950s. It was developed by adhesive companies. The
first glue stick was made from hard resins: hydrogenated terpene, polymerized
terpenes; soft rubbers: hydrocarbon polymers and butyl or natural latex; and soft waxes:
Japan wax, stearic acid, paraffins and spermaceti (Brennan, 1966). Resins contribute
largely to the 'adhesiveness' of the adhesive and control the softening point in
combination with the waxes used. Rubbers give a spinning action when the adhesive
is stroked rapidly across a sheet of paper and also add tack, especially in combination
with the resin. The most important contribution of the soft wax is to maintain the
applied film of adhesive in a softened and activated state for a period of time. They also
plasticize and enhance the tack of the resins.
Glue stick was improved using a salt of an aliphatic carboxylic acid such as sodium
8
stearate as the gel forming agent to improve rub off attributes, and to reduce processing
time and temperature (Muszik and Dierichs, 1971). Later, wax acid (Hoechst waxes) was
used as the gel forming agent so that alkali-sensitive substances could be included in
the formulation; acid-reacting active compounds may be added (Adhesive Tape Limited,
1 974 and Werke H.u.M. Fischer, G.m.b.H., 1974). There were also further developments
on removable bonding (Gollub et a1., 1987) and temporary bonding glue sticks (Palm,
1 989) for specific application using an alkali metal or ammonium salt of an aliphatic
carboxylic acid as gel-forming agent and a mixture of carboxylated -alkyl ester of acrylic
and polyethylene as adhesive.
1.3.3
Use of Starch in Glue Stick
Although the technology of making glue stick is not new as it was developed in the late
50s, this product is considered as 'new' by Thai consumers. The use of a glue stick
instead of other glue products is increasing owing to its ease of use and cleanliness.
Despite the fact that a glue stick has some advantages over other glue products, it is
still not widely used because the price is higher than the prices of other glue products
such as liquid glue, PVC glue and paste glue.
Price (Bahts f
Glue stick
1 5-20 (8 g)
Liquid Glue
1 0 (50 ml)
PVC glue
15 (60 ml)
Paste glue
5 (25 g)
1 N.Z. dollar
15 Bahts
1 U.s. dollar
25 Bahts
The price of glue stick in the Thai market is abou t 15 to 20 Bahts for a 8 g stick. In order
to compete with other glue products available in the market, it is necessary to bring the
price down either by using low price raw materials or by reducing the cost of
production.
The adhesive component in glue stick is mainly synthetic polymers which makes the
glue stick expensive. It was con.s idered that if a natural polymer was used to replace
9
all or some part of synthetic polymers it would decrease the price of the final product.
Apart from reducing the price of glue stick, the benefit from using natural polymers is
that they are safe for consumers especially children because natural polymers are non
toxic and they are edible so it will be less harmful if children should eat them by
accident. Natural polymers are a renewable resource so there will be no problem with
regards to the continuity of supply of raw materials.
Natural polymers which can be used in glue sticks are starch, starch derivatives,
cellulose derivatives, gum arabic. (Muszik and Dierichs, 1971; Werke H.u.M. Fischer,
G.m.b.H., 1 974 and Adhesive Tape Limi ted, 1974). Tapioca (cassava) starch was
considered as a suitable replacement for an adhesive substance in glue stick formulation
for the Thai market since it is produced in a large amount in Thailand . There was no
evidence of the use of tapioca starch in glue stick formulation. Therefore, this study was
conducted to explore the use of tapioca starch in the gl ue stick formulation.
1 .4
AIM AND OBJECTIVES
It was found that only physical testing was employed for measuring the product
attributes of adhesives (Ando, and Yamazaki, 1 974; and Pletcher and Wong, 1978).
There has been no evidence of using consumer testing in glue stick development nor
report of the use of either experimental designs or optimization techniques for this kind
of product.
Since glue stick is a consumer product, it is necessary that the product is accepted by
the consumers. Although there are many reports on the use of consumer data in
product development, no information was found on how the data generated from
consumers can be used efficiently in product optimization, particularly in a non-food
system.
Hence, the aim of this thesis was to study the use of consumer data as inputs in the
optimization procedure for a non-food product, especially in development of a tapioca
starch based glue stick product.
10
The objectives were to:
". Study consumers' reactions toward glue product usage particularly glue sticks
and generate a list of important attributes of glue sticks required by consumers.
". Explore the use of tapioca starch as a raw material in glue sticks and select a
suitable formulation for a starch based glue stick.
". Develop product test measures and study the effect of glue stick components
on physical and sensory attributes.
,.. Generate an ideal product profi le for glue sticks through a consumer panel.
,.. Use linear programming to generate an optimum tapioca starch based glue
stick which was highly acceptable by the Thai consumers and could compete
with the products already in the market.
,.. Confirm that the formulation and process developed in the laboratory could
be transferred into i ndustrial production.
,.. Measure product acceptability in a home-use test.
11
CHAPTER 2
LITERA TURE REVIEW
The goal of this project was to optimize a glue stick product for the Thai market. Since
glue stick is a consumer oriented product, use of consumer input during the product
opti mization procedure is discussed in this chapter. The suitable product optimization
techniques are i nvesti gated. Basic ingredients used in commercial glue stick
formulations and their properties are reviewed. The use of starches in glue products
specifically in glue stick will also be mentioned.
2.1
PRODUCT OPTIMIZATION TECHNIQUES
The major methods which have been used in optimization research are: linear
programming, multiple regression, response surface methodology, and gradient search.
Linear programming is a mathematical technique used to determine the optimum

allocation of a limited supply of resources, subject to certain constraints, to either
maximise or minimise a specified objective (Nicklin, 1979). In product formulation
problems, the resources are normally raw ma terials, the constraints are based on the
functional properties of th.e final product and the objective is usually cost minimization
or consumer acceptability maximization for a consumer oriented product.
Multiple regression involves the development of models that relate dependent and
independent variables. The regression models obtained contribute significantly to the
development of a data base concerning the relationships between the product
characteristics, acceptance (output variables) and the raw material composition and also
processing variables (input variables).
The response surface methodology systematically uses quantitative data from

appropriate experimental designs to determine and simultaneously solve multivariate
equations (Giovanni, 1983). The equations can be graphically represented as response
12
surfaces or contour diagrams. The information can be used t o describe how the
variables affect the response, determine the interrelationships between variables, and
describe the combined effect of all variables on the response. This method has been
used in the optimization of fragrances where both sensory and image properties were
optimized (Williams et a1., 1 992).
The gradient search method is a procedure for moving toward a point on a continuous
surface where all the partial derivatives of a function are at or very near zero. The
gradient points in the direction where the size of the set of acceptors is increasing the
fastest. The step size tells the technologist how far to move in the gradient direction
(Lagrange and Norback, 1987).
2.1.1
Linear Programming in Product Optimization
Linear programming is often considered to be a limited tool because of i ts assumptions

of linearity and infinite divisibility. However, the technique is much more flexible than
is generally recognized (Bender et aI., 1982). Linear programming has been successfully
used in optimization of food products for more than 30 years. IBM ( 1966) reported use
of linear programming in blended meat formulations. This method was also successfully
used in formulation of ice-cream (Dano, 1 972), mayonnaise (Bender et al., 1982), fresh
turkey bratwurst (Beausire et aI., 1 988) and corn-based snack (Almeida-Dominguez et
al., 1 990). Linear programming has also been applied in non-food product formulation:
paint and resin (Kavanagh, 1 978), light duty liquid detergent formulation (Chan and
Kavanagh, 1 988) and hand cream (Uaphithak, 1990).
2 . 1 .2
Use of Sensory Attribute Constraints in Linear Programming Models
In optimization of a consumer oriented product, sensory attributes of product have to

be taken into account in order to obtain a product which is accepted by the target
consumers. Use of sensory attributes in product optimization has been developed for
many years, however, the applications were only widely used in food products.
Although optimization of sensory attributes were conducted by using some
optimization techniques: multiple regression (Moskowitz, 1 985), and response surface
methodology (Giovanni, 1983), the use of sensory attribute constraints in l inear
13
programming model has only recently been introduced in food product development
in the formulation of a fresh turkey bratwurst (Beausire et al., 1988). The method was
la ter used in hand cream optimization (Uaphithak, 1 990).
2.2
USE OF CONSUMER SURVEY AND CONSUMER TESTING IN PRODUCT

OPTIMIZATION
In the product optimization process, information regarding the product has to be

elicited from consumers as early as possible. Various methods have been used for
getting information from the consumers. Consumer survey and testing of product with
consumers are most cominonly used.
2.2.1
Consumer Survey
Market research is classified into observation, experimentation and survey research.

Survey research is the best known source of primary data collection, not only in
marketing but the social sciences in general . Tull and Albaum (1973) stressed that
surveys are concerned with understanding or predicting behaviour and offered as their
definition: ' Survey research is the systematic gathering of information from (a sample
of) respondents for the purpose of understanding and/ or predicting some aspect of the
behaviour of the population of interest'.
Consumer survey is a method developed to get consumer reactions towards both

existing products on the market and also products which are new to the market. With
the existing products, the consumer survey identifies for researchers the weaknesses and
strengths of their own product and those of competitors - why consumers like or dislike
the products and what improvements they want in the products. Consumer survey also
helps the researchers to identify the important attributes of the product being studied.
With the new product, the consumer survey can help the researcher screen the product
ideas and obtain the product concept and also helps the researcher to foresee the
consumer reactions toward the product to be developed.
14
The following are the tasks which the consumer survey can do:
.. Collect information about the product users

- Percentage of population who use the product
- Type of product users
.. Determine information about product usage
- How often consumers use the product
- How often consumers buy the product
>/0
Determine what are the alternative products
.. Identify the problems in product usage

.. Evaluate the importance of product attributes
Alreck and Settle (1985) stated that the main advantages of surveys are: comprehensive,
customised, versatile, flexible and efficient. This means the survey is appropriate to
almost all types of consumer research and suitable for all kinds of problems and
budgets. In spite of these advantages, surveys also have their disadvantages (Hart,
1 987):
.. The unwillingness of respondents to provide the desired data. The non

response error will invalidate research findings .
.. The ability of respondents to provide data. I t is important to target individuals
who have the knowledge and experience of the subject under survey .
.. The influence of the questioning process on the respondents. Respondents may
give the answers they think the researcher will want to hear, thus distorting
the accuracy of the data.
However, these disadvantages can be overcome by selecting a suitable survey method

and the target consumers as well as the way the questions are presen ted to the
consumers for a specific product. Alpert (1971) suggested that with a product involving
more subjective buying motives, consumers should be asked indirect questions.
Lai ( 1987) used a consumer survey to obtain information from consumers in a bakery
snack development process. The results indicated that a meat pie was the possible
product to be developed. Devro Ltd . (Birn, 1990) conducted market research at the
15
beginning of sausage development. The information obtained from the survey suggested
that consumers needed a pre-cooked, ready-to reheat sausage which provided a ready
to-eat breakfast. Moskowitz ( 1984) reported that the results from a market research
study enabled the Riley Company to come up w ith an idea to develop a deep skin
scrub made from natural products, which would both clean and enrich the skin.
Durgee (1990) recommended that in research on product sensory properties in groups

or in one-on-one interviews, it is very important that respondents experience the
product first-hand d uring the interviews. If it is a food product, they should taste it. If
it is a perfume, they should smell it. Simply asking consumers to recall the taste or feel
of a product lacks the immediacy of actually lIsing the product. With a new product
which is not available in the market, a product concept can be used to trigger
consumers' opinion on the product and its attributes. However, consumers may have
d ifficulty in visualize the product. In the development of the deep skin scrub, which
was a new product, Moskowitz ( 1984) used with consumers a number of different
stimuli representing both in-market products for facial cleanSing, and also experimental
prototypes of a deep skin scrub. The researcher realized that probably no single product
tested actually would represent the final facial scrub.
In the consumer survey conducted in this study, permanent bonding glue stick was a
product already available in the market so the commercial products could be used as
reference. However, the temporary bonding glue stick was not on the market at that
time so i t was necessary to use a different product with similar temporary bonding, in
this case a self-adhering note pad was used.
2.2.2
Consumer Testing
There are many reasons that make consumers a valuable measure in product testing.
Those reasons are given below:
* Consumers evaluate the product in a different way from the trained sensory
panel. Consu mers may use product dimensions di fferent from those of a
trained panel in evaluating product preference or product acceptance. Product
acceptance should not be obtained from a trained sensory panel.
16
* In determining product differences, consumers normally use product attribute
preference to discriminate the products but this does not apply with the
trained panel who merely use difference of attributes between products.
* In the product optimization process, it is necessary for the product developer

to identify the important product attributes which consumers use in product
evaluation. These attributes can only be obtained effectively from consumers.
,.. In measurement of purchase intention, the data should be collected from the
target consumers in order to get effective results and give the correct
prediction of product purchasing.
* Trained sensory panels tend to have too much knowledge or experience about
the product being tested which could cause bias in product evaluation.
* By associating consumer data with sensory panel data and when possible
with ingredients and processing variables, and / or with instrumental or
chemical analyses of the product, the researcher can discover the relationships
between the product formulation, product's attributes and consumer
acceptance.
2.2.3
Identification of Important Product Attributes
A ttributes represent the dimensions along which consumers rate products. The product
optimization procedure depends critically on the correct selection of product attributes.
If the researchers fail to select a true set of attributes, then this can invalidate the entire
study because th panelists do not evaluate the product by the appropriate dimensions.
Attributes which are used to describe a product can be classified as use a ttributes,
sensory attributes, liking attributes and image or appropriateness. These attributes
possess various degrees of importance. The product developer usually has to limit
attention to the primary and most important attributes during development of
prototypes, otherwise there are great difficulties in analysing the data.
17
Moskowitz ( 1984) stated that a ttributes of a product are called ' important' i f changes
in those attributes markedly increase or decrease the acceptance of the product.
Attributes are called unimportant if changes in these attributes do not affect final
acceptance.
There are several methods which researchers use to measure importance of product
attributes: direct rating (Alpert, 1971 ), rank order (Moskowitz and Chandler, 1978),
conjoint measurement, open-ended elicitation approach, information search approach,
indices based on Jaccard's subjective probability and paired comparison (Jackard et aI,
1 986).
Alpert ( 1 971) stated that, for products such as one-dollar pens, d irect questioning
methods - how important is each of these athibutes in your own choice of a pen in the
one-dollar category?, may be better than the indirect ones - with 'most people'
substituted for ' you', but for other products involving more subjective buying motives,
the indirect approach was more effective.
Moskowitz and Chandler (1978) used direct rating (via magnitude estimation) to
determine important food attributes. They found that an acceptable flavour in foods
appeared to be more important than the other product attributes surveyed. Heeler et
al. ( 1979) found that 'self-reports', in which consumers rated the given attributes
according to the attributes' importance to them, was the most reliable method to
measure the importance of food blender product attributes compared with conjoint
measurement and information display board.
Rabino and Moskowitz (1 980) asked consumers to rank order attributes of skin lotion
according to their relative importance, to identify the characteristics which consumers
felt to be important as indications of an acceptable, efficacious hand lotion.
Jaccard et al. (1986) compared several methods to measure product attribute importance.
Conjoint measurement technique, open-ended elicitation approach, information search
approach, direct rating of importance, indices based on Jaccard's subjective probability
approach and paired comparison approach were compared. The products they used in
their studies were cars and birth control methods. They found that these measures have
18
relatively low levels o f convergence. The results suggested that conclusions made about
attribute importance may be quite different depending on the method used to assess
i mportance.
I t can be seen that some methods are suitable for some products. To decide which
method to use, the type of products and consumers have to be taken into account.
Open-ended elicitation method appears to be the method most suitable for a new
product or a product whose list of attributes is not available.
After the researcher obtains a set of important attributes of the product being studied,
they have to decide on which method they will use to measure these attributes during
the optimization process.
2.2.4
Testing of Product Acceptability and Important Attributes of Product with

Consumers
Since consumers determine the success or failure of a product, it is sensible that

consumers' reactions to various formulations are collected early in the product
optimization process. It can save product optimization resources a grea t deal if the
direction toward successful product is discovered as early as possible. Acceptability as
well as sensory attributes of the product should be measured. Acceptability i tself is not
enough for the product developer to make a decision on which attributes should be
improved and in which direction. Acceptability is used in evaluating new product
concepts, in selecting basic formulations during prototype development, in product
optimization and in comparison between the developed product and the competitor's
product in final product testing. Product acceptability can be measured in several ways,
either by direct measurement or by relating to the purchase intention and the price the
consumer is prepared to pay. Sensory attributes of the product are measured in order
that the researcher knows the relationships between the ingredients and the sensory
a ttributes. The models developed from these relationships can be used in the
optimization stage.
There are several methods used for measuring product acceptabili ty and the product
sensory a ttributes. The type of measurement used in consumer testing depends on the
19
stage in the product optimization process at which the test is conducted. It also depends
on the kind of information needed and how the results will be utilised. The number of
a ttributes to be evaluated also has an effect on what method is to be used.
Usually consumer tests involve not less than 50 responses per product and larger panels
are preferred rather than replicate responses (A.S.T.M., 1986). In the study conducted
by Hovenden et al. (1979), it was found that the accuracy of an untrained panel is not
different from the accuracy of trained sensory panels. This implies that the variance
observed with consumers is similar to that observed with trained judges whereas the
untrained panelists will provide information relevant for product developers as well as
by marketing researchers (Sidel and Stone, 1985)
The methods generally used in sensory evaluation with consumers are product profile
or l ine scaling, magnitude estimation, category scaling, ranking and paired comparison.
Lawless and Malone (1986a) compared the ability to discriminate differences among
products using four types of rating scales - nine-point category scales, line marking,
magnitude estimation, and a hybrid of the category and l ine scale. In their study, visuat
tactile and olfactory characteristics of products were assessed by consumers. Consumers
j udged that category scales were easy to understand and fast to complete but somewhat
restrictive. Line scales were not restrictive, moderately fast, but a little harder to
understand. Magnitude estimation appeared worst in all aspects. Category scales were
found to have a sensitivity advantage as well as user-friendliness. All scales yielded
high F-ratios for moderately clear sample differences and these number of subjects (3455 panelists). Lawless and Malone ( 1986b) also found that magnitude estimation scaling
was used less efficiently than category scales and linear scales by a heterogeneous
sample of consumers.
Pangborn et al. (1989) compared using category, graphic and magnitude estimation
scales. They found no difference between intensity estimates obtained with category
scale and graphic scale. The results indicated that magnitude estimation was
inappropriate for scaling of degree of liking.
As mentioned earlier, both the type of measurement scales and also how many
categories should be in the scale depend on the stage in the product development
20
process at which the consumer test is conducted. It also depends on the kind of
information desired and how the results will be used. The number of a ttributes to be
evaluated also has an effect on what method is to be used. For the marketing point of
view, the method used for consumer testing should give the measurement of consumer
acceptance and also the purchase intention. These methods generate information about
overall product preference not liking of individual product attributes so that the success
of the product in the market can be estimated.
2.2.5
Consumer Research and Consumer Testing in Glue Stick Optimization
Although glue sticks have been marketed for a number of years, there are no reports
about the consumer input during glue stick development or of the identification of
product attributes by consumers. There appears to be no published consumer research
on glue sticks. Glue stick is a consumer product, hence in development of this product,
reaction of consumers towards the developed prototypes have to be taken into account.
2.3
PRODUCT PROFILE
Product profile or sensory profile is a quantitative descriptive test which represents the
most sophisticated of available sensory methodologies compared with discrimination
and acceptance testing methods. Product profiles include a complete sensory description
of the test products and provide a basis for determining the sensory attributes that are
important to product acceptance, as well as an aid in identifying the effects of
underlying ingredients or process variables on specific changes in the sensory attributes
of a product. This information cannot be obtained by conducting difference testing or
preference testing .
Product profiling consists of three major procedures: Flavour Profile developed by the
A .D. Little Co., Cambridge, Massachusetts in the late 1940s (Cairncross and Sjostrom,
1 950), Texture Profile developed at the General Foods Research Centre (Brandt et al.,
1963; Szczesniak, 1963), and quantitative descriptive analysis (QDA) developed by Stone
et al. ( 1974). These methods were developed originally for evaluation of food products.
Later the method was expanded by Civille and Szczesniak (1973) and Civille and Liska
21
( 1 975) to include specific attribute descriptors including semisolid food, beverages, skin
care products, fabric and paper goods.
Product profile is normally used with a trained sensory panel. However, this method
has been successfully used with consumer panels (Szczesniak et aI., 1975 and
Uaphithak, 1990). Product profile is a useful tool for the researcher to get a picture of
how consumers perceive the product or product prototypes. Profile of a product
comprises a set of descriptive terms with the intensity of each attribute. Acceptability
may be included to generate a more complete profile for the product (Moskowitz, 1984),
but some researchers think that this should not be done as it influences the ratings on
the a ttributes.
Szczesniak et al. (1975) stated that the consumer testing techniques used for measuring
degrees of acceptance, preference or difference between samples did not give a
thorough description of the product in terms of its sensory characteristics as they were
perceived by an untrained group of persons. Although a texture profile panel was
trained to do the job, it was criticized as being too 'artificial' and too removed from
reality. Therefore they developed a consumer texture profile technique which could be
administered in home use or in a central location type of a test. The developed method
represents a combination of popular texture terminology, classified texture terms used
by the panel, and a scaling technique on a semantic differential. One to six numbered
scales were used where 1 indicated absence of a given characteristic and 6 indicated its
presence to a very high degree. Texture terms were listed in a randomized order in the
left hand side column. Alongside each word were six boxes for checking the intensity
of the given textural characteristics. The end boxes were marked 'not at all' and 'very
much so'.
They declared that this type of rating scale gave better results with consumers than a
numerical scale in that it was more fully used to express intensities of discrimination.
Their technique also permitted the description of an ideal texture for specific food items.
An average of about 30 consumers per test had been used in the testing of breakfast
cereals, puddings, dessert gels and whipped toppings. The results were presented in a
typical 'profile' graph in which profiles for several related products, including an ideal
was plotted on the same chart. Another way was to represent the ideal product as a
22
vertical straight line corresponding to a '0' rating and describing the test products in
terms of their deviations (+ or -) from this line. This linear scale data could also be
subject to factor analysis.
For glue stick testing in this study, anchor words at each end of the scale varied for
each descriptive term instead of 'not at all' and 'very much so' in order that consumers
could understand clearly what was the meaning on each part of the scale: The
descriptive terms were presented in order of occurrence. In the sensory testing of skin
care products Schwartz (1 975) also suggested that the a ttributes were tested in sequence
of occurrence.
2.3.1
Use of Line Scale in Product Profile
In the quantitative product profile or quantitative descriptive analysis, samples are

evaluated independently by each panelist using a scaling technique, i .e., category scale,
magnitude estimation scale or line scale. For the final report, data may be displayed in
tabular or graphic form. Line scale is the most commonly used especially with
consumer panels. This is owing to its simplicity to understand by consumers. Lawless
and Malone (1986a and 1986b) and Lawless ( 1989) found that the line scale and the
category scale were used more efficiently by a heterogeneous sample of consumers than
magnitude estimation scale. Magnitude estimation was found to be somewhat less
discriminative than category scaling or line scaling. Line scale was found a l ittle harder
to understand than ca tegory scale but it was not restrictive and could be used to
provide continuity in quantitative data.
2.3.2
Ideal Prod uct Profile
Ideal product profile is obtained by asking consumers to indicate their ideal product on
a set of specified attributes (normally important attributes). The ideal product profile
is compared with the profile of product prototypes in order to estimate how far the
prototypes are from the consumer product ideal. Ideal product profile has been used
in development of many products. Hoggan (1975) used ideal absolute scores with
consumers to indicate their ideal beer on specific attributes. Szczesniak et a1. (1975)
described the use of ideal absolute scores in a consumer texture profile method.
23
Sinthavalai ( 1 986) stated that ideal absolute scores had to be shown in either graphical
or numerical form alongside the test sample scores so that the differences between
sample and ideal, in both magnitude and direction, could be understand.
Since it was found that the ideal absolute score itself was not useful in quantitatively
measuring how the different product prototypes were nearing the optimal product, the
ideal ratio score was introduced to the product testing system (Cooper et al., 1 989).
Ideal ratio scores, the ratio of the product score to the ideal score, were used to decide
the size and direction of product changes required to reach the consumers' ideal
product. Ideal ratio score has been used successfully at Massey University for many
years (Sinthavalai, 1 986; Lai, 1987; and Wiriyacharee, 1990). Beausire and Earle ( 1 986)
stated that mean ratio scores could be used in factorial experimental designs to give
empirical equations which could be used to predict the levels of ingredients or
processing conditions necessary to give optimum sensory characteristics. Use of
l ogarithms of ideal ratios was introduced later in order that ratio scores could be
symmetrized, which reduced the skewness of the data considerably.
2.3.3
Use of Product Profile in Product Optimization
There have been reports on use of the product profile in many food product
development projects. However, in the area of non-food products, the use of the
product p rofile was reported only in the development of skin care products (Szczesniak
et al., 1975; Uaphithak, 1990).
Schwartz ( 1975) showed how the principles of sensory texture profiling were applied
to the evaluation of skin care products and how the basic methodology was mod ified
to accommodate problems unique to this type of product. Schwartz suggested that the
perception of texture comprised the following phases, which occur in sequence:
,. Pick-Up
- the removal of the product from the container;
,. Rub-Out
- the application of the product to the skin, and;
,. After-Feel
- the evaluation of the effect of the product on the skin.
24
In her study, a wide variety of ingredients were screened and selected for use in
finished products. Prototypes were evaluated in comparison with commercial products
and results used successfully in predicting consumer response.
For glue stick testing, al though there is no set pattern of how panelists evaluate the
product, this pattern of Schwartz can be applied. The pa ttern of testing can be as shown
below:
... Appearance
- the eva luation of appearance of the s tick;
It
Rub-out
It
Glue residue on substrate - the evaluation of amount of glue coa ted on substrate;
It
Effect on substra te
- the application of product to the surface of substrate;
- the evaluation of change occurring to the substrate

(paper, wood etc.) after applying glue, and;
It
Stickability
- the eva luation of bond strength of glue after 2 surfaces

of substra te were placed to contac t each other.
Development of the sequence and the descriptive terms for glue s tick testing is
described in Section 3.2.2.
From the sensory profiles of prototypes and competitors including the level of product
characteristics desired by the target consumers, it is possible to determine which
a ttributes of each prototype come closest to the ideal, and which a ttributes need
modification and to wha t degree. Such a profiling method helps the researcher to
determine wha t to do next in a product development project. Empirical equations
showing rela tionships between ingredients and product sensory attribu tes can be
obtained and used in the optimization stage.
2.4
DEVELOPM ENT OF MODEL USING PRODUCT PROFI LE
Product profile is a useful tool in tha t it can produce quantita tive da ta for model
development. Empirical equations show ing relationships between ingredients and
product a ttribu tes are necessary for optimiza tion of the product. These equations can
be genera ted using multiple regression. Moskowitz (1 984) sta ted that the model plays
25
a key role in optimizing because the model summarizes the data in a tractable form,
a llowed the researcher to express empirical relations in a simple form rather than
having to lay out the data in extended tables. The model also allowed the researcher to
interpolate between levels to estimate the likely sensory attributes' ratings of the
untested points.
Moskowitz (1984) showed the use of the relationships between formulation ingredients
and sensory attributes in lotion optimization. A consumer panel evaluated the product
in a home-use test using magnitude estimation scaling. Model relationships between
ingredients and product sensory attributes were developed by using multiple regression
and were used further for product optimization. Beausire et al. (1988) used an in-house
panel to test the textural attributes of fresh turkey bratwurst. Then the quantitative
relationships between the sensory attribute and the ingredients were developed for
product optimization.
2.5
FINAL PRODUCT TESTING
After the product had been made successfully in the pilot plant scale production, it
should be tested with consumers. In final product testing, researchers may want to
compare their new product with a current formula or with the market leader. Marketing
information such as price, brand name and packaging can be included in product
testing. Product acceptance and purchase intention is normally evaluated at this stage.
Aldridge et al. (1983) reported the case history of Knorr seasoning cubes in Nigeria. In
final product testing, the researcher used Maggi product, which was in a strong
monopoly position at that time, to be compared with the developed product. Monadic
test was used in the product testing. Five-point scales were employed to measure
.
consumers' perception on each product.
The product testing can . be carried out in a central location or in a home use test.
Central location tests usually differ from in-home use tests in terms of testing
environment, as well as in the nature and length of exposure to the product. Penny et
al. ( 1 972) suggested the use of 'in-home test' for:
,.. Products whose assessment must be made over a period of time.
26
* Products for which a heavy fatigue element may be involved (e.g. highly
spiced foods, strong toothpaste, flavours, etc).
* Products for which in-home factors (e.g. preparation of foods) may be crucial
in the overall assessment.
* Products where usage instructions are followed with varying conSistency.

A fter the consumers have used the product(s) for a period of time, the reaction of
consumers toward the tested product may be obtained by a self-completed
questionnaire or by an interview. Penny et al. (1972) reported on one experiment
designed to compare the results from consumer testing by self-completed postal
questionnaire and a face-to-face interview with 2 types of products: washing powder
and margarine. They concluded that the results did not differ between these tests.
2.6
GLUE STICK FORMULATION AND PROCESSING METHOD
Adhesive sticks can be divided by their performance into 3 categories: permanent

bonding, removable bonding and temporary bonding.
* Permanent Bonding (Brennan, 1966; Muszik and Dierichs, 1971; and Ando
and Yamazaki, 1974)
The adhesive is usually intended to give a permanent bond, i.e. when one surface has
been bonded to a receiving surface and the adhesive has dried the two surfaces cannot
be separated without tearing the material into pieces. It can be used to stick paper with
paper or other materials such as textiles, aluminium foil and porous wood.
* Removable Bonding (Gollub et al., 1987)

This kind of adhesive can be used for temporary bonding purposes, such as attaching
pieces of paper bearing messages to a notice board, desk window or other substrate, the
paper being removable without difficulty and without leaving a fil m of adhesive on the
substrate.
27
>I-
Temporary bonding (Palm, 1989)
For temporary bonded adhesive, the adhesive coated surface can be adhered to a
receiving surface and removed from this without leaving traces of adhesive on the
receiving surface, and this process can be repeated a number of times.
2.6.1
Raw Materials in Glue Stick Formulation
Adhesive sticks can be produced from rubber, resins and waxes (Brennan, 1 966). These
components are combined in the form of a pressure sensitive, thermoplastic adhesive
which is shaped into an elongated or crayon like body. Muszik and Dierichs (1971)
found that with crayons made from rubber, resin and waxes the surface layer must be
rubbed off these adhesive applicator crayons under relatively high pressures in order
to carry out the spreading of the adhesive. This can cause the crayon itself to break and
the paper is damaged during the rubbing of adhesive from the crayon. So they showed
that self-supporting adhesive sticks can be formed from alkali metal salts of aliphatic
carboxylic acids, water or water-miscible organic solvents and water-soluble or water
dispersible adhesive materials.
Ando and Yamazaki ( 1974) used the reaction product of sorbitol and benzaldehyde as
the gel-forming agent. By using this as gel-forming agent, an adhesive which could
maintain a constant hardness and adhesiveness in a wide temperature and humidi ty
range was obtained. It could be readily spread by rubbing it on a surface to be bonded
to form a thin and uniform film.
However, it was found that using alkali-metal salts of aliphatic carboxylic acids as gel
forming agents had some disadvantages: no alkali-sensitive substances may be
employed and no acid-reacting active components may be added because this would
destroy the gel-like supporting structure of the stick. It was recommended that free long
chain aliphatic carboxylic acids be used . Particularly preferred were wax acids obtained
by oxidation of mineral or petroleum waxes, such as peat wax, montan wax or from
ozokerite (Werke H.u.M. Fisher, G .m.b.H., 1974). Hence, adhesives which are active in
the acid range, e.g. resin acids or esters of resin acids can be employed.
28
I n general the raw materials used in adhesive sticks can be classified as follows:
>I-
Adhesive
>I-
Gel forming substance
>I-
Solvent
The following raw materials may also be added in the formulation in order to improve
some properties of the adhesive stick.
2.6.2
>I-
Plasticizers
>I-
Emulsifiers
>I-
Filler
>I-
Antifoam
>I-
Colouring
>I-
Perfume
Adhesive
A d hes i v e or fi l m fo r m i ng s l l hs t a nl'l i s t i ll! p r i m a ry com p onen t ilnd haH t he f u n c t ion o f
holding the substrates, particularly paper, together. The adhesive can b e hard resins
such as hydrogenated terpenes, polymerized terpenes, phenolic terpenes, polymerized
rosin esters, and hydrogenated rosins with melting points of approximately 65 c.
Rubber ingredients which are used as adhesive are hydrocarbon polymers with
molecular weights of from about 1,000 to 25,000 (Brennan, 1966).
Adhesive can also be any water- or alcohol-soluble adhesive either natural or synthetic,
e.g. polyvinyl alcohol, polymethacrylic acid, polyacrylamide, copolymers of acrylamide,
salts of polyacrylic acids, polyvinyl pyrrolidone, ethylene maleic anhydride copolymer,
methyl vinyl ether copolymers, carboxylated polyvinyl acetate. The preference of many
mani.l factu rers is polyvinyl pyrrolidone (Adhesive Topes Li mited, 1 974). Natural
carbohydrate polymers or modified carbohydrate polymers, such as starch, dextrin, gum
arabic, cellulose, methylcellulose, cellulose esters or carboxymethyl cellulose may also
be employed. M ixtures of adhesives may be used. It is particularly preferred to employ
polyvinyl pyrrolidone especially in conjunction with 2-6'1., by weight of a carbohydrate
29
or modified carbohydrate polymer (Werke H.u.M. Fischer, G . m.b.H., 1974). Adhesive
Tapes Limited (1974) reported the use of ] 0 percent dextrin solution with polyvinyl
alcohol as the adhesive substance in glue stick formulation.
2.6.3
Gel Forming or Shape-Giving Substance
Gel forming substances are blended with the binder in an adhesive stick to give a stick
which is easy to apply without deformation. It also gives a uniform coating, free from
lumps of adhesive on the substrate. However the gel forming agent reduces the tack of
the adhesive so the type and the amount of this ingredient must be chosen carefully
(Palm, 1 989). The following are gel forming agents commonly used:
* alkali or ammonium salts of aliphatic carboxylic acids with 8 to 36 carbon

atoms such as: sodium stearate (Muszik and Dierichs, 1 971; Gollub et aI., 1987
and Palm, ] 989)
* reaction products of sorbitol and benzaldehyde (Ando and Yamazaki, 1 974)

* wax which can be one or more partly saponified or fully saponified esters of
10 to 30 carbon atom acids with various alcohols, both natural and synthetic,
oxidised polyethylene wax, paraffin wax, chlorinated paraffin wax, glycerides
of fatty acids, hydrogenated oils and fatty alcohols. The preference is for ester
waxes (Adhesive Tapes Limited, 1974).
2.6.4
Solvent
Solvent can be water or water miscible organic solvent or both. Water miscible organic
solvents used are one or more mono or multivalent alcohols, e.g. methanol, ethanol,
isopropanol, glycerol, ethylene glycol and may include water as in water/ organic
solvent mix tures, e.g. water/acetone, ester/ethanol (Adhesive Tapes Limited, 1 974).
Some water is used in most formulations to dissolve the adhesives and other water
soluble ingredients
30
Plasticizers and Emulsifiers
2.6.5
Plasticizers or softening or moisture retaining agents can be used to tackify the adhesive
and to control drying out on the substrate to which the adhesive is applied. These
substances are also conducive to an easy, soft rubbing. Tri- and tetra-ethylene glycol,
sorbitol, mannitol, glucose, glycerol, propylene glycol, polyethylene glycols and
polypropylene glycols of molecular weight 400 to 1 500 can be used. The preference is
for glycerol - up to 25% by weight (Werke H.u.M. Fischer, G .m.b.H., 1 974).
The emulsifier can be non-ionic, cationic, anionic or mixtures of these emulsifiers

provided that, in combination with the other ingredients, it produces a water in oil
emulsion. The preference is for non-ionic emulsifiers, particularly the polyethylene oxide
esters of phenols and fatty esters. The emulsifier can also act as a lubricant in the
adhesive stick. These substances may be used to increase transparency of the product.
The following are examples of the types of emulsifiers which can be used (Adhesive
Tapes Limited, 1974):
Nonionics:
polyoxyethylene esters of a phenol, polyoxyethylene esters of a fatty acid,

polyoxypropylene esters of phenols and fatty acids, polyoxyethylene
esters of phenols and fatty acids, glycerides of fatty acids, sorbitan fatty
acid esters, fatty acid amides, natural occurring emulsifiers - lanolin and
cholesterol derivatives.
Cationics:
ethoxylated amines, quaternary ammonium compounds.
Anionics:
soaps, sulphonates, phosphates.
2.6.6
Fillers, Antifoaming Agents, Colours, Perfumes
Fillers are nonadhesive substances added to the adhesive to improve its working
properties, strength, permanence, or other qualities. Fillers are also used to reduce
materials costs. Considerable changes can be made in the properties of the adhesive by
selective use of fillers such as clay, chalk (Pletcher and Wong, 1978).
31
Antifoam Nopco 8034, a n antifoam agent containing silica, mineral oil and a
methacrylate copolymer can be added (Palm, 1989). Dye-stuffs, e.g. phthalocyanine blue,
may be used to add colour to glue stick (Pletcher and Wong, 1 978). Odour improving
compounds such as, pine-needle oit eucalyptus oil, aniseed oil, benzaldehyde, may be
used (Muszik and Dierichs, 1971).
2.6.7
Formulation, Processing and Packaging
The composition of the adhesive sticks must be so chosen, that a not too soft mass is
formed which may be readily rubbed off. This is usually achieved when the content of
the gel forming substance is between 3 and 60% preferably from 10 to 40%. The volatile
liquid components of the adhesive sticks normally amount to about 20-80%, particularly
30 to 65% and the proportion of adhesive components normally lies somewhere
between 5 and 50%, usually between 20 and 40% (Werke H.u.M. Fischer, G .m.b.H.,
1 974).
The tougher and more cohesive the formulation, the more difficul t it is to separate a
film of the material from the crayon body. It has been found that the higher the initial
strength of a formulation tested, the more difficult the film of adhesive will be to
separate from the crayon (Brennan, 1966). Brennan also added that the formulation with
the lower softening temperature will apply a film with less physical effort than the
formulation with the higher softening point. For this reason the softening point should
be generally kept as low as practical.
Processing methods vary according to the raw materials used in glue stick. The
adhesive sticks are usually prepared by mixing the individual components while
heating them to 0-95 C until homogeneous under reflux conditions, the process taking
about 7 hours. Then the mixture is allowed to cool in a mould or after extrusion in the
desired shape form (Muszik and Dierichs, 1971).
The stick may be conveniently mounted in an applicator of the lip stick holder type. In
such a holder i t may be readily carried among personal possessions or kept in an office
desk for use in light gluing applications such as the sticking together of sheets of paper
or cardboard or thin layers of flexible plastic materials. The stick can also be placed in
32
2.7
USE OF STARCHES IN GLUE STICK
Starch adhesives are suitable for labelling and sealing by high-speed, automatic
machinery for manufacturing corrugated boxwood, plywood, envelopes, stamps and
gummed tapes. They possess the advantages of low price and can be applied cold or
;.
at moderately low temperatures, being reasonably lacking in odour and taste. The latter
point makes them especially useful in the food packaging industry or for any use where
the adhesive may make contact with the tongue, e.g. envelopes and stamps. The
adhesive properties of starch are developed when the starch is gelatinised and then
cooled to a gel. The starch gelatinization involves firstly hydration of the starch granules
.,
which swell to several times their original size, followed by a change from opaque to
clear solution, with a rapid increase in viscosity. Finally the linear molecules of amylose
dissolve and diffuse from the ruptured granules. Tapioca starch has a gelatinization
temperature higher than potato starch but lower than maize starch, and develops a
moderate viscosity during cooking.
. i
.'
,
.'
" .
Upon cooling the uniformly dispersed matrix forms a gel or a paste-like mass. This
three-dimensional gel network is a mixture of swollen granules, microgels of
amylopectin and a soluble fraction of macromolecules and amylose. The strength of this
network determines the strength and rigidity of the gel. The elasticity of the starch gels
is due to the ability of the gel to be stretched to a certain extent without breaking. The
gel stability of tapioca starch is much higher compared to cereal starches and hence is
preferred for adhesive products. In addition, the higher paste clarity is a desirable
property of tapioca starch. For adhesive purposes, the starch gel should retain its water
solubility for considerable periods of time. Starch gel obtained from corn starch shows
an excessive loss of adhesive properties on aging. This appears to be due to
retrogradation of starch molecules. Tuber, root and waxy starch products give starch
gels that retain their adhesiveness for long periods of time because of slower rates of
retrogradation (Swinkles, 1985).
2.7.1
Uses of Starch and its Derivatives in Adhesive
The physical nature of the native unmodified starches limit the usefulness of starch in
many commercial applications. These shortCOmings may include the lack of free-flowing
properties; insolubility or failure of the granules to swell and develop viscosity in cold
water; excess or uncontrolled viscosity after cooking; and the sensitivity of the cooked
33
starch to break down during extended cooking, when exposed to shear, to low pHs or
high pHs (Wurzburg, 1986). There are many effective ways to overcome these
shortcomings and thus expand the usefulness of starch for a number of industrial
applications. Starch can be modified by depolymerization reactions, cross-linking
reactions and stabilization (Fleche, 1985).
Depolymerizing a starch reduces its viscosi ty and consequently, allows its use at a
higher level of dry solids. This can be done by dextrinification, acid conversion, alkaline
conversion
or
oxidation,
thermochemical
liquefaction,
a nd
enzyme-prod uced
liquefaction.
Cross-linking or bridging the molecular chains, leads to a more rigid macromolecular

network inside the granule. This reaction, while limi ting granule retention capacity, sets
the viscosity level to the desired value, provides greater resistance to thermomechanical
shearing, improves stability in acid media (Fleche, 1985). In industry, those starches
with gels resistant to mechanical shearing and to various media (acid, basic, or salts)
can be used in corrugated cardboard, in textile printing pastes, in adhesives and in
other applications.
The stabilization of starch reduces the hydroxyl number and decreases the reassociation
of the molecules. Consequently, retrogradation is slowed down and starch stability
increases, particularly a t low temperatures. Esterification or etherification provides
stabilization, although most of the commercially modified starches are cross-linked and
stabilized.
Apart from modification, starch can be made into dextrin. The term dextrin covers
products produced in a variety of ways. In its broader sense it refers to all degradation
products of starch without reference to the manner in which they are produced. Thus,
dextrin may refer to the degradation product produced by enzymatic, as well as acidic
catalyzed,
h y d ro l y s i s
rlln i n
aqueous rned i u m o n
granular or peptide starch or by
pyrolysis of starch in the dry granular form. They may range from slightly degraded
starch polymers to highly degraded po ly m ers of anhyd rogl ucose units. Three primary
groups of dextrins are now known: Bri tish gu ms, white dextrins and yellow dextrins.
All are made by heating powdered starch. They differ in the manner in which the starch
34
is treated prior to the heat treatment, the manner and extent of heating, and the
properties of the resulting products (Wurzburg, 1986).
These modified starch and dextrins are used in numerous adhesive applications. Cross
linked starches are used in corrugating adhesives to provide high viscosity under
strongly alkaline conditions (Rutenberg, and Solarek, 1984). Dextrins are used in Tube
winding, laminating, case and carton sealing, bottle labelling, flat gumming, envelope
sealing, and gummed tape manufacture.
2.7.2
Use of Tapioca Starch in Thailand
Tapioca starch has the same general properties as other kinds of starch and so it makes
a good substitute. It does have one feature which is of interest. Tapioca starch contains
very little protein, thus it is a fine material to start with in industrial processes which
transform starch into other substances. Fewer steps are required and this means lower
production costs.
Tapioca starch can be converted into sugars - glucose and fructose and because of this,
it is widely used in the preparation of a variety of foods. It is also used as a raw
material in producing monosodium glutamate. In addition, thanks to i ts qualities as a
thickener, a hardener, and adhesive, and a moisture stabilizer, tapioca starch finds
applications in the manufacture of many different materials. including glues, plywood,
textiles, paper and medicines.
Tapioca starch has been used in making many kinds of tasty Thai sweets and other
foods and nowadays tapioca starch is also widely used in many of the commercially
prepared foods and confectionary products sold in shops and department stores.
Tapioca starch is an ingredient in, for example, bakery products, sausages, bean
vermicelli, rice noodles, tonic food beverages, ice cream, and many types of candies
(TTTA, 1990).
One of the important end use of tapioca starch is adhesive application. It can be simply
gelatinized in hot water or with the help of chemicals. Tapioca is considered more
suitable for the manufacture of adhesive in that it gives adhesives which are more
35
viscous, smoother in working and more easily prepared, whilst the joints made from
these adhesives give a higher tensile strength than .those made from potato or maize
starches (Radley, 1 976).
2.7.3
Use of Starches in Glue Stick Formulation
Adhesives which are now commonly used in glue stick can be either water- or alcohol
soluble. Although polyvinyl pyrrolidone (Adhesive Tapes Limited, 1 974) is more
commonly used, mixtures of adhesives may be used. It i s preferable to employ

polyvinyl pyrrolidone in conjunction with 2-6% by weight of a carbohydrate or
modified carbohydrate polymer (Werke H.u.M. Fischer, G.m.b.H., 1974).
Although it was mentioned that starch and starch derivative could be used in glue stick
formulation (Werke H.u.M. Fischer, G.m.b.H., 1 974), there was no report on which type
or form of starch was suitable and at what levl. Tapioca starch was chosen as a
substitute for synthetic adhesive in glue stick formulation, firstly because tapioca is a
raw material produced largely in Thailand, secondly, tapioca starch has low
retrogradation tendency, and good sol stability. Tapioca starch products give starch
films that retain their adhesiveness for longer periods of time compared with starch
films from other starch such as corn and wheat starch which show an excessive loss of
adhesive properties on aging due to retrogradation of starch molecules (Swinkels, 1 985).
Therefore, tapioca starch or a derivative of tapioca starch was considered suitable for
using as an adhesive substance in glue stick. There was no mention found in the
literature on the use of tapioca starch in glue stick. Hence, the suitable level of the
tapioca starch as well as other starch compatible basic ingredients in glue stick
formulation had to be investigated.
2.8
TESTING OF G LUE STICK PROPERTIES
No standard method for glue stick testing was found in the literature. There were
reports showed that researchers had used physical testing to determine the performance
of the products during the development of glue sticks (Ando and Yamazaki, 1 974; and
Pletcher and Wong, 1978). Although the usage properties perceived by the consumers
36
had been mentioned (Muszik and Dierichs, 1971; and Palm, 1989), no sensory evaluation
had been involved in the development of this product.
The important physical properties of glue stick which are normally tested can be
grouped into 4 main categories: characteristics of the stick itself, amount of glue applied
on paper, initial adhesion, and adhesion after drying. The physical properties which
have been tested are as follows:
2.8.1
Stick characteristics
The . softening temperature was determined using a Fisher-Johns melting point

apparatus set to rise in temperature at a rate of 2 C per minute. The temperature a t
which the material softens and starts t o turn translucent. The softening temperature o f
at least about 40 C was recommended s o that the stick will not become too tacky or
fluid to hand hold or store. Although no mention was found on maximum softening
temperature, an unduly high softening temperature will render the adhesive difficult
or impossible to friction activate and will detract from other properties (Pletcher and
Wong, 1978).
Hardness was measured according to Japanese Industrial Standard GIS) K 2530 by using
a penetrometer as used for asphalt and was shown by a depth of penetration for 3
seconds of a needle loaded up to 46.3 g in total weight (Ando and Yamazaki, 1974).
The shape-stability was shown by the number of cycles of putting an adhesive crayon
out and in repeatedly, which was packed in a cylindrical container of 15
mm
in
diameter, made of polyethylene and provided with a screw used for putting out and
in a content, by operating the screw until it became impossible to set the crayon in the
container (Ando and Yamazaki, 1974).
2.8.2
Amount of glue applied on paper
The write-on factor relates to the quantity of adhesive which is deposited on the
substrate while rubbing. The physical effort required in rubbing should be neither too
37
high nor too low. if the effort required to transfer a given amount of adhesive is too
small, it is likely that too much adhesive will be applied (which is wasteful and can be
messy). On the other hand, if the effort required to transfer a given quantity of adhesive
is too great, inadequate adhesive will be transferred to the surface to be adhered to
allow a firm bond to be achieved.
The write on factor is measured utilizing an apparatus with an oscillating base and in
which an adhesive stick can be mounted vertically above and resting on the base. An
adhesive stick approximately 1 cm by 1 cm in cross section and about 5 to 8 cm long

is mounted in the apparatus and the bottom is flattened by running i t against a piece
of paper clamped to the oscillating base. The adhesive stick is then weighed and
replaced in the apparatus. A sample of 100% rag paper with a 20 Ib basis weight and
a cockle finish is then clamped to the base and the base is set into motion at about 220
cycles per minute, the total distance travelled by the base being 6.9 cm per cycle. The
preflattened end of the adhesive stick to be tested is lowered onto the oscillating base,
the stick being pressed against the base with a force of about 2,720 g. After 18 cycles the
adhesive stick is lifted from the paper and reweighed. Its loss of weight in grams
during the 18 cycles is the write on factor. The recommended write-on factor was a t
least 0.06 g per 1 8 cycle (Pletcher and Wong, 1 978).
2.8.3
Initial adhesion
The open-time is the interval during which the adhesive remains bondable a fter being
cooled below the softening temperature i.e. after rubbing finished. This should be
sufficiently long to allow the pieces which are to be adhered to be positioned in contact
but not so long that there is an undue wait for the firm bond to develop. An open time
of from about 20 seconds to 1 0 minutes was recommended. Preferably the open time
of the glue stick is at least about one minute.
The open time is determined utilizing the same apparatus and sample preparation as
in the procedure for determining the write on factor. The adhesive stick is run against
the rag paper for 18 cycles to heat the adhesive on the end of the stick above its
softening temperature and to transfer it to the rag paper sample on the base. As soon
as the rubbing is stopped the adhesive on the rag paper sample cools to a temperature
38
below its softening temperature. Thus, the open time for a pa rticular adhesive stick is
the elapsed time between the end of the rubbing cycles and the point at which the
adhesive on the rag paper reverts to its non-bondable state. Whether the adhesive is still
bondable after a particular interval is determined by placing a strip of 80 lb basis weight
construction paper in contact with the adhesive on the rag paper sample and pressing
it down with four passes (twice in each direction) of a two kilogram roller. The
composite sample is allowed to stand for thirty minutes and the two pieces of paper are
then pulled apart. If delamination occurs, the adhesive was is the open state when the
construction paper was applied. The open time for an adhesive stick is the maximum
interval between the end of the rubbing cycle and the application of the construction
paper to the adhesive on the rag paper which resul ts in bonding. To determine this,
several composite sample must be prepared and tested for delamination. The open time
is ordinarily determined to an accuracy of at least about 10% (Pletcher and Wong, 1978).
The initial adhesion was measured by observing the breaking state of paper layer,
which is occurred by peeling rapidly a sheet of craft paper of 25 mm x 1 00 mm in size,
0 . 1 3 mm in thickness and 68.5 g / m2 in weight, bonded to another sheet of the same
craft paper by applying an adhesive and fixing the both sheets and then pressing for
a certain period by means of a rubber roll loaded up to 30 kg in total weight (Ando and
Yamazaki, 1974).
The peel adhesion or peel strength in the open state is desirably great enough to hold
the two pieces to be adhered (e.g. paper) together against their own weight, but low
enough to allow repositioning of the pieces without damaging them by delamination.
The 1 800 peel adhesion is tested on an apparatus of the type described in 'Test Methods
for Pressure Sensitive Tapes' by the Pressure Sensitive Tape Council, Illinois, and using
a sample prepared as follows:
An 0.13 mm (5 mil) thick film of the adhesive to be tested is knife coated onto a silicone
release liner, the coating knife and the (molten) adhesive being at about 120 0c. After
cooling, the coated material is cut into 2.54 cm strips approximately 20 em long. A
composite test sample is prepared by placing the following on a 10 cm by 20 cm tin
plated steel panel in the l isted order: a 2.5
ern
by 20 cm 100% rag paper as described
39
above, the 2.54 cm by 20 cm adhesive strip (with silicone l iner removed), a 2.5 cm by
20 cm by 0.05 mm thick strip of aluminum foil (the dull side of the foil against the
adhesive strip) and a 1 0 cm by 20 cm glass plate. The assembled test sample is placed
in an oven at about 80 C for eight minutes. Thereafter the steel panel and the glass
plate are removed and the sample is placed in a 21 C, 50%
relative humidity
atmosphere for testing. The open state peel adhesion is run 5 minutes after removal of
the sample from the oven. The carriage of the adhesion machine is set to move at a rate
of about 229 cm (90 inches) per minute. The strip of 100% rag paper is a ttached to the
carriage with double coated tape. The adapter ring leading from the meter on the
machine is attached to the aluminum foil. The carriage is started and the value observed
from the gauge is recorded as the adhesion value in grams per centimetre. The peel
adhesion in the open state should not less than about 50 g/ cm. (Pletcher and Wong,
1 978).
2.8.4
Adhesion after drying
The peel adhesion in the closed state is desirably great enough to hold the pieces
together strongly but not so great that the adhesive cannot be removed from desk tops
and the like.
The closed state peel adhesion is run 24 hours after removal of the sample from the
oven using the same procedure as the open state peel adhesion. The peel adhesion in
the closed state should not less than about 100 g/cm. (Pletcher and Wong, 1 978).
Ando and Yamazaki (1974) measured the peeling strength by using an Instron in the
manner of peeling an aluminum foil of 0.06
mm
in thickness, which was bonded to
another same foil of 25 mm X 1 00 mm by applying an adhesive on the surface of either

foil within an area of 25 mm x 25 mm at 20 C and 65% relative humidity and fixing
two foils and then pressing at a pressure of 50 kg/ 25 mm x 25 mm, under the
condi tions of tensile rate of 200 mm/mm and peeling angle of 1 80.
The shearing strength w as measured on the sample, prepared in the same manner as
the sample used in the test of peeling strength, under the conditions of a tensile rate of
10 mm/mm and a shearing angle of 0 (Ando and Yamazaki, 1974).
40
2.9
CONCLUSIONS
Although tapioca starch is widely used in the adhesive industry, no mention was found
on the use of tapioca starch in glue stick formulation. According to the literature,
natural carbohydrate polymers such as starch and starch derivatives could be used upto
6% in combination with polyvinyl pyrrolidone. Therefore, i t was decided to explore the

use of tapioca starch as substitute of polyvinyl pyrrolidone which was the most
commonly used adhesive substance in glue stick.
In the development of glue stick using tapioca starch as adhesive substance, it was
necessary to select the suitable basic raw materials at the right levels using proper
processing method. The effects of ingredients on glue stick attributes had to be
identified. The suitable testing methods for glue stick attributes, both the physical and
sensory attributes, had to be established.
In this project, consumer acceptance of the product was the vital key to its success.
Although consumers have been employed in optimization of various products, there has
been no investigation into the stages where consumer input is useful in developing
adhesive products especially glue stick products. The important attributes of this
product also had to be identified so it was possible to assess the consumer acceptability
of the product.
Linear programming was selected as optimization technique because it provided a

means of determining optimum formulation relatively quickly and efficiently and
sensory constraints including acceptability constraint obtained from consumer panel can
be included.
41
CHAPTER 3
PROJECT METHODS
In this chapter, the optimization procedure used in the research is outlined and details
of some of the stages in the project shown in Figure 3.1 will be described. Since the
methods used in this project were different in nature, this chapter discusses the project
methods in general, while specific methods and techniques for certain stages are
discussed in the chapters associated with them.
INITIAL DEVELOPMENT
Consumer survey
Determination of important attribu tes
Selection of basic formulation
PROTOTY PE DEVELOPMENT
Prototype testing
Optimization of formulation by linear programming
Prod uction trial
Product costing
Test of final product
FINAL PRODUCT TESTING

In home use test
Figure 3.1 Stages of glue stick optimization
42
3.1
CONSUMER STUDY
Two major types of consumer study methods were used in the project: consumer survey
and consumer panel testing.
3.1.1
Consumer Survey
Consumer survey was used at the beginning of the glue stick optimization process in
order to gain information on how consumers perceived glue sticks already on the
market and what were the important attributes of this product according to the
consumers.
The consumer survey was done in two sections: preliminary survey in New Zealand
with Thai consumers and New Zealand consumers and the survey with the target
consumers in Thailand. Details are given in Section 4.2.
3 . 1 .2
Consumer Testing
The consumer testing was conducted in Bangkok with the target consumers - students
(school and university students) and office workers. The consumer testing was divided
into laboratory testing and home-use testing. Ninety consumers were used in laboratory
testing and 1 72 consumers were used in home-use testing. In laboratory testing, the
sensory product profile technique was used. In home-use testing, a self-administered
questionnaire with multiple-choice questions as well as open-ended questions was
employed. Details are given in Section 6.4.
3.2
SENSORY TESTING OF G LUE STICKS BY TRAINED SENSORY PANEL
A trained sensory panel was used d uring selection of the basic formulation. This enable
the author to be able to identify the suitable basic formulation for starch based glue
stick before the prototypes were tested by the consumer panel. The sensory panel
training comprised 3 parts: selection of panelists, development of descriptive terms, and
training of panelists.
------- --
43
3.2.1
Panel Selection
The questionnaire used for panel selection was adapted from a prescreening
questionnaire for a tactile panel (Meilgaard et al., 1987b) and a scaling questionnaire
was used to test the candidate's potential to learn scaling (see Appendix 3.1).
In order to be qualified at this stage the candidates had to:

* Be available for the training sessions .
.. A nswer 80% of the verbal questions correctly and clearly.
* Assign scalar ratings which were within 1 0% of the correct value for all figures.
Candidates, 15 students in the Consumer Technology Department, Massey University,
New Zealand, were invited to attend the panelist selection session. These candidates
were those who were interested in participating in the sensory project and were
available for sensory testing when needed. They had never been trained as a sensory
trained panelist, but most of them had done sensory testing and knew how to use the
line scale which was employed in this study. They were informed that 7 dollars per
hour would be paid for those who attended the panel training. Fifteen panelists were
screened at this stage and only one candidate was not qualified.
The triangle test was used to select panelists who had abili ty to discriminate products.
Sequential triangle tests as described by Meilgaard et al. ( 1987a) were used in order to
economize in the number of evaluations required. These tests are very practical and
efficient because they take into consideration the possibility that the evidence derived
from the first few evaluations may be quite sufficient to draw a conclusion. Four
commercial glue sticks and four glue stick prototypes prepared in the laboratory were
used in the test. Fou r subjects were rejected from this testing. The other subjects were
asked to attend the training session for glue stick sensory testing and test prototype
products during the development of the basic formulation for starch based glue stick.
44
3.2.2
Development of Descriptive Terms
Ten selected panelists, 6 men and 4 women, were asked to attend the descriptive term
development which comprised two one-hour sessions. The objective of this part of the
training was to develop descriptive terms and their definitions for the sensory attributes
of glue stick.
During the first session, panelists were presented with 6 commercial glue sticks and
were asked to use the products in the same manner as they normally did, then they
wrote down the sensory attributes of the products for the following categories: appearance, sensory attributes while applying glue, glue residue on paper, stickability,
effect on paper. Odour and packaging attributes were not included in the study. Fifty
three terms were developed. Panelists then discussed the terms developed, deleted the
terms which had the saine meaning and changed some terms to make them easier to
understand both for the trained panelists and for consumers in the fu ture work. There
were 25 terms altogether at the end of the session (see Appendix 3.2) .
In the second session, panelists developed the definitions for the descriptive terms. It
was necessary that the definition of each term was developed so that every panelist
agreed on the definition, and used these terms in the same way during the sensory
testing of the product. In the first place, some panelists did not agree with the
developed definition, they had to discllss it and modify the definition until everyone
agreed with it.
In this session, panelists also developed the adjective terms describing the intensity of
the sensory attributes. They had to make decisions on which terms should be on each
end of the scale in order that everyone understood the scale and did not mark on the
w rong part of the scale.
3.2.3
Training of the Panel
Panelists were invited to the training sessions, each session lasted about one hour. The
panel was divided into two groups with 5 panelists in each group to make it easy for
the panel leader to manage and so that the panelists could chose to join the session
45
when they were available. Panelists were circulated between groups so they had a
chance to work with d ifferent panel members.
A 15 cm semi-structured line scale anchored at both ends was used. For descriptive
terms whose intensity could be varied from none to all, the line scale was anchored at
both ends. For the terms whose intensity varied from very weak to very strong, the
scale was anchored at 1 .5 cm from both ends.
The descriptive terms for product a ttributes developed in Section 3.2.2 were used
during the panel training. Only those terms describing sensory attributes perceived
while applying, glue residue on paper, effect on paper, and stickability were used.
There were 18 terms used in the training.
Samples used in the training were commercial glue sticks. They varied in many sensory
a ttributes so that the panelists could experience what constituted the extreme intensities
for many of the a ttributes. Each sample was coded with 3 digit random number.
The testing procedure started with a panelist orientation on the sensory testing method.
Each panelist was asked to read the instructions and the questionnaire before
performing the test. The definition of each sensory attribute term in the questionnaire
was explained by the panel leader to make sure that everyone understand it clearly. If
any panelist did not fully understand any sensory attribute term, it was explained until
the term was clearly perceived.
In general, sensory testing should be conducted in a room which is partitioned into

separate booths, in order to avoid subject-to-subject influences. However, in this study,
no such room was available and also during the training an overhead projector was
needed so a seminar room was used. Lighting for the testing area was uniformed and
provided by daylight fluorescent lamps so that panelists could perceived the glue
residue on paper and the effect of glue on paper. The panelists were seated separately
during sample evaluation to avoid any distraction.
A set of samples was presented to the panelists, they were told to wash their hands
before the testing so that there was no dirty mark on the paper during testing. Every
46
subject was instructed to evaluate the product in the same manner. Firstly wind up the
stick so that the end of the stick comes out about 0.3-0.5 cm, and hold the stick 90
degree to the paper surface then rub the stick along the surface of paper and spread
adhesive to cover the area to be bonded using back and forth strokes. After finishing
one set of rubbing, panelists were asked to clean the tip of the stick every time in order
to get rid of the part that might disintegrate from the stick.
For the effect on paper and stickability attributes, the panelists were asked to apply glue
on a piece of paper then place this coa ted paper on top of another piece of paper and
rub repeatedly with fingers, then evaluated the attributes.
Bond strength was evaluated both before drying (as soon as the rubbing finished) and
after drying (30 minutes after rubbing). In order to allow 30 minutes drying time for
bond strength after drying, panelists were asked to use the sample to stick the given
papers together at the beginning of the test then evaluate the bond strength after drying
at the end of the test.
In the training method, the panelists were trained in four sessions. Each session lasted
about one hour. At the end of each session, the panelists compared their scores by
transferring their scores onto the overhead projector. If their score was different from
the others they were asked to test the sample again and explain how they tested and
evaluated the sensory attributes. If their testing method was different from the other
panelists they had to change their method. After that they could change their score if
they wanted.
In the first three sessions, 3 samples were presented to the panel. For the first training,
the panelists evaluated the samples in an open session so they could compare their
scores with other members in the group, discuss the testing method and the terms
which were used to describe glue stick sensory attributes. The panel leader also
motivated the panelists to use the whole scale, if from the results, they used only some
part of the scale.
At the end of the third training, one of the three samples was selected to be a reference.
The average scores obtained from the panel for that sample were used as reference
47
a ttribute scores. Panelists were asked to test the reference sample again. Then they were
asked whether they agreed with the given scores. The scores were adjusted until
everyone agreed.
In order to reduce the number of sensory attributes to be tested, some attributes were
dropped at this stage. Those attributes were spreadability, stickiness, evenness, wetness,
and visibility of glue residue trail the paper.
In the fourth training session, only 13 sensory attributes were used. These terms and
their definitions are shown in Figure 3.2. Reference attribute scores were marked on the
line scale in the questionnaire beforehand. The questionnaire is shown in Appendix 3.3.
Panelists were asked to test the reference sample and they were asked to mark the
scores for the reference if they did not agree with the specified scores. They were then
asked to test 4 samples, 2 samples were previously used in the first three sessions, the
other two samples the panel had not tested before.
The reference was used in order that panelists had something to refer to during testing
the samples. The results from different testing could be compared; if scores of the
reference were much different in different tests, the scores had to be analyzed carefully.
3.2.4
Use of Trained Panel
The trained panel was used to test the product prototypes developed in New Zealand,
during the selection of the basic formulation for starch based glue stick.
48
SENSORY AITRIBUTES WHILE APPLYING G LUE STICK
.. Ease of applying
Rub the stick along the surface of paper and spread adhesive
to cover the area to be bonded.
Slipperiness
- The feeling of force between working surface of the stick

and paper while rubbing the stick along the paper. Rated
as 'drags - slips' .
.. Perceive attribute of the stick while a pp lying
While using the stick the following attributes are evaluated:

Hardness
- Perceive hardness of the stick. Rated as 'very soft - very

hard'.
Deformation
- Tendency to deform when apply with hard force. Rated as

'low - high'.
Disintegration
- The tendency to disintegrate when apply w i th hard force.

Rated as 'low - high'.
GLUE RESI DUE ON PAPER
The following attributes are evaluated by visual

inspection
Degree of coverage
- Degree of adhesive cover the area of paper after applying

one coat. Rated as 'none - total'.
Thickness
- Amount of adhesive left on paper after applying one coat.

Rated as 'very thin - very thick'.
Visibility of glue trail
- Ease of seeing the glue trail. Rated as 'invisible - very

visible'.
EFFECT ON PAPER
Place another p iece of paper on top of the coated paper and

rub repeatedly with fingers then evaluate effect on paper by
feeling the surface of the papers with fingers and using
visual inspection.
Smoothness
- Unevenness of paper surface may cause by moisture or

lump of adhesive. Rated as 'very wrinkly - very smooth'.
Cleanliness of work
- Adhesive residue left on top surface of (degree of stain)

paper which makes paper dirty and sticky. Rated as 'very
dirty - very clean'.
STlCKABI LlTY
Adjustability
- Ease of repositioning the paper. Rated as 'very difficult very easy'.
Bond strength
- Strength of adhesive bond between two surfaces. Rated as

'very weak - very strong'.
Before drying
- Evaluate by peeling bonded paper immediately after two

surfaces arc pl aced tu contact each other.
After drying
- Evaluate by peeling after adhesive dried out (30 minutes

after placing two surfaces to contact each other).
Delamination of paper
Figure 3.2
- Degree of delamination of paper after peeling the bonded

papers. Rated as 'none - all'.
Descriptives and their definitions used by trained panel
49
3.3
PHYSICAL TESTING
The test methods used during development of glue stick products were chosen from the
existing methods such as American Standard Testing Method (ASTM, 1 991), standard
test methods from 3M, a reliable organization (Pletcher and Wong, 1978), test methods
developed by researchers or companies for similar products (Ando and Yamazaki,
1974). However, in some cases the new methods were developed for the product testing
in order to cover all the tests necessary for the product.
3.3.1
Choosing the Physical Testing Methods for the Product
Several factors were recognised in choosing the physical test methods for glue sticks correlation of physical tests and consumer reactions, need for simple tests for routine
quality control, reliability and precision. As glue stick is a consumer product, the
physical test should allow the researcher to estimate the reaction of consumers toward
the use of product. The substrate used in product testing should be the substrate which
represents the one that will be used with finished product, in this case paper. The
testing methods should also be appropriate for routine testing since they are used as
performance quality control tests. The number of test methods should be as small as
possible but covered all the necessary product characteristics. Reliability and precision
of the methods were important factors to be considered as well.
The physical testing methods used in this project were:
,.. Properties of the stick
- hardness
- melting pOint
- moisture content
- water activity
,.. Amount of glue applied per area - wet glue per area
- dry glue per area
,.. Stickability
- open time
- peel strength
50
3.3.2
Hardness
Hardness of glue stick had an impact on ease of applying. If the stick was too hard,
high pressure was required for applica tion, the stick itself could break or the parts to
be glued together, such as thin paper might be damaged during a rubbing. If the
product was too soft, there would be too much glue left on the paper.
The Instron
Universal Testing Instrument Model 4502 (Instron Corporation,
Massachusetts) was used to measure hardness of glue stick. Compression force

measurement was conducted, see diagram in Figure 3.3. A 3
rnrn
diameter probe was
driven onto the sample ( 1 .5 cm diameter and 3 cm length) at crosshead speed 1 0

mm/min. The depth o f penetration was preset s o that each test had the same
mechanical parameters. The maximum load was measured. Three replications were
conducted for each sample and the average data was used.
P r ob e
Figure 3.3
--
Hardness testing diagram
51
3.3.3
Melting Point
Melting point was measured to determine the temperature at which the stick began to
change from solid phase to liquid phase. This indicated the temperature at which the
stick became soft and might lose its shape if applied with a hard force.
Melting point was measured using Sofi 6920 hot stage melting point measurement
(Leitz Wetzlar, Germany). A small sample was mounted on a slide and covered with
a coverglass. This sample was then gradually heated at the rate of 2 C per minute. The
temperature at which glue became transparent was measured. Two replications was
conducted for each sample.
3.3.4
Moisture content
Moisture content measured in this study included moisture and other volatile
substances in glue stick. The moisture content was determined by weighing
approxima tely 2 g of the glue stick sample into an aluminium moisture dish, which had
been dried in the drying oven at 1002 C for three hours and cooled for one hour in
a desiccator beforehand. The moisture dish and the sample was weighed, with the lid
in place, to the accuracy of 0.0001 g. Then the moisture dish which contained the
sample was dried in the oven at 1 002 C for 24 hours with the lid opened. After
cool ing in a desiccator for one hour, the lid was replaced and the moisture dish with
the dried sample was weighed. The weight loss was calculated as the moisture content
in percentage. Two replications were conducted and the average data was used.
3.3.5
Water activity
Water activity of the sample was measured using CX-1 water activity meter (Decagon
Device, Inc., Pul man, Washington, D.C.). The sample was placed in a disposable sample
c u p u n ti l i t fi l l ed h a l f the
CLIp
(abo u t 3-4 grams
of s a m p l e
was used . )
Then the sa mp l e
cup was put in the sample drawer. When the sample reached equilibrium, the water
activity was read from the display.
52
3.3.6
Amount of Glue Appl i ed per Area
This method was adapted from the ASTM standard testing method DS9S-90 Applied
I
weight per unit area of dried adhesive solids' (ASTM, 1 991a). The amount of glue
applied per area indicated how much glue was deposited on the paper while applying.
The amount of glue should be neither too high nor too low. If too much glue was left
on the paper, it was wasteful, messy and caused wrinkling of the paper, if too small an
amount of glue was left on the paper there was inadequate glue to make a strong bond
between surfaces and the consumer had to rub the glue onto the paper many times
which could cause damage to the paper.

The paper used for this test was the SO g/m2 white paper normally used for report
writing or photocopying. The surface area of each test paper (6 ern x 6 em) was
calculated to an accuracy of 1 %. Then the paper was conditioned at 231 C and 505%
relative humidity for 48 hour. The weight of the test paper was determined to the
accuracy of 0.0001 g.
Thirty six strokes of adhesive were applied on the conditioned paper. The paper was
reweighed immediately. Then the coated paper was dried in the drying oven at 1 001
C for 14 hours. At the end of the heating period, the paper was removed to a
desiccator and cooled to room temperature. Then the paper was weighed immediately
upon removal from the desiccator. Three replications were tested for each sample and
the average data was used.
The weight of adhesive applied was calculated as follows:
where
S
weight of wet adhesive applied, expressed in g / m2 of surface

area
Wo
original uncoated weight of the paper (g)
WI
weight of paper (g) immediately after application of the

adhesive
area of test paper, m2
53
This measurement gave the amount of glue applied per surface area (wet glue per area).
From this data, it was possible to know how much glue residue was left on ,the paper
after a number of applications.
The weight of dry adhesive applied was calculated as follows:
= [ W2 - Wo( 1 - k)/A]
where
weight of dry adhesive applied, expressed in g/m2 of surface
area
W2
= weight of paper (g) after application of the adhesive and
elimination of solvents,
k
= a factor applied to correct for changes in paper weight that
occur during the solvent elimination process. It was obtained by

weighing an uncoated paper and then exposing i t
simultaneously with the coated paper to the solvent evaporation
proced ure. The average k was obtained from five replication.
k was calcula ted as follows:

where
Mo
= original weight of duplicate paper, and M J = weight after
exposure to solvent elimination.
This measurement gave the amount of dry glue applied per surface area (dry glue per
area). This indicated how much glue residue was left on the paper after drying.
3.3.7
Open time
The open time was the interval during which the adhesive remained bondable after
being applied onto the paper. The open time had to be sufficiently long to allow the
pieces which were to be adhered to be positioned in contact but not so long that there
was an undue wai t for the firm bond to develop. Preferably the open time of the glue
stick was at least about one minute, with a maximum of 10 minutes (Pletcher and
54
Wong, 1 978).
The paper used for testing of open time was 5 cm x 5 cm of 80 g / m2 paper. To
determine open time, 18 strokes of the glue stick sample were applied on 3 cm x 3 cm
area of the test paper. The open time for an adhesive stick was the elapsed time
between the end of the rubbing and the point at which the adhesive on the paper
reverted to its non-bondable state. Whether the adhesive was still bondable after a
particular interval was determined by placing a piece of 3 cm x 3 cm 80 g / m2 paper in
contact with the adhesive on the first piece of paper and pressing i t down with four
passes (twice in each direction) of a 2 kg roller. The composite sample was allowed to
stand for 30 minutes and the two pieces of paper are then pulled apart by hands. If
delamination occurred, the adhesive was in open state when the other piece of paper
was applied. If delamination did not occur, the adhesive had already reverted to the
closed, nonbondable state before the other piece of paper was applied. The open time
for an adhesive stick was the maximum interval between the end of the rubbing cycle
and the application of the paper to the coated paper which resulted in bonding - the
adhesive was still in bondable stage. To determine this, several composite samples had
to be prepared at different intervals and tested for delamination.
3.3.8
Peel Strength of the Adhesive Bond
Stickability is one of the most important attributes of glue products. This attribute was
tested to make sure that the glue did its job properly i.e. stuck two pieces of substrates
together. The substrate used in the test depends on the purpose of glue usage. Since the
glue stick developed in this research was to stick paper together, paper was used as test
substrate.
The test method was from the ASTM standard testing method 0903-49 (ASTM, 1991b).
The test sample consisted of one piece of paper, 1 in. x 12 in. (25
mm
x 304.8 mm),
coated with 10 strokes of adhesive for 6 in. (152.4 mm) at one end and bonded to
another piece of paper, 1 in. x 8 in. (25 mm x 203.2 mm), with the unbonded portions
of each member being face to face. Samples were conditioned for 7 days by exposure
to a relative humidity of, 502 % at 231 dc.
55
Gr ip
aper
A l i gnme n t
plate
Figure 3.4
Peel strength testing diagram
The sample was placed in the Instron Universal Testing Instrument Model 4502 (Instron
Corporation, Massachusetts) by clamping the free end of the 8 in.-Iong paper in one
grip, the free end of the 12 in.-Iong paper was turned back and clamped in the other
grip (see Figure 3.4). The test paper was maintained approximately in the plane of the
clamps during the test. This was done by holding the paper against an alignment plate
attached to the stationary clamp.
56
The separating members of the sample were stripped approximately at an angle of 180
degree. The rate of travel of the power actuated grip was 12 in. (305 mm) / min. This
rate which provided a separation of 6 in. (152.4 mm) /min had to be uniform
throughout the test. The sample was peeled at least one half of the bonded area, even
though a peel strength value had been indicated before this point. Th maximum peel
strength was recorded. This measurement indicated the strength of bond after drying.
3.4
INGREDIENTS AND THEIR PROPERTIES
3.4.1
Starches
The raw starch used in this study was obtained from First Victor Co., Ltd., Bangkok,
Thailand: The modi fied tapioca starch used in this study was 'National Frigex' which
was cross-linked and stabilized starch. This type of starch could withstand the high
temperature in an alkaline system wi thout changing in colour. The modified tapioca
was obtained from National and Chemical Starch Ltd., Auckland, New Zealand for the
laboratory scale experimentation and from National and Chemical Starch Ltd., Bangkok,
Thailand for the pilot scale production.
3.4.2
Polyvinyl Pyrrolidone
Polyvinyl pyrrolidone (PVP) is prepared by the polymerization of N-vinyl-2pyrrolidone, a colourless liquid. It is manufactured in four viscosity grades identified
by their K-value, which are approximately K-15, K-30, K-60, and K-90. The number
average of the molecular weights for these grades are about 10,000, 40,000, 160,000, and
360,000, respectively.
The PVP used in this study was GAF PYP K-90 with average molecular weight 360,000.
The sample was obtained from ISP (Australasia) Pty. Ltd . and ISP (Hongkong) Ltd.,
Bangkok representative office.
57
3.4.3
Glycerol
Glycerol is used as plasticizer in adhesives to impart the finished film characteristics to

adhesives. It makes the film dry slowly withou t becoming brittle. The quality of the
plasticizer and its level of use must be such that it gives the desired bond flexibility
without excessive stickiness or blocking. Glycerol is a better plasticizer than are sugars
and does not cause darkening of the bond with age. The glycerol used in this study was
Glycerine BP (GLY 510) from Bronson and Jacobs Limited, Auckland, New Zealand.
3.4.4
Stearic Acid
Stearic acid was used in the glue stick in order to react with sodium hydroxide in the
system to give sodium stearate which acted as a gel forming agent in glue stick. Stearic
acid is reported to increase starch gelatinization temperature. The stearic acid used in
this study was stearic acid STE 922 obtained from Bronson and Jacobs Limited,
Auckland, New Zealand.
3.4.5
Sodium Hydroxide
In the glue stick formulation, sodium hydroxide reacted with stearic acid and gave
sodium stearate which performed as the gel forming agent in the system. The sodium
hydroxide used in the laboratory scale experimentation was laboratory grade sodium
hydroxide from BDH Limited, England. The sodium hydroxide used in the pilot scale
. was industrial grade obtained from Arsrom Co., Ltd., Bangkok.
3.4.6
Glyceryl Monostearate
Glyceryl monostearate was used as an emulsifying agent in the system. The glyceryl
monostearate was Lexemul 561 (LEX 657) obtained from Bronson and Jacobs Limited,
Auckland, New Zealand.
3.4.7
Brij 35
Brij 35, polyoxyethylene lauryl, was used as an emulSifying agent. It was a non-ionic
58
surfactant stable in acid and alkaline solution. The Brij 35 used in this study was
obtained from BDH Limited, England.
3.4.8
Dextrin
Dextrin sample was Dextrin MW obtained from N.Z. Starch Products, Ltd., Auckland,
New Zealand. It was white dextrin with 1 0% moisture content. Its viscosity at 1 7%
solid, 40 C, was 25.0-40.0 cpo
3.4.9
Casein
Sulfuric acid casein was obtained from New Zealand Dairy Research Institute. Its
molecular weight was 20,000. The casein sample contained 1 0% moisture content. Its
viscosity at 15% solid, 25 DC was 30 poise.
3.5
PROCESSING OF GLUE STICK
The processing method described here is the method used in the laboratory
experimentation. The method which was used in the production trial is described in
Chapter 8.
3.5 . 1
Equipment
Equipment used in the laboratory scale production of glue stick was as follows:
>I-
A 500 ml round bottom flask with a 5-neck lid
>I-
Motor driven stirrer
>I-
Thermometer
>I-
Condenser
>I-
Water bath
59
Figure 3.5
Laboratory scale equipment
60
The mixing equipment comprised of the 500 ml round bottom flask with 5-neck lid
which was connected to a reflux system, see Figure 3.5. The reflux system included a
spiral glass condenser fitted with two rubber tubes, one connected to cold water tap and
the other one to let the water out from the system. A blade motor-driven stirrer was
inserted through the neck of the lid. The water-bath comprised an aluminium bath,
thermostat, pump for water-circulation, thermometer, and aluminium foil lid.
3.5 . 2
.
Method of Processing
At the beginning of the process, starch was mixed with cold water in the flask to form
a starch slurry. Then the other ingredients were added. The mixture was stirred
manually so that the ingredients were thoroughly mixed together before being exposed
to the heat. The flask containing the mixture was then put in the water-bath, held at
temperature 901 C and the stirrer was connected. Condenser was fitted and cold
water was turned on. The glue mixture was heated under reflux. The water in the water
bath was then gradually heated up to 991 dc. The mixture was stirred using high
speed (30 rpm) for the first 10 minutes then low speed (15 rpm) until the process was
finished. About every 10 minutes, the stirrer was stopped in order to let the mixture be
stirred by hand lIsing a glass-rod stirrer. This had to be done regularly because some
part of the mixture at the bottom was not mixed by the stirrer.
Speed of the stirrer was important for emulsification of the ingredients, too Iow a speed
would not give good emulsification but too high a speed might break the emulsion
system.
The reason that the mixture was heated at a low temperature at the beginning was to
a llow the starch to slowly gelatinize and at the same time stearic acid, glyceryl
monostearate, sodium hydroxide and PVP dissolved into the mixture. If the starch was
heated up rapidly, it tended to give lumps of gelatinized starch and this could not be
easily mixed with other ingredients.
At the end of the process, the viscous hot mass was poured into a 10 ml plastic beaker.
The finished product was allowed to cool down at room temperature for an hour. Then
it was cut into sticks using a 1 .5 cm diameter plastic tube and mounted in the
61
containers normally used for 8 g commercial glue sticks. Glue stick was aged a t least
5 days before testing.
3.6
EXPERIMENTAL PLANS
Constrained mixture designs were used as the experimental plan, since in the
experiments changing the proportion of one component affected the proportion of other
components in the system. Snee and Marquardt (1974) recommended tha t the extreme
vertices for the mixture design containing 'q' components can be computed using the
XVERT algorithm described below.
(i)
Rank the components in order of increasing ranges (upper limit - lower

limit) or (bi-a} X I has the smallest range, and X q has the largest range.
(ii)
Form a two-level design from the upper and lower bounds of the q-1
components with the smallest ranges.
(iii)
Compute the level of the qlh component
(vi)
A given point is an extreme vertex if aqSXqSbq . For those points which

are outside of the constraint limits, set Xq equal to the upper or the lower
limit, whichever is closest to the computed value.
(v)
From each point originally outside of the limits, generate additional

points (max =q-l) by adjusting the level of one component by an amount
equal to the difference between the computed value for Xq and the
substituted upper or lower limit. Addi tional points are generated only
from those components whose adjusted level remain within the limits of
the components.
62
Table 3 . 1
Six components extreme vertices

Stearic
NaOH
Starch
PYP
Water
Min
10
10
38
Max
16
16
45
14
12
17
10
16
45
14
12
HJ
16
44
14
12
16
10
45
14
12
16
10
44
14
12
16
16
42
14
16
16
41
14
16
16
45
12
16
16
39
14
12
16
16
44
12
10
16
16
38
14
12
C.P.l
1 4.8
14.8
42.7
1 2.8
1 1 .4
3.5
C. P.2
1 4.8
14.8
42.7
1 2.8
1 1 .4
3.5
RUN
Range
Glycerin
CORE MATRIX
ADDITIONAL POINTS
11
15
10
45
14
12
12
10
15
45
14
12
13
13
16
45
14
14
10
16
45
14
11
15
12
16
45
14
16
10
16
45
13
12
17
15
16
45
12
18
16
13
45
14
19
16
10
45
14
11
20
16
12
45
14
21
16
10
45
13
12
22
16
15
45
12
23
16
16
45
11
24
16
16
45
11
25
16
16
45
10
63
The results shown in Table 3 . 1 were obtained from the experimental plan used for
prototype development in Section 5.2.2. Only core points and 2 centre points were
chosen for conducting the experiment. Centre points were calculated by averaging the
coordinates of the 10 vertices.
Twelve runs (core matrix and 2 runs from the centre point) were used to develop
prototype products. Other constrained mixture designs used in this project were
calculated in the same manner and are shown in Sections 5.2.3 and 5.2.4.
3.7
DATA PROCESSING METHOD
There were 5 main computer programmes used
111
the processing of data obtained
during the optimization process:

* VP Planner Plus (Paper Software International, California)
* Minitab 8.2 (Minitab Inc., Pennsylvania)
* Stat-Packets (Walonick Associates, Inc., Minneapolis, MN)
* LP88 (Eastern Software Products, Inc., Alexandria, Virginia)
* SPSS / PC+ version 4.0 (SPSS Inc., Chicago, IL)
Data were put into VP Planner before being analyzed by the other programmes. VP
Planner was used to calculate means and standard deviations of the data. It was also
used for transformation of data into ideal ratios and logs of ideal ratios.
Minitab 8.2 was used for correlation, stepwise multiple regression and analysis of
variance of the data obtained from physical testing and sensory testing of the product
prototypes.
Stat-Packets was used for determination of significance using the t-test. LP88, the linear
programming package, was used to generate optimum glue stick formulation in the
o p t i m i za t i o n s tage.
SPSS/ PCt
was lIsed fo r cross-tabu l a t i on o f the d a ta
the final consumer testing in Chapter 9.
obtained from
64
CHAPTER
CONSUMER STUDY
The consumer study was conducted in order to study the consumers' perspective
towards glue products especially glue sticks. The primary consumer study was carried
out with Thai and New Zealand consumers in New Zealand so as to identify the
important attributes of the glue sticks. The problems of using glue products were also
identified. Then a consumer survey was conducted with Thai consumers in Thailand
in order to find information on glue product usage, and the importance of glue product
a ttributes particularly of glue sticks and the possibility of introducing the developed
product to the market.
4.1
AIM AND OBJECTIVES
The aim of this study was to develop the product concept with consumers. The
objectives were to:
* Obtain information regarding glue product usage:

* Identify important attributes of glue products.
* Identify the differences between the two cul tures as regards glue usage.
* Generate direction for improvement of glue products.
* Generate a product concept.
65
4.2
METHOD OF CONSUMER STUDY
4.2.1
Products of Interest
This study was focused on the direction of improvement of glue stick. As mentioned
before, glue stick could be in the form of permanent bonding or temporary bonding so
the decision on which type of glue stick to be developed had to be made. The
permanent bonding glue stick is a solid adhesive in lipstick-type
container. The
adhesive is usually intended to give a permanent bond when applied to paper or

cardboard. The advantages of this product when compared to other glue products used
for paper work are: easy to apply, does not seem to soak through paper, less messy,
short drying time, easy to carry in pocket. For the temporary bonding glue stick, the
glue coated surface can be adhered to a receiving surface such as paper, notice board,
desk-top and can be removed from this without leaving traces of adhesive on the
receiving surface, and this can be repeated a number of times. This product has
advantage over the self-adhering note pad in that the latter has a limited field of
utiliza tion, although i t in itself functions very well.
As permanent bonding glue stick products were already on the market, the commercial
products were given to consumers as a reference for generating their opinions towards
the product. However, temporary bonding glue stick was not available on the market
at the time, so the self-adhering note pad precoated with glue, which had similar
properties as that from temporary bonding glue stick, was used.
Although the study was focused on glue stick, other glue products such as: liquid glue,
paste glue, PYA glue, cellotape and 2-sided tape were also included in the study in
order to get general information on pattern of glue usage from consumers.
4.2.2
Stages of Survey in New Zealand and Thailand
The first consumer study was conducted in New Zealand with Thai and New Zealand
consumers. They were 21 Thai students in various Faculties at Massey University and
21 New Zealand students in the Technology Faculty, Massey University. This survey
was done in order to preliminary explore the general glue product usage by consumers
66
and to identify any problems consumers had when using glue products. Important
a ttributes of glue stick and self-adhering note pad were generated by the consumers.
The second consumer study was conducted in Thailand with Thai consumers. They
were 1 7 undergraduate students, 17 post-graduate students (both groups were studying
at Kasetsart University, Bangkok, Thailand) and 17 government office staff. In this
study, the importance of the product attributes elicited from the former study were
measured to confirm that they were important according to the consumers in Thailand.
Buying intention, p rice and size or the glue stick product to be developed were also
obtained. This was further used to develop the concept of the product.
4.2.3
Method Used in Generation of Important Attributes
Normally the important attributes of products can be obtained from the literature in the
same a rea or from t h e
company
re p ortt;.
However, i f
t he re
is no such information
available the list of the important attributes has to be developed for that specific
product. It is widely known that the important attributes perceived by consumers are
d ifferent from those perceived by researchers. In order that the i mportant a ttributes to
be used in further study were the same as the ones consumers used in making decisions
of glue selection, consumers were employed in establishing the l ist of important
a ttributes. In this study, the consumers in New Zealand were asked to think abou t any
problems they had with using glue products and write it down in the questionnaire (see
Section 4.2.5). Samples of a commercial glue stick, 'UHU' (GmbH, Germany), and self
adhering note pads, 'Post-it' (3M), were also given to consumers as references. After
that they were told to give a list of attributes they considered important for the
products. With this method, the consumers were reminded to think of the problems and
they described them in the first part of the questionnaire, so they could identify the
attributes they thought were important in the following part of the questionnaire.
4.2.4
Method of Measuring Importance of Attributes
There were two methods used in this study: open-ended elicitation measure and direct
rating method. Both methods were used in the survey conducted in New Zealand. Only
the direct-rating method was used with consumers in Thailand.
67
In the elicitation measure, consumers were asked to give lists of product attributes that
were important to them in evaluating a glue stick and a self-adhering note pad.
Responses were analyzed to determine what attributes were mentioned and in what
order. An importance index for a given attribute was defined for each individual to
incorporate order of elicitation. This involved dividing the rank order of elicitation of
the attribute (where 1
last elicited attribute through n
first elicited attribute) by the
number of a ttributes elicited by the subject. If an attribute was not mentioned, the index
number for that attribute was set to zero (Jaccard et aI., 1986).
In the direct-rating method conducted in New Zealand, consumers were asked to

evaluate the importance of the given attributes of products using the linear scale
containing scores from 0 to 1 0; 0
not important, 10
very important. Product
a ttributes used were obtained from a literature survey and the researcher's opinion. In
the study conducted in Thailand, scores 1 (not important) to 5 (very important) were
used and the list of attributes were obtained from the results of the consumer study in
New Zealand.
4.2.5
Questionnaires for Consumer Survey in New Zealand and Thailand
The questionnaire for the consumer survey in New Zealand comprised questions related
to glue product usage especially glue stick, the problems consumers had with product
usage, improvement of glue products, important attributes of glue products. The
questionnaire used with Thai consumers was translated into Thai. Questionnaire testing
was conducted before the survey with 3 Thai students and 3 New Zealand students.
Alteration of the questionnaire was made where necessary. The questionnaire used in
the survey is shown in Appendix 4.1 .
The questionnaire for the consumer survey in Thailand was written in Thai. It
comprised questions related to glue product usage, rating of glue a ttributes' importance,
buying intention, size and recommended price of product to be developed. The
questionnaire was pre-tested with 6 Thai students at Massey University. The
questionnaire including the English translation are shown in Appendix 4.2.
68
4.2.6
Survey Method
A self-administered survey was used in both surveys. In the survey conducted in New
Zealand, as the number of respondents was small a 'drop and collect' survey was used.
The questionnaires as well as the product samples were handed out to the respondents.
They were allow enough time to fill out the questionnaire, approximately 3-4 days then
return it back.
I n the survey conducted in Thailand, the questionnaires and the productsamples were
given to the representatives of each group of consumers who handed them to the
consumers. After the questionnaires were finished the consumers returned them back
to their representatives.
4.3
DATA PROCESSING
The data from the consumer study were coded and input in the VP Planner then
analyzed using SPSS/PC+ computer programme. For the answers from open-ended
questions, the coding frames for all the open-ended questions were developed, each
response was read and a judgement made as to which code frame category it matched.
The appropriate code was then given to that response ready to be input. This method
was described by Hague and Jackson (1990).
4.4
PATTERN OF GLUE USAGE
The percentages of users of each glue product are shown Table 4.1 . The pattern of glue
usage of the students and the office workers in Thailand were not significantly different
so the data were grouped and presented as the total . Of the 93 respondents, 98 % used
cellotape and 95 % used glue stick and only 56 % used self-adhering note pad. There
were differences in the pattern of usage of self-adhering note pad between Thais in
Thailand and Thais in New Zealand, only 25 % used them in Thailand but 1 00 % used
them in New Zealand.
69
Table 4.1
Users of each glue product
Products
Total
Cello tape
Thais (in Thailand)
Thais (in New Zealand)
91 (98%)
49 (96%)
21 (100%)
21 (100%)
Glue stick
88 (95%)
50 (98%)
20 (95%)
18 (86%)
PYA glue
75 (81 %)
39 (76%)
1 9 (90%)
17 (81%)
2-sided tape
74 (80%)
37 (73%)
1 8 (86%)
19 (90%)
Liquid glue
70 (75%)
31 (61%)
21 (100%)
18 (86%)
Paste glue
54 (58%)
21 (4] %)
17 (8] %)
16 (79%)
Self-adhering
note pad
52 (56%)
13 (25%)
21 (100%)
18 (86%)
Note:
New Zealanders
Percentage is given out of the total number of respondents, i.e. 51 Thais (in Thailand), 21 Thais (in
New Zealand), 2] New Zealanders and 93 total.
The frequency of using the different glue products also varied as shown in Table 4.2.
Consumers used cellotape most often, only 9 percent of the respondents used it less
than once a month. Glue stick was used more often than the other glue products (64%
used it more than once a month) and paste glue was used least often only 1 1 percent
used it once a month or more. The frequency of using self-adhering note pad, PVA
glue, liquid glue and 2-sided tape were not very different - 34-41% of the respondents
used them more than once a month.
70
Table 4.2
Frequency of glue product usage
Products
Consumers
3 times a week
or more
Cellotape
Thais in Thailand
23 (45%)
22 (43%)
6 (12%)
Thais in N .Z.
6 (29%)
1 3 (62%)
2 (10%)
New Zealanders
8 (38%)
13 (62%)
TOTAL
37 (40%)
48 (52%)
8 (9%)
Thais in Thailand
14 (28%)
32 (63%)
5 (10%)
Thais in N.Z.
3 (14%)
7 (33%)
1 1 (52%)
New Zealanders
2 (10%)
2 (10%)
17 (81%)
TOTAL
19 (20%)
41 (44%)
33 (35%)
Thais in Thailand
3 (6%)
25 (49%)
23 (45%)
Thais in N.Z.
2 (10%)
1 (5%)
18 (86%)
New Zealanders
2 (10%)
TOTAL
7 (8%)
Glue stick
PYA glue
2-sided tape
Liquid glue
Thais in Thailand
19 (90%)
26 (28%)
60 (65%)
25 (49%)
26 (51%)
1 7 (81%)
4 (19%)
New Zealanders
7 (33%)
2 (10%)
12 (57%)
TOTAL
1 1 (12%)
27 (29%)
55 (59%)
Thais in Thailand
2 (4%)
17 (33%)
32 (63%)
Thais in N .Z.
5 (24%)
2 (10%)
14 (67%)
New Zealanders
4 (19%)
2 (10%)
15 (71%)
1 1 (12%)
21 (23%)
6 1 (66%)
1 (2%)
8 (6%)
42 (82%)
Thais in Thailand
21 (100%)
Thais in N .Z.
New Zealanders
TOTAL
Self-adhering note
pad
Less than once a

month and not used
Thais in N .Z.
TOTAL
Paste glue
. Once a month to
2 times a week
20 (95%)
(5%)
2 (2%)
Thais in Thailand
8 (9%)
83 (89%)
13 (25%)
38 (75%)
Thais in N .Z.
5 (24%)
6 (29%)
10 (48%)
New Zealanders
3 (1 4%)
5 (24%)
13 (62%)
TOTAL
8 (9%)
24 (26%)
61 (66%)
71
The purposes of glue product usage obtained from the survey in New Zealand are
shown in Table 4.3.
Table 4.3
Purposes of glue product usage
Purposes
Cellotape
Glue
stick
Reports (photos, graph,

diagram)
10 (26%)
1 5 (75%)
4 (27%)
Seal envelope flap
1 1 (29%)
7 (35%)
2 (13%)
Wrapping presents
5 (1 3%)
Other p urpose for sticking

paper together
19 (50%)
9 (45%)
10 (67%)
Put up notice on board
1 1 (29%)
Stick paper to nonpaper
3 (8%)
Stick non paper to non paper
6 (56%)
Liquid
glue
1 (7%)
PYA
glue
Paste
glue
2-sided
tape
3 (38%)
8 (67%)
3 (38%)
1 (100%) 4 (33%)
3 (38%)
2 (17%)
3 (38%)
1 (8%)
19 (1 00%)
Writing message or note

Total
Note:
Note pad
38
20
15
12
19
The percentage is given out of the total number of respondents who used the product as one of the
three products they used most often.
The purposes of glue usage were grouped into categories that had a similar meaning.
Most of the products were used for sticking paper together. However, some products
were also used with non-paper materials as well . It was found that cellotape seemed
to have more applications than other products followed by glue stick, 2-sided tape, PVA
glue and liquid glue. Self-adhering note pad was used for writing messages or as a
reminder.
4.5
ATTRIBUTES OF GLUE STICKS WHICH SHOULD BE IMPROVED
Consumers were asked to define the attributes of glue stick that should be improved.
The responses were grouped into categories and the results are shown in Table 4.4.
--- ---
72
Table 4.4
Attributes of glue stick product needing improvement
A ttributes
Total
Thais (in Thailand)
Thais (in N .Z.)
New Zealanders
Ease of use
38 (52%)
23 (45%)
1 3 (72%)
2 (50%)
Stickability
16 (22%)
10 (20%)
5 (28%)
1 (25%)
Aesthetic
1 5 (21%)
10 (20%)
5 (28%)
Uniformity of
coating
1 3 ( 18%)
6 ( 1 2%)
6 (33%)
1 (25%)
Price
1 1 ( 15%)
1 1 (22%)
Cleanliness
10 (1 4%)
8 ( 1 6%)
1 (6%)
1 (25%)
Size
10 ( 14%)
5 ( 1 0%)
5 (28%)
Effect on paper
6 (8%)
5 (1 0%)
(6%)
Drying time
4 (5%)
3 (6%)
(6%)
Versatility
3 (4%)
2 (4%)
Refil l
2 (3%)
2 (4%)
Keeping quality
2 (3%)
2 (4%)
Heat/water
resistance
Note:
(1;'.)
1 (25%)
1 (6%)
The percentage is given out of the total number of responJents who answered the question, i.e. 51
Thais (in Thailand), 18 Thais (in New Zealand), 4 New Zealanders, and 73 total.
Ease of use was the attribute that the most consumers recommended should be
improved followed by stickability, aesthetic and uniformity of coating.
The attributes of self-adhering note pad which should be improved according to the
results obtained from consumer study in New Zealand are shown in Table 4.5.
73
Table 4.5
Attributes of self-adhering note pad needing improvement (Panel in

New Zealand only)
Attributes
Total
Thais (in N .Z.)
New Zealanders
Stickability
12 (63%)
8 (67%)
4 (57%)
Size
7 (37%)
5 (42%)
2 (29%)
Aesthetic
5 (26%)
5 (42%)
Price
2 (1 1 %)
1 (8%)
Ease of use
1 (5%)
Removability
(5%)
1 ( 1 4%)
1 (14%)
1 (8%)
Stickability was the attribute that 63% of the New Zealand panel wanted to be
i mproved, they wanted the product to have a stronger bond .
4.6
IMPORTANT ATTRIBUTES OF GLUE STICK AND SELF-ADHERING NOTE

PAD
As no information was found on the important attributes of glue products, it was

necessary that the list of important attributes was generated at the beginning of the
product optimization process. The New Zealand panel including Thais and New
Zealanders were employed to generate the list of important attributes of glue stick and
self-adhering note pad products. In the consumer study in New Zealand, consumers
were asked to describe the attTibutes they considered important for a glue stick and a
self-adhering note pad. There were 48 important attributes identified by New
Zealanders and 28 attributes identified by Thais for glue stick. There were 32 and 17
a ttributes respectively identified as i mportant for self-adhering note pads. The list of
important attributes of a glue stick and a self-adhering note pad given by consumers
are shown in Table 4.6 and Table 4.7 respectively.
74
Table 4.6
The attributes of a glue stick product
Attributes
Description
.. Appearance
Colour
Size
.. Smell
- Colour of glue
- Volume and quantity
- N ice shape
- N ice perfume
- No awful smell
.. Ease of use
Lid
Screw bottom
- Ease of getting lid on and off / Using a screw off or flip-top

cap
- Not get 'glued up'
- Ease of wind up and down
The stick
- Hard ness, does not go out of shape when

applied
- Call be used to the very end
Dispensation
Shape
Adjustability
Dispensing technique
Ease of spreading
Smooth roll up
Uniformity of coating
- Easy to carry
- H andsize for accuracy
- Ability to slip the paper around to position it
Ease of applying on paper - Slipperiness / ease of spreading

- Uniformity of coating
.. Effect on paper
Not soak through the paper

Must not colour or discolour the paper
Must not affect the print on the paper
Cannot be seen through the paper
Does not wrinkle the paper
.. Stickability
- Strength of bond
- Long lasting
.. Cleanliness
- Cleanliness of work
- Cleanliness of container after use
- Ease of cleansing from hands.
.. Other attributes
Heat resistant
Water solubility
Keeping quality / storage life
Multipurpose
Amount of glue needed to stick paper together
Drying time / Ability to d ry evenly
.. Image
Price/ Value for money

Non toxic/environmental friendly
Manufacturer/Brand name
Country of manufacture
75
Table 4.7
Important attributes of self-adhering note pads given by the consumers

Attributes
* Appearance
Colour
Bright colour / easy to see

Variety of colour
'
Decent size / variety of size
Number of pages per pack
Rectangle or square
* Ease of use
Ease of pulling from the pads

Easy to write on
A good quality paper
Ease of pulling away from the receiving surface
* Performance
Stickiness
Reattachability
Versatility / ability to stick on fabric
Ability to stick well on surface without curling of edge
* Damage of substrate
- No damage on the receiving surface

- Clean removal
- No discolouring of surface
Size
Shape
Image
* Others
4.7
Descriptions
Price/value for money

Environmental friendly
Manufacturer/ Brand name
Country of manufacture
- Quality of paper
- Ability to be used on both sides / gum on both sides
IMPORTANCE MEASURE OF GLUE PRODUCT AITRIBUTES
The New Zealand panel were employed to evaluate the importance of attributes of glue
stick and self-adhering note pad products by open-ended elicitation and d irect rating
methods. In the consumer survey in Thailand, the consumers evaluated the importance
of the attributes of glue products in general and of the glue stick products to be
developed in this study by direct rating method.
76
4.7.1
Importance of Attributes Measured by Open-ended Elicitation Method with

Panel in New Zealand
In this method, it was assumed that order of elicitation was sensitive to attribute
i mportance. An index number was given to the attribute according to the order of
elicitation (see Section 4.2.4). Sum of index numbers of each glue stick attributes and self
adhering note pad are shown in Table 4.8 and 4.9 respectively. The higher the index
number, the more important the attribute was to the consumers.
Table 4.8
Importance of glue stick attributes from open-ended elicitation method
Attributes
Total (42)
Thais (21)
New Zealanders (21 )
Ease o f use
18.9 1
5. 92
1 2.99
Stickability
1 7.48
7.26
1 0.22
Cleanliness
16.33
5.80
1 0.53
Price
1 4.49
6.76
7.73
Size
10.16
4.42
5.74
Uniformity of glue coating
7.75
5. 95
1 .80
Hardness of the stick
5.57
3.35
2.22
Drying time
3.64
0.25
3.39
E ffect on paper
3.02
1 .92
1.10
N o awful smell
3.01
1 .58
1 .43
Nice perfume
2.27
0.67
1 .60
Colour of the stick
2.26
0.83
1 .43
Versatility
1 .94
0.50
1 .44
Keeping quality
1 .32
0.99
0.33
Brand name
1 .57
1 .57
0.00
Amount of glue needed
1 .20
0.00
1 .20
Thickness of adhesive film
0.33
0.33
0.00
The results show that ease of use, stickability, cleanliness, and price were the important
attributes of glue stick according to both groups of consumers.
77
Table 4.9
Importance of self-adhering note pad attributes from open-ended

elicita tion method
A ttribute
Total (42)
Thais (21 )
Size o f paper
22.46
1 ] .05
1 1 .41
Stickability
1 8.20
4.50
1 3.70
Ease of use
11 .78
8.88
2.90
Price
1 0. 1 7
4.89
5.28
Damage of receiving surface
8.21
4.76
3.45
Colour of paper
7.84
3.30
4.54
Force needed to pull paper
4. 9 1
1 .33
3.58
Versatility
3.53
2.53
1 .00
Brand name
0.88
0.50
0.38
0.50
0.50
0.00
Keeping quality
0.44
0.31
0.13
The results show that size of paper had a high index number so i t was a very i mportant
a ttribute for most consumers. However stickabiJity seemed to be very important
according to New Zealand consumers. Ease of use was also considered important by
Thai consumers. Brand name, uni formity of coating and keeping quality were not very
i mportant as they gained very low index numbers.
4.7.2
Importance of Attributes Measured by Direct-rating Method with Panels in

New Zealand and Thailand
The results of importance measure using direct-rating method obtained from the survey
conducted in New Zealand are shown in Tables 4.10 and 4.1 1 .
E ffect on paper, ease of use, drying time, uniformity of coating, keeping quality and
amount of glue needed to stick paper together were considered as important attributes
for glue stick by the consumers (see Table 4.10) . However hardness and no awful smell
were regarded as important by Thai consumers and price was also considered
important by New Zealand consumers.
78
Table 4.10
Importance of glue stick product attributes from panel in New Zealand

(scores varied from 0
not important to 10
very important)
Attributes
Total (42)
Thais (21 )
New Zealanders (21)
Stickability
8,8'
9 .3'
8.4'
Effect on paper
8.8'
9 .3'
8.3ab
Ease of use
8.7ab
8.7'b
8.7ab
Drying time
8.2be
8.6'b
7.8,be
U niformity of glue coating
8.1 abe
8,6,b
7.6'bed
Keeping quality
'd
8,Oatx
8.6ab
7.3 bed
7.8 bed
8.1 ,bc
7.5, bed
Price
7.6 cd
7.5 be
7.8ab
7.5
cde
8 . 1 ,be
6.8 cd
Versatility
7.4
cd.
B .O be
6. 9 cd
No awful smell
7.4
cd.
8.1 be
6.8 cd
Thickness of adhesive film
7.1
de
7.6 be
6.7 cd
Size
6.7
6. 9 c
6.5
Nice perfume
3,1
3.7 d
2.6
Colour of the stick
3.7
4.0
3.3
Brand name
3.7
3.7 d
Note:
Mean Scores followed by a different letter are significantly different at p
Table 4.1 1
d
.
3.0
0.05
<
Importance of self-adhering note pads attributes from panel in New

Zealand (scores varied from 0
not important to 1 0
very i mportant)
A ttribute
Total (42)
Thais (21 )
Damage o f receiving surface
9 .1'
9.1'
9.1'
Ease of use
8.1 b
8 . 4 b
7.9b
Stickability
B. l b
B ,2b
8.O'b
Keeping quality
8.0 b
B .7b
7.2 h
Price
7.5 be
7.5b
7.5 b
Versatility
7.4 be
7 .6b
7.2 bed
7.0 be
7.5b
6.5 bed
7,0 bed
7.9ab
6.0 ed
Colour of paper
6.4
7.0e
5.8
cd
Brand name
3.2
3.7"
2.8
Note:
cd
c
Mean scores followed by a different letter are Significantly different at p
<
0,05
79
It can be seen that damage of the receiving surface is the most important attributes for
self-adhering note pad followed by ease of use, keeping quality, price, versatility, force
needed to pull paper away from receiving surface and uniformity of glue coating.
4.7.3
Importance of Attributes of Glue Products with Panel in Thailand
The consumers in Thailand were asked to rate the importance of attributes of the glue
products they normally used to stick paper together. The results from the consumer
study (see Table 4. 12) indicated that effect on paper and stickability were the most
i mportant attributes of glue products followed by uniformity of coating, cleanliness and
ease of use.
Table 4.12
Importance of glue product attributes from Thailand panel (scores varied

from 0
not important to 5
very important)
Attributes
Mean scores
Effect on paper
4.82a
Stickability
4.78ab
4.45 be
Cleanliness
4.37
Ease of use
4. 1 0
Drying time
4.00
3.78
Keeping quality
3.71
Versatility
3.49
Price
3.4 1
Odour
2.51
cd
de
el
elg
Ig
Igh
gh
h
Note: Mean score followed by a different letter are significantly different at p < 0.05,
using t-test.
They were also asked to rate the importance of permanent bonding and temporary
bonding glue sticks. The results are given in Tables 4.13 and 4.14 and compared with
the results obtained from the New Zealand panel.
80
Table 4.13
Importance of permanent bonding glue stick attributes
Attribute
Thais in Thailand (51)
Thais in N.Z. (21 )
9.3a
Effect on paper
4.8"
Cleanliness of work
4.8"b
Ease of use
4.7ab
B.7ab
B.7'
Stickability
4.7ab
9.3'
B.4'
4.5 b
B.6ab
7.6'bed
Thickness of adhesive fil m
4.2 c
7.6 be
6.7 cd
Price
4.
7.5 be
b
7.Ba
Drying time
4.2 c
7.B'be
4.0 c
7.5'bed
Cleanliness of container
4.0 c
3.5 d
Note:
6.8 cd
The numbers in the parentheses are number of consumers
Table 4.14
Importance of temporary bonding glue stick attributes
Attribute
Thai i n Thailand (51 )
Thai in N .Z. (21 )
Damage on surface
4.8"
9.1"
9.1"
Ease of use
4.3 b
B.4ab
7.9'b
Stickability
4.2 b
B.2ab
8 .O'b
Reattachability
4.2 b
Versa ti lity
4.1 be
7.6b
7.2ab
3.8 c
7.5b
6.5 bed
Note:
The numbers in the parentheses are numbers of consumers
4.8
ACCEPTABILITY OF ATTRIBUTES OF GLUE PRODUCTS BY PANEL IN

THAILAND
Thai consumers in Thailand were asked to rate their acceptability towards the attributes
of the existing glue products which included glue stick, l iquid glue, PYA glue and
cello tape. The acceptability of each attribute of the glue products was measured using
a 5 point category scaling: 1
not acceptable to 5
very acceptable.
----
----
-- --
--
81
From Table 4.15, it can b e seen that the average acceptability o f glue stick was close to
that of cellotape. Glue stick obtained higher acceptabil ity in terms of ease of use and
effect on paper but had lower acceptability in terms of drying time, versatility, price and
odour. Price of glue stick had lowest acceptability compared with other products.
Table 4.15
Acceptability of glue products by panel in Thailand
Attributes
Glue stick
Liquid glue
PYA glue
Cellotape
Ease of use
4.5
2.3
2.5
4.1
Cleanliness
4.3
2..0
2.1
4.1
E ffect on paper
4.2
1 .8
1.9
3.7
Drying time
4.0
2.0
2.2
4.7
Odour
3.7
3.0
2.4
4.1
3.7
2.2
2.6
3.9
Keeping quality
3.7
3.1
3.2
3.6
3.7
2.5
2.9
3.4
Stickability
3.6
3.1
3.6
3.8
Price
2.9
3.9
3.1
3.3
Versatility
2.7
2.5
3.4
3.3
M EA N
3.7
2.6
2.7
3.8
Note:
The number of respondents was 51
4.9
BUYING INTENTION, SIZE AND PRICE FROM PANEL IN THAILAND
The Thai consumers in Thailand were asked to indicate their buying intention for the
glue stick to be developed. Size of glue stick as well as the price which consumers were
prepared to pay were also asked.
82
Table 4.16
Buying intention of consumers towards the new product
Buying intention
Permanent bonding
Temporary bonding
Definitely buy
27 (54%)
21 (42%)
Probably buy
23 (46%)
22 (44%)
Not sure
6 ( 1 2%)
Probably not
1 (2%)
Definitely not
Note;
The number of respondent was SO because one consumer did not answer this question
Fifty four percent of the consumers said that they would definitely buy the improved
permanent bonding glue stick and 46 % of the consumers said that they would probably
buy the product. This indicates that if the product is i mproved according to what
consumers recommended it is possible that the product will get a high market share.
The number of consu mers who said they were going to buy (definitely and probably
buy) temporary bonding glue stick were a little lower (86 %) than those of permanent
bonding.
Consumers were asked to estimate how long it took them to use up one 8 g glue stick.
The results are shown in Table 4.17.
Table 4.17
Usage time for one stick of glue
Usage time
less than one month
Permanent bonding
(49)
Temporary bonding
(50)
3 (6%)
1 month
21 (42%)
14 (28%)
2 month
10 (20%)
1 7 (34%)
3-4 month
11 (22%)
1 2 (24%)
6 month
4 (8%)
6 ( 1 2%)
more than 6 month
1 (2%)
1 (2%)
Note;
The number of respondents for permanent bonding and temporary bonding were 49 and 50
respectively because some consumers did not answer the questions
Most consumers (48 %) stated that one permanent bonding glue stick would last them
for about one month or less. Consumers seemed to think that they would used
temporary bonding glue stick less often than permanent glue stick only 28 % would use
83
one stick up within one month or less.
The price consumers were prepared to pay for the new products are shown in Table
4.18.
Table 4.18
Price of glue stick consumer prepared to pay
Price
Permanent bonding (51 )
Temporary bonding (49)
less than 10 Bahts
27 (53%)
17 (35%)
1 0-12 Bahts
16 (31%)
14 (28%)
1 2- 1 5 Bahts
8 ( 1 6%)
15 (31%)
1 6-18 Bahts
2 (4%)
more than 18 Bahts

Note:
(2%)
The number of respondents for temporary bonding was 49 bcause some consumers did not answer
the question
Fifty three percent would buy the permanent bonding glue stick at a price less than 1 0
Bahts, no one would buy the product a t a price higher than the average price of
commercial products (15 Bahts). The study indicated that consumers were prepared to
pay more for temporary bonding glue stick 37 % would buy the product at the same
price or higher than the commercial permanent bonding glue stick.
The results from the consumer study indicated that most consumers in Thailand (61 %)
preferred small size glue stick
1 .9 cm diameter X 8 cm high containing 8 g of glue
rather than a large stick.
4.10
DISCUSSION
4 . 1 0 . 1 Pattern of Glue Usage
Cello tape seemed to be the glue product most consumers used (98 %). There were 95
% of the consumers who used glue sticks. Most consumers in Thailand in this study
used glue sticks (98 %), hence there is no trouble of introducing an improved glue stick
to the Thai consu mers since the consumer needs for this kind of product already exists
84
product already exists and the consumers are familiar with the product.
Self-adhering note pads were widely used by the New Zealand panel, both Thais ( 1 00%)
and New Zealanders (86%), but only 26% of the consumers in Thailand used this
product.
The frequency of usage of cellotape by consumers in Thailand was similar to that of

glue stick. PVA glue and l iquid glue were used less often.
4.10.2 Importance of Attributes
It was found that ease of use, stickability, cleanliness and uniformity of coating came
up as very important attributes for permanent bonding glue stick in both methods.
Effect on paper was judged as very important by direct rating but not by elicitation
measure. Price and size were considered important by elicitation measure. As price of
glue stick obtained very low acceptability by the consumers in Thailand, it should be
included in the important attribute list as well.
Damage of receiving surface, ease of use, stickability and price were shown as
i mportant attributes for self-adhering note pad by both methods. However, size of
paper came out as important in elicitation measure. Keeping quality, versatility, force
needed to pull paper and uniformity of glue coating on paper were judged as important
by direct rating but were not noted as important attributes by elicitation measure.
It can be seen that for most of the attributes which were more important than the others
they came out as important in both methods. This indicated that both methods can be
used to measure importance of attributes. However, elicitation measure is useful in the
case that there are other important attributes of the product being studied which are not
included in the list suggested by the researcher. With some attributes such as price,
elici tation measure is considered more suitable for measuring of importance.
The results from the study showed that New Zealand consumers were able to generate
more important attributes than Thai consumers. This might have resulted from the
difference in tendency to express their opinion. Nevertheless, it was found that there
85
were no significant d ifferences between Thai and New Zealand consumers in
importance rating using elicita tion measure both for glue stick (p=0.18) and self
adhering note pad (p=0.76). With direct rating, there were highly significant differences
between the two groups for glue stick (p=O.OO) and self-adhering note pad (p=O.OO).
Price was judged as important for both products by New Zealand consumers but not
by Thai consumers. Keeping quality for glue stick and uniformity of glue coating on
self-adhering note pad were judged as important by Thai consumers not by New
Zealand consumers.
There were no d ifference in rating of importance of attributes by Thai consumers from

both group: in New Zealand and in Thailand.
4.10.3 Differences between Cultures in Glue Usage
When comparing the results obtained from the Thais in New Zealand and the New
Zealanders, only slightly d ifferences were found between cultures in glue usage in
terms of type of glues, frequency, and purposes of usage. However, there were
d ifferences between the usage patterns of the consumers in Thailand and New Zealand.
A slightly higher percentage of the consumers in Thailand used glue sticks and used
them a lot more often than the consumers in New Zealand. It was found that the
consumers in Thailand used self-adhering note pads less than the consumers in New
Zealand.
4. 1 0.4 Comparison between Glue Products
Although there were some problems in terms of ease of use according to the consumers,
cello tape was the most popular amongst the products used for sticking paper together.
This might have occurred because cello tape has a wider application than the other
glues. It can be used very effectively with non-paper materials. To develop a product
which can overcome these defects it might be possible to gain some market share from
this product. Other glue products (liquid glue, PYA glue and paste glue) had some
disadvantages compared to glue stick in terms of drying time, effect on paper, ease of
use, uniformity of coating and cleanliness. However, consumers found that glue stick
was expensive compared with those products. In order to increase market share of the
86
glue stick product, the price factor should be considered.
4. 1 0.5 Consumers Intention of Buying between the Two Products - Permanent and
Temporary Bonding Glue Sticks
Consumers in Thailand indicated that they wanted to buy the developed permanent
bonding glue stick at a lower price than the commercial products. Most of them wanted
to buy at the price lower than 10 Baht. This implied that consumers considered that
glue sticks in the market were too expensive. Although price did not seem to be a very
important attribute according to the panel in Thailand, the existing glue stick's price
obtained lowest acceptability by the consumers. This indicated that if the glue stick
could be developed and sold at a lower price, there is high opportunity to compete with
competitors.
For temporary bonding glue stick, 86 percent of consumers said that they might buy the
product. Moreover, some of them (36 percent) would buy the product at a price equal
to or higher than the commercial glue sticks. This might have happened because the
self-adhering note pad is an expensive product and if consumers could used this glue
instead of the note pad, they would be prepared to pay more for it even though this
product had a narrower range of usage.
4. 10.6 Development of Product Concept
The results of the consumer study revealed that both permanent bonding glue stick and
temporary bonding glue stick had si milar opportunity to be further developed. More
consumers were willing to buy permanent bonding glue stick and would use it more
often.
Since the consumer study showed that there was the need for a permanent bonding
glue stick especially a glue stick which would be sold at a cheaper price than glue stick
already on the market, it was decided that in further study, emphasize would be given
to permanent bonding glue stick. Most consumers were willing to buy the developed
product if it could overcome some defects which the present commercial glue sticks
possess. Therefore, the concept of the product to be developed was:
87
'A glue stick which can be applied smoothly and easily onto paper and give a uniform
coating. The paper can be repositioned within a few minutes of applying, then it would
form a firm bond between the substrates to be bonded together. The finished work
would be free from wrinkle or curling. The product should be water soluble, easy to
clean off hands or other materials. The 8 g glue stick would be sold for 10 to 12 Bahts.'
4 . 1 0.7 Using the Consumer Study for Development of Glue Stick Products
The consumer study was used successfully in obtaining information regarding glue
usage from consumers. Although consumers were not especially trained to describe the
i mportant attributes of glue products, the evidence from the results showed that
consumers could be used for generating important attributes of glue products.
J. Walter Thompson Company Ltd . used a 'sensitivity' panel which comprised 8

housewives, aged 25-45 with at least one child at home in development of adhesive
products. These panelists were given some training in perception and responsiveness
in a research situation. The respondents were asked to make notes on the perceived
a dvantages and d isadvantages of glue products, which members of the family used
a dhesives, for what purposes, and so on. It was found that packaging and systems of
application tuned out to be as important as the products themselves, and many
suggestions were made for improvement (Fuller, 1984).
However, in the area of glue products or any consumer products, the information
regarding product usage should be elicited from the users themselves rather than the
non-users. This could lead the research to the wrong d irection and end up with a
product failure.
88
4.1 1
CONCLUSIONS
Permanent bonding glue stick was finally chosen as the suitable glue product to be
developed for the Thai market. With the use of tapioca starch in the formulation, the
price of the final product could be reduced. This would persuade .more consumers to
buy the product and the government offices as well as the private companies who did
not buy glue stick to use in their offices owing to the high price could take the new
price into account.
From the consumer study, it can be summarised tha t the important attributes of
permanent bonding glue sticks are as follows:
,.. Effect on paper

,.. Cleanliness of work
,.. Ease of use
,.. Stickability
,.. Uniformity of coating
,.. Price
Since there were no differences between cultures in the pattern of glue usage, if the
product was developed and was found successful in Thai market, there was
opportunity that the product should be able to be exported to other countries such as
New Zealand.
At this stage of the research project, a final decision was made to choose the glue stick
product to be developed. The following chapter discusses the selection of the basic
formulation for tapioca based glue stick and the development of testing procedures for
glue stick attributes.
I.
89
CHAPTER 5
INITIAL DEVELOPMENT STUDY ON GLUE STICK FORMULATION
This preliminary study on starch based glue stick formulation was conducted in order
to explore the appropriate ratio of ingredients in the formulation to be used in the
development of prototypes. Glue stick comprises three major parts: adhesive substance,
solvent and gel-forming agent, and the suitable ratios of these components were
investigated. Since starch was to be used as adhesive in the formulation, the ingredients
which were compatible with starch were focused.
5.1
A I M AND OBJECTIVES
The aim of this study was to explore the use of tapioca starch in glue stick and to select
a suitable basic formulation for a starch based glue stick. The objectives were to:
* Explore the suitable levels of basic ingredients for starch based glue stick.
* Investigate the relationships between ingredients and attributes of glue stick.
* Study the relationships between physical attributes and sensory attributes.
* Select a set of attributes for product testing by consumers.
* Select physical attribute measures.
5.2
SELECTION OF PRELIMINARY FORMULATJON
Although glue stick was developed many years ago, the knowledge on formulation and
method of glue stick processing are not widely known. Therefore, the main purpose of
this formulation selection was to obtain a simple glue stick formulation which could be
made with simple laboratory equipment and did not require a complicated processing
method. This formulation would be used as a basic formulation for further
development.
90
5.2 . 1
Criteria for Selection of Preliminary Formulation
The decision on glue stick formulation was made based on the following:
,.. Availability of ingredients
- Some companies in New Zealand did not sell ingredients in small
amounts and did not have a small sample of ingredients for
experimentation. Therefore this limited the choice of ingredients which
could be used in the formulation
- Only the ingredients which were available in New Zealand where the
experiments took place were used
... Properties of ingredients
- Non-toxic
- Price
- Compatibility with starch
... Processing method
- Availability of processing equipment in the laboratory
- Time and temperature used in the process
... Basic properties of finished glue stick
- Colour
- Stickability
- Structure / maintain shape when rubbing on paper
- Homogeneity of the stick
From the conditions mentioned above, the following initial formulation was chosen
(Muszik and Dierichs, 1971):
%
Polyvinyl pyrrolidone
Sodium stearate
32
War
Glycerine
14
This formulation was chosen because it had simple ingredients and contained polyvinyl
pyrrolidone as adhesive substance which could be used with carbohydrate or modified
carbohydrate (Werke H.u.M. Fischer, G.m.b.H., 1974). The finished product was water
91
soluble so i t could be easily cleaned off hands or other materials.
5.2.2
Preliminary Experimentation
In the preliminary experiment, tapioca starch (raw starch and modified starch) was used
to replace some part of the polyvinyl pyrrolidone in the preliminary formulation. Glue
stick samples from the formulations shown in Table 5 . 1 were made in the laboratory
using the method described in Section 3.5.2.
Table 5 . 1
Formulations used in preliminary experimentation
Ingredients
FI
F2
F3
1 6.0
22.0
20.0
F4
F5
F6
1 6.0
10.0
1 0.0
A D H ESIVE
Raw starch
Mod i fied starch
4.0
6.0
5.0
4.0
20.0
20.0
Water
56.0
48.5
45.5
50.5
46.5
46.5
Glycerin
14.0
10.0
10.0
14.0
8.0
8.0
10.0
10.0
Stearic acid
1 2.0
9.0
9.0
9.0
Sodium hydroxide
4.0
3.0
3.0
3.0
3.5
3.5
3.5
SOLVENT
GEL-FORMING AGENT
Sodium stearate
EMULSIFYING AGENT
Brij 35
Glyceryl monostearate
3.5
3.5
The purpose of the study at this stage was to obtain the suitable formulation for starch
based glue stick and the samples generated from the formulations in Table 5 . 1 were
very much di fferent in their properties, so they could be easily j udged by the author.
Hardness of the sample was judged by pressing fingers on sample. The sample was
rated as soft, medium hard and hard. Stickability was assessed by the ability to stick
paper together after drying and was rated as poor, fair and good. Homogeneity and
colour were judged by eye according to appearance of the finished product. The results
are shown in Table 5.2.
92
Table 5.2
Properties of glue stick samples from preliminary experimentation
Formulation
Hardness
Stickability
Homogeneity
Colour
FI
Soft
Poor
Poor
Pale brown
F2
Soft
Poor
Fair
Pale brown
F3
Soft
Good
Fair
Pale brown
F4
Medium hard
Poor
Fair
Off white
FS
Hard
Fair
Fair
White
F6
Hard
Fair
Good
White
It was found that sodium stearate did not give a product which could be made into a
stick in the starch based system (FI and F2) therefore stearic acid and sodium hydroxide
were used to form the gel-forming agent. When modi fied tapioca starch was used (F4,
FS and F6) instead of raw starch, the colour of the stick was improved and it also gave
a harder stick. In the formulations which used higher amounts of polyvinyl pyrrolidone,
the stickability was improved. The emulsifying agent increased homogeneity of the
stick, glyceryl monostearate (F6) gave better homogeneity than Brij 35.
From the results of the preliminary experimentation, formu lation F6 was chosen for
further study. Since this formulation gave only fair stickability, in order to improve this
property, it was decided to try adding other natural adhesives in the formulation to
increase stickability. Dextrin and casein were considered.
5.2.3
Dextrin Experimentation
In the system using stearic acid and sodium hydroxide to form the gel-forming agent,
the alkalinity of the system was very high. This caused dextrin to change to very dark
brown colour when heated. It was decided that the amount of sodium hydroxide used
in the formulation should be decreased and amount of stearic acid should be increased.
93
Based on the formulation F6, a constrained mixture design with 3 variables: modified
tapioca starch (14-16%), polyvinyl pyrrolidone (14-16%) and dextrin (4 8%) was
-
developed and samples were made in the laboratory using the method described in
Section 3.5.2. The other ingredients were fixed as follows:
%
10.0
Stearic acid
Sodium hydroxide
Water
1 .5
43.0
Glycerin
8.0
3.5
The experimentation plan is shown in Table 5.3.

Table 5.3
Experimentation plan for dextrin experiment
Run
Starch
Dextrin
Others
01
1 6 .0
14.0
4.0
66.0
02
10.0
16 .0
8.0
66 .0
03
16.0
1 0.0
8.0
66.0
04
14.0
1 6.0
4.0
66.0
05 (centre point)
1 4 .0
14 .0
6.0
66.0
06 (centre point)
14.0
14.0
6.0
66 0
.
The hardness and peel strength of the glue stick samples from this experimentation
were measured lIsing the method described in Section 3.3. The homogeneity and colour
of glue stick was judged subjectively by eye. The results are shown in Table 5.4
Table 5.4
Physical attributes of dextrin glue sticks
Samples
Hardness
Peel strength
Homogeneity
Colour
01
1 .62
4 .93
Fair
Pale brown
02
1 . 98
3.26
Fair
Brown
D3
2.89
3.28
Fair
Brown
04
2 .48
4.40
Fair
Pale brown
05
2.41
4.20
Fair
Brown
06
2.43
4.53
Fair
Brown
94
The correlations between ingredients and physical attributes are shown in Appendix 5. 1 .
The empirical equations relating physical attributes to the ingredients are shown in
Table 5.5.
Table 5.5

attributes of glue sticks from dextrin experiment
A ttributes
Equations
Hardness
Peel strength
+
Note:
0.1 62*Starch
t-ratio
1 1. 21 '''
0.95
0.197*Starch
4.53'
0.99
0. 167*PVP
3.85
- 0.167*Dextrin
-2.25
'
'
- Significant at 0.05 2 P 2 0.01
" - Significant at 0.01 2 P 2 0.001
... - Significant at p < 0.001
The empirical equations showing the relationships between ingredients and glue stick
attributes showed that starch increased hardness of glue stick. Starch and polyvinyl
pyrrolidone increased peel strength. From the equation, dextrin decreased peel strength
of glue stick and also showed negative correlation with peel strength (p=O.Ol), see
Appendix 5.1.
Since i t was found that dextrin did not increase stickability of glue stick and the
finished product was brown in colour, it was decided that dextrin was not to be
included in the glue stick formulation.
5.2.4
Casein Experimentation
Casein was considered to be used as adhesive substance in the tapioca based glue stick.
The experiment was conducted to investigate the effect of casein on glue stick
properties. Effects of using sorbitol instead of glycerin in the formulation were also
studied. The mixture experimentation plan of the casein experiment is shown in Table
5.6.
------ -- -
95
Table 5.6
Experimentation plan for casein experiment
Runs
PVP
Casein
C1
14.0
0.0
1 3.0
0.0
73
C2
8.0
6.0
13.0
0.0
73
C3
14.0
0.0
0.0
13.0
73
C4
8.0
6.0
0.0
13.0
73
C5 (centre point)
1 1 .0
3.0
6.5
6.5
73
C6 (centre point)
1 1 .0
3.0
6.5
6.5
73
Glycerin
Sorbitol
others
Other ingredients in the formulation were fixed as follows:

%
Modified tapioca starch
18.0
Stearic acid
12.0
Sodium hydroxide
Water
3.0
41.0
Glyceryl monostearate, 3.5 g, was added to 1 00 g of the complete formulation. The

casein solution used in the experiment was obtained from adding 20 g of casein into the
mixture of 80 grams of water and 0.6 grams of sodium hydroxide. The solution which
gave the amount of casein required was weighed then more water and sodium
hydroxide was added later to make up the percentage required in the formulation. The
samples were made in the laboratory using the method as described in Section 3.5.2
The samples were tested by physical testing, see Section 3.3 and sensory testing by
trained sensory panel as described in Section 3.2.3. The results are shown in Appendix
5.2. The correlations between ingredients and glue stick attributes are shown in
Appendix 5.3. The empirical equations relating ingredients to physical attributes and
sensory attributes are presented in Table 5.7 and 5.8 respectively.
------- -
-----
96
Table 5.7
physical attributes from casein experiment

Physical attributes
Ingredien ts
t-ratio
Dry glue/Area
1 .333*PVP
1 .754*Casein
12.72'"
5.75"
'"
40.27
"
15.2 7
-2.36
'"
15.72
"
4.60
'"
1 0.93
'
3.94
%Moisture
3.362*PVP
3.034*Casein
-0.260*Glycerin
Peel strength
0.254*PVP
0.216*Casein
Physical hardness
0.412*PVP
0.432*Casein
Wet glue/ Area
2.418*PVP
3.256*Casein
Open time
0.384*PVP
10.15'"
4.70"
'"
8.73
0.99
0.99
0.99
0.98
0.98
0.93
Note: '
- Significant at 0.05 P 0.01

Significant at 0.01 P 0.001
...
- Significant at p < 0.001
From the empirical equations, it can be concluded that polyvinyl pyrrolidone and casein
increased physical hardness, wet glue and dry glue per area, moisture content and peel
strength of the samples. Polyvinyl pyrrolidone also had significant effect on open time.
It was found that neither glycerin nor sorbitol showed significant effect on physical
attributes of the samples.
From the empirical equations in Table 5.8, it could be seen that casein increased
thickness, smoothness, cleanliness, perceived hardness, and bond strength before drying.
Use of casein increased deformation and disintegration of glue stick samples and it was
found that casein did not dramatically increase stickability therefore casein was not
included in the formulation in further study.
97
Table 5.8

sensory attributes of glue stick from casein experiment
Sensory Attributes
Thickness
Ingredients
+
0.621*PVP
0.771 *Casein
Smoothness
0.775*PVP
+ 0.708*Casein
Cleanliness
0.815*PVP
0.7S7*Casein
+ 0.027*Sorbitol
Slipperiness
Hardness
Deformation
0.283*PVP
0.107*Sorbitol
0.549*PVP
+ 0.46S*Casein
+ 0. 154 *Sorbitol
+
0.676*PVP
0.784*Casein
Visibility
0.660*PVP
+ 0.73S*Casein
Adjustability
0.63S*GI ycerin
+ 0.470*Sorbitol
Disintegration
0.635*PVP
+ 0.868*Casein
Coverage
0.969*PVP
Bond strength 1
0.405*PVP
+ O.564*Casein
Bond strength 2
0.43S*PVP
+ 0.368*Casein
Delamination
Note:
0.221*PVP
+ 0 . 1 12*Glycerin
t-ratio
'"
1 3.30
5.68"
72.50'"
22.77'"
83.89'"
32.72'"
2.10
'"
14.63
'
4.12
1 3.29'"
'
4.74
2.83
12.8'"
"
5.10
'"
1 1 .80
'
4.52
"
12.98'
'"
9.60
9.91'"
4.66"
1 1 .44'"
7.17"
'
3.43
8.41'"
0.99
0.99
0.99
0.99
0.99
0.99
0.98
0.98
0.98
0.96
0.96
0.96
2.45
6.05
2.29
"
0.96

.., - Significant at p < 0.001
Sorbitol was positively correlated (p<O.OS) with perceived hardness and glycerin was
negatively correlated (p<O.OS) with perceived hardness. Hence, it can be said that
sorbitol increased hardness while glycerin decreased hardness of glue stick. It was
found that .use of sorbitol in the formulation resulted in decreasing of delamination and
adjustability so it was decided to continue using glycerin in the formulation not the
sorbitol.
98
5.3
ATTRIBUTES OF COMMERCIAL GLUE STICKS
Since there was no consumers' ideal product profile or the standard on glue stick
attributes which could be used as guidance in the development of glue stick prototypes
at this stage, it was necessary that the attributes of commercial glue sticks were
investigated. Six of the commercial glue sticks available in the market (see Appendix
5 .4) were bought and tested using the physical methods described in Section 3.3 and
sensory attributes were evaluated by the trained panel (see Section 3.2.3). The physical
a ttributes and sensory attributes of the commercial glue sticks are shown in Tables 5.9
and 5.10 respectively.
Table 5.9
Samples
Physical attributes of the commercial glue sticks
Wet glue/area
(g/m' )
Dry glue/area
(g/m')
PRlTI
32.6
1 1 .6
UHU
36.2
AMOS
Mp
Peel strength
(Newton)
Hardness
(Newton)
Open time
(minute)
57.0
3.52
1 2.25
64.3
3.65
13.2
59.8
3.55
21 .75
52.5
4.99
45.8
14.9
67.2
2.86
24.00
68.0
4.46
BOSTIK
31 .9
10.3
67.3
2.93
12.75
64.0
4.71
PELIFIX
36.6
1 4.8
47.9
2.67
99.00
68.0
6.30
ESSELTE
35.0
1 1 .6
65.5
2.42
1 6.75
56.3
4.86
Moisture
("!o)
("c)
Some of the physical attributes of the commercial glue sticks varied in a wide range
such as open time, moisture content and peel strength. Most of the physical attributes
of UHU, the glue stick which most Thai consumers used, were in the middle of the
ranges except that the hardness was the highest and melting point was the lowest.
99
Table 5.10 Sensory attributes of the commercial glue sticks (mean scores varied
from 0-15)
Samples
Slip Hardness Deform Disintegr. Cover
PRITT
9.6
9.1
6.6
7.0
1 0.2
6.6
8.3
UHU
5.9
9.2
1 0.2
8.1
1 1 .0
6.2
AMOS
9.5
7.0
8.9
6.4
1 1 .4
PELIFIX
6.8
7.3
8.8
5.2
ESSELTE
6.5
8.1
5.8
3.8
Thick
Visible Smooth
Clean
Adjust
Bond
Bond Delaminate
strengthl strength2
1 2 .0
1 2.6
8.5
6.7
10.7
1 0.5
9.4
1 1 .8
1 2.0
9.6
7.6
9.9
9.1
8.0
9.7
1 0.5
12.4
1 0.3
8.2
1 0.2
9.6
1 0.5
8.1
12.9
1 2. 1
12.2
10.4
6.7
9.4
1 0.3
1 0.8
7.8
6.1
1 2.2
1 2.4
9.3
6.3
9.7
9.5
Although, the sensory attributes of commercial glue sticks varied in a wide range. These
attributes could be used as guidance at this stage. Then the ideal product profile which
was developed later by the target consumers would be used to guide the direction in
which the prototypes should be improved in the optimization stage.
5.4
DEVELOPMENT OF GLUE STICK PROTOTYPES
In order to make the decision on the suitable levels of the ingredients in the starch
based glue stick, the attributes of the prototypes developed at this stage would be
compared with the attributes of commercial glue sticks. The formulation providing
sample with the attributes in the ranges of commercial glue sticks would be chosen as
a basic formulation. This basic formulation was to be used later to developed prototypes
for consumer testing.
At this stage it was decided that the formulation F6 from Section 5.2.2 should be used
as the basic formulation to develop product prototypes. This formulation contained
modified tapioca starch and polyvinyl pyrrolidone as adhesive substances, water and
glycerin as solvents, stearic acid and sodium hydroxide as gel forming agent. The
suitable lower and upper level of each ingredients were specified in order that the effect
of each ingredient on product attributes could be assessed. The ranges of each
ingredient used in the experimentation are shown below:
1 00
%
Low level
High level
10.0
1 6.0
10.0
16.0
Water
35.0
45.0
Glycerin
8.0
14.0
Stearic acid
9.0
12.0
Sodium hydroxide
3.0
4.0
Amount of glyceryl monostearate was fixed at 3.5 grams per 1 00 grams of glue mixture.
The experimentation plan as described in Section 3.b is shown in Table 5.1 1 .
Table 5.1 1
Experimental plan for prototype development
Total
Starch
PVP
1 6.0
Run
W ater
Glycerin
1 6.0
42.0
14.0
9.0
3.0
32.0
56.0
1 2.0
1 6.0
16.0
45.0
8.0
1 2.0
3.0
32.0
53.0
1 5.0
10.0
16.0
45.0
14.0
1 2.0
3.0
26.0
59.0
1 5.0
1 6.0
10.0
45.0
14.0
1 2.0
3.0
26.0
59.0
1 5.0
1 6.0
1 6 .0
39.0
14.0
1 2.0
3.0
32.0
53.0
1 5.0
1 6.0
1 6 .0
4 l .0
1 4.0
9.0
4.0
32.0
55.0
13.0
16.0
16.0
44.0
8.0
12.0
4.0
32.0
52.0
1 6.0
1 0.0
1 6.0
44.0
1 4 .0
1 2.0
4.0
26.0
58.0
1 6.0
1 6.0
10.0
44.0
1 4.0
1 2.0
4.0
26.0
58.0
1 6 .0
10
16.0
1 6.0
38 . 0
1 4 .0
1 2.0
4.0
32.0
52.0
16.0
11
1 4 .8
1 4 .8
42.7
1 2.8
1 l .4
3.5
29.6
55.5
1 4 .9
12
1 4 .8
1 4.8
42.7
1 2.8
1 1 .4
3.5
29.6
55.5
1 4 .9
Note:
Stearic
NaOH
Adhesive
Solvent
Gel forming agent
Runs 11 and 12 are centre points
Glue sticks from each run were made in the laboratory lIsing the method as described
in Section 3..2. The sticks were mounted in the containers and kept at room
temperature before testing. The samples were aged at least 5 days before being
evaluated.
101
5.5
PROPERTIES OF GLUE STICK PROTOTYPES
The glue stick samples from each experimental run were subjected to physical testing
as described in Section 3.3, the results are shown in Table 5.12.
Table 5.12
Physical attributes of the glue stick samples from constrained mixture

design experiment
Samples Wet glue/area Dry glue/area Moisture

(g / m 2 )
(g / m 2)
(%)
Hardness
(Newton)
Open time
(minute)
Mp
(
Oe
)
Peel strength
(Newton)
Run 1
48.3
2l.3
54. 1
4.53
1 0.25
80.5
3.70
Run 2
23.9
1 2.5
50.5
6.67
1 .50
7 1 .0
4.24
Run 3
20.2
9.0
53.5
5.95
0.00
7 1 .5
4.21
Run 4
29.5
1 2.5
53.3
5.63
0.00
73.0
4.55
Run 5
30.9
1 4.3
50.3
6.39
2.50
75.3
3.96
Run 6
64.2
28.4
5 1 .3
1 .85
36.00
79.3
4 .64
Run 7
42.6
19.3
54.5
5.1 1
1 .25
78.3
4.58
Run 8
34. 1
13.7
57.3
4.01
0.00
79.3
3.49
Run 9
45.5
18.1
57.4
4.29
3.25
73.0
4.38
Run 1 0
43.3
1 9.8
51 .6
4.89
4.50
80.5
5.66
Run 1 1
46.7
21.8
47.7
4.43
5.5
78.3
3.48
Run 1 2
37.8
1 8.5
47.9
4.92
5.0
78.0
3. 1 4
A trained sensory panel of 1 0 panelists was used for sensory testing. The panelists
evaluated the samples as they did in the panel training (see Section 3.2.3). Samples were
randomly presented to the panelists. Each panelist attended 2 sessions with 6 samples
in each session. The mean of sensory scores obtained from the trained sensory panel are
tabulated in Table 5. 1 3.
102
Table 5.13
Sensory attributes of the glue stick samples from mixture design

experiment (mean scores on a scale from 0-15)
Samples
Slip Hardness Deform Disintegr. Cover
Thick
Visible
Smooth
Clean
Adjust
Bond
strength1
Bond
Delaminate
strength2
5.1
9.1
1 1 .6
1 1 .9
( 1 .7)
( 1.6)
( 1 .8)
( 1 .3)
1 0.9
(2. 1 )
9.0
(1 .0)
5.7
(2.2)
1 1 .0
( 1 .3)
1 1 .7
(0.9)
8.6
( 1 .9)
6.3
(1 .7)
7.4
( 1 .7)
4.9
(2.5)
Run 2
3.7
(1 .5)
10.6
( 1 .9)
8.2
(2.1 )
7.3
(2.4)
7.4
(2.3)
7.8
( 1 . 9)
4.8
(1 .7)
1 1 .8
(1 .5)
1 2.2
(0.7)
9.7
( 1 .7)
6.0
(2.2)
4.9
( 1 .6)
2.1
( 1 .4)
Run 3
6.1
( 1.0)
9.6
(1 .7)
9.1
(2.3)
9.2
(2.3)
8.1
(2.4)
5.8
( 1 .4)
7.0
(1 .9)
1 2.0
(0.5)
1 1 .9
(0.8)
1 0.8
(2. 1 )
4.4
( 1 .8)
3.5
( 1 .5)
1 .2
(0.9)
9.6
( 1 .4)
6.5
(2.3)
5.9
(2 .3)
1 1 .7
(1 .5)
1 2.2
(0.4)
10.0
(2. 1 )
3.3
( 1 .4)
3.2
(1.1)
0.4
(0.5)
8.0
( 1 .9)
7.7
(1 .3)
5.8
(2.4)
1 1 .8
(0 .4)
1 1 .8
(0.8)
8.4
(2.2)
5.4
( 1 .9)
4.7
( 1 .3)
2.2
( 1 .8)
1 2.6
( 1 .4)
1 1 .8
( 1 .8)
1 0.5
(2. 1 )
9.8
(2.0)
1 1 .0
(1.1)
8.8
(2.0)
8.1
(2.7)
1 0.5
(1 .5)
8.7
(1 .7)
8.3
(2.3)
7.5
(2. 1 )
8.8
(0.9)
1 1 .3
(1 .4)
1 2. 1
(0.6)
6.9
(2.0)
6.7
(1 .9)
5.3
(2.0)
3.7
(2.4)
9.7
(2.5)
6.0
(1 .7)
7.4
(2.4)
1 1 .3
(1.1)
12.1
(0.3)
9.8
(1 .7)
5.3
( 1 .9)
3.3
( 1 .3)
1 .0
(1 .2)
Run 1
Run 4
Run 5
Run 6
Run 7
Run 8
Run 9
Hun 10
Run 1 1
Run 1 2
Note:
5.6
6.3
9.2
9.8
9.9
(2.2)
(2.0)
(2.4)
(2.6)
4.8
8.9
( 1 .8)
(2.0)
4.9
5.6
( 1 .7)
( 1 .8)
4.0
9.2
( 1 .4)
(2.9)
10.0
(2.3)
9.3
(2.8)
11.1
12.8
(2.5)
( 1 .8 )
10.2
(2.2)
9.5
( 1 .2)
7.4
8.3
8.6
9.3
(2.0)
(1 .9)
(2.1 )
( 1 .7)
7.3
9.0
9.9
(1 .3)
(1 .6)
(2.3)
(1 .2)
9.6
( 1 .5)
8.2
(1 .7)
8.2
(2.2)
1 1 .7
(1 .0)
1 2.2
(0.3)
8.2
(2. 1 )
5.3
(1 .7)
4.4
(2. 1 )
1.1
(1 .0)
4.7
( 1 .5)
8.4
(2.4)
8.8
(1 .6)
9.5
(1 .5)
9.0
(2.2)
9.1
(2.0)
8. 1
(2.4)
1 1 .4
(0.8)
'12.0
(0.3)
8.7
( 1 .9)
6.2
(2. 1 )
6.0
(2.0)
4. 9
(2.7)
9.1
8.9
(2.3)
8.2
(1 .2)
1 0. 2
(1 .4)
1 1 .4
(1 . 1 )
1 1 .8
(0.7)
8.3
(2.2)
6.9
(1 .6)
7.7
(2.0)
7.0
(2.7)
8.9
(2.3)
8.3
(1 .9)
1 0.0
(1.1)
1 1 .3
(1 .2)
1 1 .8
(0.6)
8.3
( 1 .7)
6.6
(1 .5)
6.9
(2. 1 )
6.0
(2.4)
5.3
8.0
( 1 .5)
(2.4)
10.6
( 1 .6)
10.6
(1 .5)
5.1
8.7
7.6
8.6
(1 .9)
(1 .5)
(2.3)
(1 .6)
The numbers in parentheses are standard deviations
COMPARISON B ETWEEN PROTOTYPE PRODUCTS AND COMMERCIAL

P RODUCTS
Comparison of physical measurement and sensory attribute results of prototype

products and commercial products are shown in Table 5.14 and 5.15 respectively. Since
the sample from Run 6 gave a very soft stick and a high amount of glue applied onto
1 03
paper, it was decided tha t the results from experimental Run 6 not be included in the
comparison.
Table 5.14 Comparison between physical attributes of prototype products and

commercial products
Physical Attributes
Commercial
Prototype
Wet glue per area
31 .9-45.8
20.2-48.3
Dry glue per area
1 0.3-14.9
9.0-2 1 .8
%Moisture content
57.0-67.3
47.7-57.4
Hardness
2.42-3.55
4.01 -6.67
Open time
12.3-99.0
0.0-10.3
Melting point
52.5-68.0
71 .0-80.5
Peel strength
3.65-6.30
3.14-5.66
Some physical attributes of the prototypes: wet glue per area, dry glue per area and
hardness were in the ranges of the commercial products' attributes. Melting point of
the prototypes (71-81 DC) were much higher than those of the commercial products (5368 DC) . The high melting points indicated that the prototypes would not melt in hand
during usage or during storage at room temperature but also gave a harder glue sticks.
Open time of the prototypes were between 0 to 10 minutes while those of the
commercial ones were between 1 2 to 99 minutes. The prototypes had slightly lower peel
strength.
1 04
Table 5.15 Comparison of sensory attributes of prototype products and commercial
products
Sensory Attributes
Commercial
Prototype
Coverage
1 0.2-1 1 .4
7.4-10.9
Thickness
6.2-8.1
5.8-9.1
Visibility
6. 1-12.9
4.8-10.2
Smoothness
1 0.5-12.2
1 1 .0-12.0
Cleanliness
1 2.0-12.6
1 1 .7-12.2
8.5-10.4
6.9-10.8
Bond strength 1
6.3-8.2
3.3-6.9
Slipperiness
5.9-9.6
3.7-7.4
Hardness
7.0-9.2
5.6-10.6
5.8-10.2
7.6-1 1 .6
3.8-8. 1
7.3-1 1 .9
Bond strength 2
9.4-10.7
3.2-7.7
Delamination
9 . 1 -1 0.5
0.4-7.0
Adjustability
Deformation
Disintegration
Note: The mean scores were on line scales from 0-15

Most sensory a ttributes of the prototypes were in the ranges that covered the ranges of
the commercial products except bond strength before drying, disintegration, bond
strength after drying and delamination. Bond strength before drying, bond strength
after drying and delamination were lower than those of the commercia l ones.
Disintegration was higher than those of the commercial products. The profiles of
samples from centre point (average mean scores from Runs 1 1 and 12) and Run 1 0
which possessed the attributes closest to those o f the commercial ones are shown i n
Figure 5 . 1 .
From the comparison between the prototypes and the commercial products, the glue
stick had to be improved in terms of diSintegration, bond strength before and after
drying and delamination. Although melting point was high, this was not the main
problem as long as there was no problem in applying glue onto paper. The
recommended open time of a glue stick was at least one minute since prolonged open
time was unnecessary (Pletcher and Wong, 1978). Therefore with the right levels of
ingredients, it was possible to obtain the required open time from this basic
formulation.
105
Attributes
Coverage
Thlckneaa
Visibility
Smoothneaa
Cleanllneaa
AdJustabll1ty
Bond strength 1
Slipperlneaa
Hardneaa
Defonnatlon
Disintegration
Bond strength 2
Delamination
10
12
14
Seneory Sco....
1 --
Run10
Cenlr8pOlnt -- Comm- level -- Comm-upper level
Figure 5.1
Product profiles of glue stick prototypes compared with

commercial product attributes
106
5.7
RELATIONSHIPS BETWEEN GLUE STICK ATTRIBUTES
5.7.1
Correlations between Glue Stick Physical Attributes
The correlations between physical attributes were determined and are shown in Table
5 . 1 6.
The relationships between the physical attributes can be shown as the diagram in
Figure
Table
5.2.
Correlations between physical attributes of glue stick samples from
5. 16
Wet glue/area Dry glue/ area %Moisture
Dry glue/ area
Hardness
Open time Melting point
0.97T
%Moisture
-0.034
-0.226
Hardness
-0.872'"
-0.802"
-0.153
Open time
0.78T
0.798"
-0.193
-0.793"
Mp
0.694'
0.708'
-0.085
-0.609'
0.405
Peel st
0. 136
0. 1 1 1
0. 199
-0.029
0. 144
Note:
- Significant at 0.05 P om
"
- Significant at o m p 0.001
- Significant at p
Figure
5.2
<
-0.036
0.001
Diagram Showing Relationships between Physical Attributes
1 07
Hardness of glue stick affected the amount of glue applied per area (wet glue and dry
glue per area). The harder the stick, the less amount of glue was applied onto the
surface. The amount of glue applied per area affected the open time. In consequence,
hardness also affected open time.
It was found that the melting point of the sample was negatively correlated with
hardness of the stick (p<O.05). Generally, it was expected that increasing melting point
should be related to increasing hardness. However, it has been mentioned that there
was often no relation between melting point of mixtures and hardness (Bennett, 1963).
Therefore the melting point could not be used to indicate hardness of glue stick.
5.7.2
Correlations between Glue Stick Sensory Attributes
The correlations between sensory attributes are shown in Table 5.17. The d iagram in
Figure 5.3 shows relationships between the sensory attributes.
Table 5.17 Correlations between sensory attributes of glue stick samples from mixture
design experiment
Cover
Thick
0.656'
0.375
Smooth -0.862'"
Clean
-0.695'
Adjust
-0.059
Bondstl
0.380
Slip
0.245
Hard
-0.790"
Deform
0.576
Disintegr. 0.91 1 '"
Bondst2 0.621 '
Delamin. 0.499
Visible
Smooth
Clean
Adjust
Bond
strength1
Slip
Hardness
Deform
Disintegr.
Bond
s trength2
Thick
Visible
Note:
0.465
-0.810"
-0.756"
-0.436
0.794"
-0.406
-0.687'
0.433
0.616'
0.910'"
0.819"
-0.554
-0.488
-0.461
0.632'
0.030
-0.71 2"
0.086
0.267
0.605'
0.705'
0.828'"
0.279
-0.736"
0.167
0.857'"
-0.465
-0.737"'
-0.834'"
-0.765"
0.1 1 5
-0.658'
0.229
0.81 1"
-0.480
-0.661'
-0.847""
-0.785"

" - Significant at 0.01 P 0.001
, - Significant at p < 0.001
'
..
-0.597
0.378
0.205
-0.318
-0. 177
-0.424
-0.455
-0.526
-0.583'
0.259
0.301
0.880'"
0.885'"
-0.097
-0.038
0.196
-0.438
-0.466
-0.41 2
-0.679'
-0.714"
-0.698'
0.775"
0.458
0.333
0.544
0.399
0.%5'"
1 08
Perceived Hardness
Deformation
Bond strength
before drying
Cleanliness
Figure 5.3
Delamination
Diagram Showing Relationships between Sensory Attributes
From the d iagram, i t can be seen that perceived hardness of a glue stick affected
disintegration and deformation of the glue stick. Disintegration affected thickness and
degree of glue coverage on the paper. Glue stick with high disintegration gave high
thickness and high degree of coverage. As a resul t, these also increased bond strength
before drying, adjustability, delamination and decreased smoothness and cleanliness of
finished work.
1 09
5.7 .3
Relationships between Physical Attributes and Sensory Attributes
Normally only physical testing methods are used to assess consumer response in the
quality control of product manufacturing because sensory evaluations involve
substantial time and money expenditures, and often exhibit poor reproducibility
(Szczesniak, 1987). Since there were no standard methods set for glue stick testing,
therefore the relationships between physical attributes and sensory attributes were
investigated. The correlations between physical attributes and sensory attributes are
shown in Table 5.1B. Amount of glue applied per area (wet glue and dry glue per area)
were correlated with most of the sensory attributes of glue stick samples, except
slipperiness. Therefore, it can be said that amount of glue applied per area could be
used to assess sensory attributes of glue stick in terms of glue residue on paper,
stickability and effect on paper. Physical hardness was correlated with most of the
sensory a ttributes except adjustability, slipperiness and deformation. Hence, hardness
can be used to predict some sensory attributes of glue stick. A hard glue stick would
give low degree of glue cover on paper, low stickability, but smooth and clean finished
work. Moisture content was correlated with slipperiness and delamination. A glue
which had a high moisture content was easy to apply with high slipperiness but gave
low delamination.
Although open time was expected to correlate with adjustability, the results showed
that they were not correlated. Hence, the adjustability judged by the sensory panel was
not related to open time. As it was found that adjustability was negatively correlated
with bond strength before drying this meant that adjustability only depended on the
tack of the glue before the drying occurred. Peel strength was not correlated with any
sensory attributes.
110
Table 5.18
Conelations between physical and sensory attributes of glue stick

samples from mixture design experiment
Wet glue/area
0.793"
Coverage
'
Dry glue/area Moisture Hardness Open time
0.71 1 "
0.198
Mp
Peel strength
-0.888"
0.820'"
0.516
0.059
0.487
0.259
Thickness
0.848 "
0.897"
-0.328
-0.655'
0.877'"
Visibility
0.654'
0.677
-0.275
-0.694'
0.479
0.491
-0.069
Smoothness
-0.855'"
-0.856'"
0.085
0.880'"
-0.925'"
-0.650'
-0.068
Cleanliness
-0.662'
-0.70 1 '
0.368
0. 668'
-0.914'"
-0.456
0.063
Adjustability
-0.58 1 '
-0.618'
0 . 'l 32
0.231
-0. 149
-0.496
-0.0 1 2
0.747"
0.840'"
-0.399
-0.587'
0.681 "
0.623'
-0.048
-0.248
-0.198
-0. 1 36
-0.220
-0.836'"
-0.600'
-0.085
Bond strengthl
Slipperiness
-0.081
-0.261
0.6 1 3'
Hardness
-0.802"
-0.775"
0.134
Deformation
0.61 2'
0.552
0.139
-0.432
0.498
0.302
0.090
Disintegration
0.766"
0.664'
0.303
-0.759"
0.747'"
0.438
0.186
Bond strength2
0.826'"
0.91 0'"
-0.499
-0.671 '
0.857"
0.584'
-0.025
Delamination
0.767'
0.874'"
-0.598'
-0.618'
0.744"
0.660'
-0:046
Note:
..
,..
5,8
0.893'"

- Significant at 0.01 P ;::: 0.001
- Significant at p
<
0.001
RELATIONSHIPS BETWEEN INGREDIENTS AND GLUE STICK

ATTRIBUTES
The relationships between ingredients and the attributes of the glue sticks had t o be
identified in order that the direction to improve the glue stick could be obtained .
Correlations between ingredients and glue stick attributes are shown in Table 5,19,
111
Table 5.19
Correlations between ingredients and output variables
Ingredient
Output Variable
Stearic acid
Open time
Coverage
Smoothness
Cleanliness
Bond strength 2
Disintegration
Thickness
Dry glue/Area
Wet glue/ Area
Delamination
Hardness(P)
Deformation
Hardness(S)
NaOH
Visibility
Hardness
Wet glue/ Area
-0.823'"
-0.821 '
0.816"
0.809"
-0.798"
-0.794"
-0.731 "
-0.71 4"
-0.708"
'
-0.668
0.652 '
'
-0.646
0.577'
'
0.683
-0.675'
Starch
Thickness
Adjustability
Glycerin
Slipperiness
Note:
0.604"
0.606'
-0.601'
0.644'
, - Significant at 0.05 P 0.01

... - Significant at p < 0.001
I t was found that stearic acid was the only ingredient which had significant effects on
most of the attributes. Stearic acid increased physical hardness, perceived hardness,
smoothness and cleanliness. It decreased disintegration, deformation, degree of
coverage, wet glue and dry glue per area, open time, bond strength after drying and
delamination. Sodium hydroxide decreased hardness and increased wet glue per area
and visibility of glue trail. Starch increased thickness of glue coated on paper and
decreased adjustability. Glycerin increased slipperiness while applying glue onto paper.
In the limits used for this experimentation, polyvinyl pyrrolidone and water were not
correlated with any glue stick attributes.
Multiple regressions between ingredients and glue stick attributes were conducted and
the results are shown in Table 5.20 and 5.21 .
112
Table 5.20

physical attributes
Physical a ttribute
Ingredient
t-ratio
Wet glue/ Area
16.303*NaOH
-5.429*Stearic acid
2.064*Starch
1 .026*Glycerin
8.13'"
-8.77'"
'"
5.50
2.64'
0.99
0.99
1 .1 06*Starch
5.84'"
-5.55'"
'"
4.62
0.929*Water
3.635*NaOH
'"
6.86
2.22 '
Dry glue/Area
7.134*NaoH
-2.104*Stearic acid
%Moisture
R2
0.99
1 1 .60'"
'
-8.94 "
2.42'
2.33'
0.99
0.99
0.999*Glycerin
0.873*Starch
0.406*Water
2.82'
"
4.47
3.44'
3.06'
2.59 '
Peel Strength
0.608*NaOH
0.1 37*Starch
2.23'
2.14'
0.98
Open time
12.745*NaOH
3.31 "
-2.90'
0.54
Hardness
0.767*Stearic acid
-1 .897*NaOH
0.096*PVP
0.093*Starch
Melting Point
4.184*NaOH
1 .273*PVP
-3.437*Stearic
Note:
'
.., - Significant at p < 0.001
113
Table 5.21

sensory a ttributes
Sensory Attribute
Ingredient
t-ratio
Smoothness
0.140*Water
0.471 *Stearic
4.41 "
3.97"
Cleanliness
0.1 17Water
0.302*Stearic
0.101 *Starch
6.34'"
5.60'"
0.076*Glycerin
4.31"
3.17'
0.68*PVP
2.89 '
0.99
0.99
Adjustabili ty
0.208*Water
3 1 .52'"
0.99
Hardness(S)
0.763*Stearic
'
30.6 1 ''
0.99
Deforma tion
0.129*Water
0.277*Starch
2.67'
2.00
0.98
Disintegration
0.426*Glycerin
0.289*Water
3.93"
3.07'
0.98
Coverage
1 .342*NaOH
0.354*Glycerin
2.92'
2.83'
0.98
Thickness
0.538*Starch
2 1 .41 '"
0.97
Visibility
2 .209*NaOH
1 8.86'"
0.97
Bond Strength before drying
0.395*PVP
1 8.87'"
0.97
Slipperiness
0.419*Glycerin
20.50' "
0.97
Bond Strength after drying
0.382*Starch
10.07'"
0.89
Delamination
0.248*PVP
5 . 1 0' "
0.68
Note:
- Significant at 0.05 ;::>: P ;::>: 0.01

.. - Significant at 0.01 ;::>: P ;::>: 0.001
.. , - Significant at p < 0.001
From the empirical equations, stearic acid decreased the amount of glue applied per
area but increased hardness. Starch increased hardness, peel strength and amount of
glue applied per area. In order to increase stickability (bond strength before and after
drying and delamination) polyvinyl pyrrolidone and starch had to be increased.
1 14
However, increasing amount of starch would also resul t in increasing of glue stick
deformation. Water increased smoothness, cleanliness, adjustability and also increased
disintegration and deformation. Glycerin increased slipperiness and also increased
disintegration.
5.9
DISCUSSION
5.9.1
Effect of Ingredients on Attributes of Glue Sticks
Most of the sensory attributes of the glue stick prototype from the formulation at the
centre point of the design were close to those of the commercial ones. The attributes
tha t had to be improved were slipperiness, disintegration and bond strength after
drying. From the experimentation, it was found that in order to reduce disintegration,
stearic acid had to be increased, starch and solvent had to be decreased. However, if
bond strength after drying were to be increased, amount of starch in the formulation
had to be increased and stearic acid had to be decreased . Slipperiness can be increased
by increasing the amount of solvent in the formulation.
It was decided at this stage that the ingredients were grouped into the major
components; adhesive, solvent and gel-forming agent. With these components, it should
be easier to assess the effect of these components on glue stick attribu tes and to find the
optimum level of each component.
5.9.2
The Suitable Levels of Basic Ingredients for Starch Based Glue Stick
From the comparison with commercial products and the product profiles, the attributes
of the sample from centre point formulation were closest to those of the commercial
glue sticks so it was decided that this formulation would be used as the basic
formulation to develop prototypes for consumer testing. This formulation was: 14.8%
modified tapioca starch, 14.8% polyvinyl pyrrolidone, 42.7% water, 1 2.0% glycerin,
1 1 .4% stearic acid, 3.5% sodium hydroxide and 3.5 grams of glyceryl monostearate for
every 1 00 grams of the glue mixture, i .e. 29.6% adhesive, 55.5% solvent, and 14.9% gel
forming agent.
1 15
5.9.3
Selection of Physical Attribute Measures
It was found that some physical a ttributes were highly related to sensory attributes of
glue sticks. Therefore those physical attributes could be used to assess the intensity of
the sensory attributes.
The physical attribu tes to be used for testing of the prototypes in the further study
were:
Amount of wet glue applied per area
5.9.4
,.
Amount of dry glue applied per area
,.
Percent moisture content
,.
,.
Open time
,.
Peel strength
Selection of Sensory Attributes for Consumers Testing
As there were 1 3 sensory attributes used in the sensory evaluation of glue stick, it was
necessary to reduce the number of attributes to be tested. The decision on which
a ttributes to be dropped from the test was based on the correlations between the
attributes.
Disintegration and deformation were significantly correlated. Since
disintegration could also be used to predicted other attribu tes like effect on paper
(smoothness and cleanliness) and stickability of glue (adjustability, bond strength before
and bond strength after drying), disintegration was chosen for further study. Hardness
was also related with disintegration. However, the trained panel found that it was
di fficult to rate perceived hardness if there was only a small di fference between samples
and it was not considered as a very important attribute by the consumers so it was
d ropped from the test.
Adjustability was correlated with bond strength before d rying. Both of them could be
used to predict the ability to reposition the paper after applying. Adjustability was
considered more suitable for consumer testing since it was the term which consumers
used to describe important attributes of glue stick in the consumer study. Bond strength
after drying and delamination were highly correlated. Bond strength after drying was
116
chosen because i t represented the ability o f glue to stick paper together. I t was found
that degree of coverage and thickness were significantly correlated, and from the
consumer study, degree of coverage was consider important so only degree of coverage
was selected to be used for further sensory testing. Slipperiness of glue when applying
was not correlated to any other attributes and it was consider a very important
attributes so it would be used further. Smoothness and cleanliness were highly
correlated, smoothness was chosen since it could also be used for evaluating wrinkling
caused by lumps of glue and also too much moisture on the paper.
The sensory attributes to be used for consumer testing of glue stick were:
*
Slipperiness of glue when applying
Disintegration of glue stick
Degree of glue coverage on paper
Adjustability of paper after applying glue
Smoothness of finished work
Bond strength after drying of glue
Although some of sensory attributes could be predicted using physical attributes, those
sensory a ttributes were still used in consumer testing in order that the study of
consumer evaluation of glue sensory attributes could be obtained.
5.10
CONCLUSIONS
It could be concluded that modified tapioca starch could be used as adhesive substance
in glue stick product together with polyvinyl pyrrolidone. The glue stick prototypes
obtained from adding tapioca starch in the formulation possessed attributes in the
ranges of the commercial glue sticks. However, there were some attributes that had to
be i mproved: slipperiness, disintegration and bond strength after drying. The next
experiment was conducted in order to investigate the effect of the glue stick major
components on the attributes of glue sticks particularly acceptability to the target
consumers.
117
CHAPTER 6
PRODUCT FORMULATION DEVELOPMENT WITH CONSUMERS
From the previous mixture design experiments, based on the formulation of

commercially available glue sticks, a glue stick basic formulation using modi fied tapioca
starch as the adhesive substance was developed . In this part of the thesis, further
experimentation was conducted in order to study the relationships between the
ingredients and the glue stick attributes in the modified tapioca starch based system.
The first part of this experiment was conducted at Massey University where glue stick
product prototypes were made. As the product was targeted towards Thai consumers,
the glue stick samples were then taken to Thailand to be tested with a Thai consumer
panel. Three consumer panels - university students, office workers and school children
were used to evaluate sensory attributes of the glue stick prototypes. The panel used
a sensory profile method to eval uate the affective attributes and sensory attributes of
the glue sticks on linea r scales. Affective attribu tes included 'acceptability', 'purchase
intention' and 'price to buy'. Sensory attribute scores for the prototypes and for the
ideal glue stick were also determined . The effects of glue stick components on attributes
were identified and empirical equations showing relationships between glue stick
components and the glue stick attributes as determined by consumers were developed.
6.1
AIM AND OBJECTIVES
The aim of this part of the project was to study the effect of formulation components
(adhesive, solvent, gel-forming agent) on glue stick sensory attributes using a consumer
panel as the subjective method and to generate empirical equations showing the
relationships between components and consumer sensory evaluation scores for use in
product optimization.

* Make prototypes derived from a systematic experimental plan for use in
consumer panel testing in Thailand.
1 18
>I-
Study consumers' performance in the sensory testing of a glue product.
* Evaluate the differences between consumers from different target market

segments in their response to the glue stick.
>I-
Study the relationships between components and sensory evaluation scores.
6.2
EXPERIMENTAL METHODS
6.2. 1
Basic formulation
The basic formulation obtained from the previous experimentation was:

grams
1 4.8
1 4.8
Glycerin
] 2.8
Water
42.7
Stearic acid
1 1 .4
Sodium hydroxide
3.5
1 00.0
3.5
In this experiment, the ingredients used were grouped into three major components:
adhesive, solvent and gel-forming agent. The 'adhesive' was starch and polyvinyl
pyrrolidone, the 'solvent' was glycerin and water, and the 'gel-forming agent' was
stearic acid in combinalion with sodium hydroxide. Using the formulation above as the
basic formulation, the ratio of ingredients in each component was:
Components
Ingredients
Ratio
Adhesive
Starch : PVP
1 .00:1 .00
Solvent
Glycerin : Water
1 .00:3.34
Gel-forming agent
NaOH : Stearic acid
1 .00:3.26
Glyceryl monosteafate was held constant at 3.5 grams for every 100 grams of the 3
component mixture.
1 19
6.2.2
Experimental Design
A constrained mixture design was used for this experiment. Using the results from the
previous experiment, the constraints for the three components were defined and the
ingredient levels of the basic formulation became the centre point of the design. The
ranges of the three components used in this experiment are shown below:
High level
Low level
Adhesive
23
38
Solvent
42
62
Gel-forming agent .
10
20
The complete mixture space showing the feasible area is shown in Figure 6.1. The limits
on the three components restrained the experimentation to the shaded feasible region
with the vertices shown in Table 6. 1 .
2 3 % A d h esive
\p .
3 8 % Adhesive
<.Q
4 2 % Solvent
----r.r,rr.rk-T--
-+
20o G e l
----------
__
__
__
__
__
__
__
__
1 00/0 G e l
Gel-fo r m i n g agent
f22l .
Figure 6.1
F e a s i b l e re g i o n
Complete mixture space showing feasible area for experimentation
120
Table 6.1
Experimental plan for mixture design with three components
Experimental run
Adhesive
Solvent
Gel-forming agent
38
52
10
23
57
20
38
42
20
23
62
15
28
62
10
6 (centre point)
30
55
15
7 (centre point)
30
55
15
The seven samples were made i n the laboratory using the method described in Section
3.S. Each sample was made in a 100 g batch. Samples were left to cool down at room
temperature for one hour then cut into sticks and placed in commercial type containers.
The samples were aged for at least 5 days before any testing was conducted.
6.3
PHYSICAL TESTING
The physical testing methods described in Section 3.3 were used to measure the
following attributes of the glue stick prototypes:
* wet glue per area

* dry glue per area
* moisture content
* hardness
* open time
* peel strength
6.4
SENSORY EY ALVATION BY CONSUMER PANEL IN THAILAND
The sensory evaluation of the prototypes were conducted with the target consumers in
Thailand. In this project, quantitative descriptive analysis using line scale was employed
in order to obtain quantitative data which would be used for generation of models
121
relating product ingredients with sensory attributes. During the test the consumers
indicated their ideal product which would be used as a guideline to optimize the
product. Floating ideal - consumers indicated their own ideal product - was used since
they were not trained for this particular testing.
6.4.1
Consumer Panel in Thailand
Three groups of consumers were selected for the consumer panel to be used for product
testing of the prototypes. These were:
* University students - undergraduate and post-graduate students at Kasetsart

University, Bangkok, Thailand aged between 18-25 year old.
* School students - high school students at Kasetsart University Demonstration

School, Bangkok, aged between 15-17 years old.
* Office workers - clerks and typists who worked at Kasetsart University and the
government offices in Kasetsart University campus.
These panelists were glue stick users and were expected to be the target market
segments for the new glue stick product. There were 30 panelists in each group.
6.4.2
Sample Preparation
Glue stick samples were coded with 3 digit numbers. Each panelist was given fresh
samples for the test. The seven samples were tested by university students. Only six
samples were tested with the other two groups; school students and office staff. This
was because the glue stick mixture from experimental run No. 3 was not homogeneous
and gave a soft stick which deformed easily. However this sample was tested by the
university students in order that its attributes could be evaluated . Only the university
student panel was selected for testing this prototype because of the limita tion of the
samples available and the time available from university students. To ensure that this
glue stick prototype would not affect the evaluation of the other samples, it was tested
a fter the university students had finished testing the other six samples.
122
6.4.3
Questionnaire
The questionnaire used for sensory evaluation of the glue stick prototypes is in Thai
la nguage. The questionnaire and the English translation are given in Append i x 6. 1 . The
questionnaire included an introduction and instructions for glue stick testing as well as
the set of testing forms. The first form included 3 line scales for evaluation of
'acceptability', 'purchase intention' and 'price to buy'. The second form included 5 line
scales for evaluation of sensory attributes: slipperiness, disintegration, degree of
coverage, adjustability and smoothness. The third form included a line scale for
stickability evaluation. At the end of the test, consumers were asked to indicate their
ideal product on the same line scale for each attribute in the second and third forms.
6.4.4
Preparation for the Consumer Panel Test
The necessary materials were prepared the day before the test date as follows:
Samples
Questionnaires
Paper for testing the glue samples which included two pieces - one large piece
(14.9 cm x 2 1 .0 cm) and one smal l piece (7.4 cm x 10.5 cm) of 80 gsm white
paper
Bonded paper prepared from the glue samples
Pencils and rubbers
Mechanical pencils as gifts
The bonded paper was prepared by rubbing a glue stick sample on a small piece of
paper (7.4 cm x 10.5 cm), six coats on each edge of paper. The coated paper was then
p laced on a large piece of paper ( 14.9 cm x 21.0 cm) and pressed down with four passes
( twice in each direction) of a 2 kg roller. This bonded paper was prepared about 24
hours before the test.
123
6.4.5
Consumer Testing Procedure
Two post-graduate students from the Product Development Department, Kasetsart

University and the author were the interviewers. Training of the interviewers was done
the day before the test date.
The test was conducted in three different places, one for each consumer group, because
of the need to make the testing as convenient as possible for the consumers.
* University students
The test was conducted in the Product Development Department, Kasetsart University.
Panelists were invited to the test room which was a room lit with daylight fluorescent.
They were seated separately to avoid any distraction. The test was done in 5 sessions
with 6 panelists in each session so the panelists could chose to come at the time they
were available.
* Office workers
Permission for a consumer test was requested one week in advance. Seven offices were
contacted and 4-5 panelists from each office took part in the test. The test was done in
either a meeting room or in their own office whichever was suitable. The rooms were
also lit with fluorescent light.
* School students
Permission for conducting the consumer test in Kasetsart University Demonstration

School was requested two weeks in advance. The test was conducted after school hours
at 4:00-5:00 p.m. in a classroom where there was no interference. The room was also lit
with fluorescent light. The test was done in two sections with 1 5 students in each
section.
124
A set of samples, a pencil, a rubber and test papers including bonded papers were
distributed to each panelist. Before starting the test, the panelists were asked to read the
instructions given at the beginning of the test. Then the meaning of each attribute term
was explained by the interviewer to avoid misinterpretation. The interviewers had all
been trained to give the same explanation for each term. The panelists were told how
the questionnaire should be completed as well as how the samples should be tested. A
large number of samples could be evaluated as it did not result in testing fatigue. The
samples were simultaneously presented to the panelists in random order. Panelists were
asked to rub the sample on the given paper before starting the test in order to smooth
the tip of the stick and also to remove any small bits of glue which might affect the
evaluation of the sample.
In the first part of the test, panelists were given the first form and were asked to test
glue stick samples on the given paper and evaluate 'acceptability', 'purchase intention'
and 'price to buy' for every sample. Panelists were told to mark the line at the position
representing their perception of the attribute with the coded number of the particular
sample. For 'price to buy', the price of the commercial product, UHU, was indicated on
the scale in order that the panelist would give the price of the sample in comparison
with the commercial product.
A fter panelists finished the first part, the first form was collected and they were given
the second form. They were asked to test the glue stick samples again and to evaluate
the sensory attributes of the samples. This was done after panelists finished the first
part so that the sensory attributes would not affect the evaluation of the affective
testing. They were also asked to mark their ideal point on the scale for every attribute.
In the third part, panelists were given the bonded paper prepared from each sample.
They were asked to peel the bonded paper apart and evaluate the stickability for each
sample as well as the ideal for stickability.
Panelists could ask questions during the test from the interviewers. After the panelist
finished testing, the interviewers checked the results to make sure that panelists had
completed all the questionnaires. After the test, each panelist was reward with a pencil
for their participation. The test lasted about 45 to 60 minutes.
125
6.5
The scores on the line scale for each attribute were measured in centimetres from the
zero end of the scale and were called 'sample scores'. 'Ideal scores' for each a ttribute
were also measured . 'Ideal ratio scores' were calculated as the ratio of the 'sample score'
to the 'ideal score' . This was done using the scores from individual panelists. An ideal
ratio score less than 1 meant that the attribute was less than the ideal. An ideal ratio
score greater than 1 meant that the attribute was greater than the ideal, an ideal ratio
score of 1 indicated that the attribute was 'ideal'. 'Log of ideal ratio scores' were
obtained by taking logarithms of the ideal ratio scores.
The mean score and standard deviation for each glue stick sensory attribute for each
group of panelists as well as for all the panelists were computed for sample and ideal
scores as well as ideal ratio scores and log of ideal ratio scores. Means of log of ideal
ratio scores were transformed back to sample ideal ratio scores. VP Planner was used
to calculate the means and standard deviations of the scores. Di fferences in the mean
attribute scores for each group in the consumer panel were analyzed using analysis of
variance (ANOVA) which was run on the Minitab 8.2 computer programme. The mean
sensory scores of the sensory profiles from the three groups of consumers in the panel
were correlated using Minitab 8.2. This programme was also used to generate
correlations and the empirical equations showing the relationships between glue stick
attributes and the components in the formulation.
6.6
PHYSICAL ATTRIBUTES OF THE PROTOTYPES
Physical testing results of the glue product prototypes are shown in Table 6.2.
126
Table 6.2
Run
No.
Physical attributes of the prototypes

Wet glue
per area
(g/m2)
Dry glue
per area
(g/m2)
Moisture
content
(%)
Hardness
Open time
Peel strength
(Newton)
(minute)
(Newton)
237.9 (6.8)"
1 33.0 (3.7)
39.3 (0.2)1
3.66 (0.65)'
1 4.00 (0.00)-
4.55 (0.07)"
50.8 (5.4)d
25.6 (3.2)'
47.9 (0.7)'
3.78 (0.44)'
0.25 (0.00)1
4.77 (0.32)-
1 52.9 (1 4.3)b
1 02.1 (1 2.W
29.3 (O.4)g
3.59 (0.85)'
1 2.00 (0.35)b
4.32 (0.28)-
40.6 (3.2)"
1 7.2 ( 1 .6)"
51 .9 (0.1 )"
5.28 (0.55)'
0.25 (0.00)1
4.70 (0.52)"
59.0 (0.9),d
25.7 (0.2)'
50.2 (O. l )b
4.57 (0.49) b
1 0.25 (0.35)'
5.20 (0.72)-
32.3 (0.6)'
45.2 (O.4)d
4.58 (0.57)b
6.00 (0.00)"
5.30 (0.89)"
4.36 (0.74)b
6.50 (O.OO)d
5 . 1 0 (0.85)"
6
7
Note:
64.6 ( 1 . 1 )'
64.3 (2.2)'
3 1 .7 (0.8)'
43.7 (0.5)"
( 1 ) The numbers in the parentheses are standard deviations.

(2) Mean scores w i thin the column followed by a different letter are significantly d ifferent
at p<0.05
-
Prototype Run 1 gave the highest amount of wet glue and dry glue per area and open
time followed by Run 3. Both prototypes had high amounts of adhesive in the
formulations. Prototype Runs 2 and 4 had the lowest amounts of adhesive and gave the
lowest wet glue and dry glue per area and open time. Peel strengths were not
significantly different between the samples. Hardness of Run 4 was highest followed
by Runs 5, 6, and 7. Moisture contents of the samples were significantly different. Run
3 had lowest moisture content (29.3%) followed by Run 1 (39.3%); the other samples
ranged from 43.7 to 51 .9%.
Table 6.3
Comparison between physical
attributes of the prototypes and
commercial products
Commercial
Prototype
Wet glue per area (g/ m2 )
31 .9-45.8
40.6-237.9
Dry glue per area (g/ m2)
10.3-14.9
17.2-133.0
Moisture content (%)
57.0-67.3
29.3-5 1 .9
Hardness (Newtons)
2.42-3.55
3.59-5.28
Open time (Minutes)
1 2.25-99.00
0.25-14.00
3.65-6.30
4.32-5.30
Physical Attributes
Peel strength (Newtons)
127
The physical properties of these prototypes were compared with those of the
commercial ones from Section 5.3. I t was found that amount of glue residue per area
(wet glue and dry glue per area) of the prototypes were very much higher than the
commercial glue sticks. Only Run 4 gave wet glue per area in the commercial range.
The moisture contents were very low compared to those of the commercial ones.
Although hardness was higher than the commercial glue stick, it was consider
acceptable since lower hardness would give high amount of glue residue per area. Only
open time of Runs 2 and 4 were too low, the other prototypes' open time were
acceptable. Peel strength of the prototypes was in the acceptable range.
Therefore in the optimization stage, i t was necessary to reduce the amount of glue
residue per area and increase the moisture content of the glue stick. Open time should
be at least one minute as recommended by Pletcher and Wong (1978).
6.7
SENSORY ATrRlBUTES OF THE PROTOTYPES EVALUATED BY CONSUMER

PANEL
From the consumer testing, each panelist rated each sample on affective and sensory
attributes. The sensory attributes' scores of each glue stick sample includi ng the scores
of the consumer's ideal product obtained from each individual from the three consumer
panels are shown in Appendix 6.2.
6.7. 1
Affective Attributes
Group mean scores and total mean scores of the glue stick acceptability, purchase
intention and price to buy were computed and are shown in Table 6.4.
1 28
Table 6.4
Acceptability, purchase intention and price to buy of the prototypes
Samples
Groups
Acceptabili ty
Purchase Intention
Price t o buy
University
5.5 (3.4)
4.6 (3.5)
2.4 (2.4)
Office
5.1 (4.4)
4.0 (4.3)
2.4 (3. 1 )
School
4.9 (3.6)
3.6 (3.4)
2 . 6 (3.2)
TOTAL
5.2 (3.W
4.1 (3.7)"
2.4 (2.W
University
6.7 (3.8)
5.9 (4.0)
3.3 (2.6)
Office
5.3 (3.9)
4.9 (4.7)
2.9 (2.8)
School
5.2 (4.0)
4.4 (4.6)
2.4 (3. 1 )
TOTA L
5.7 (3.9)d
5.1 (4.4)d
2.9 (2.9)d
University
9.9 (3.8)
9.4 (4.3)
5.4 (3.4)
Office
9.5 (3.5)
7.7 (5. 1 )
5.0 (3.8)
School
8.8 (4. 1 )
8.6 (4.9)
5.1 (3.9)
TOTA L
9.4 (3.8)b
8.6 (4.8)b
5.2 (3.6)b
University
1 1 .2 (2.8)
10.8 (3.3)
6.5 (2.9)
Office
1 0.4 (3.2)
9.4 (4.2)
5.8 (3.6)
School
1 0.0 (3.3)
9.5 (3.5)
6.5 (3.7)
TOTA L
1 0.5 (3. 1 )"
9.9 (3.7)"
6.3 (3.4)"
University
7.2 (3.9)
6.2 (4.2)
3.6 (2.8)
Office
8.1 (3.5)
6.9 (4.4)
4.4 (3.3)
School
7.0 (3.5)
6.0 (3.7)
4.1 (3.5)
TOTAL
7.5 (3.6)<
6.4 (4. 1 )'
4.1 (3.2)'
University
6.8 (3.6)
6.1 (4.0)
3.6 (2.8)
Office
7.3 (3.7)
5.8 (4.7)
3.9 (3.9)
School
7.2 (3.5)
6.3 (4.1 )
4.2 (3.7)
TOTAL
7.1 (3.6)'
6.1 (4.2)C
3.9 (3.5)'
Note:
(1)
(2)
(3)
(4)
Samples from Run 3 were not tested.

The number in the parentheses are standard deviations
Scores are on scale 0 to 1 5
Acceptability
15 very acceptable
0 not acceptable
Purchase intention
15 certainly buy
0 never buy
1 5 20 baht
Price to buy
0 8 baht
Mean scores within the column followed by a different letter are significantly different at
p<0.05
=
Although the results had high standard deviations, which was expected in the test with
consumers, the analysis of variance (shown in Appendix 6.2) showed tha t the three
consumer groups were not significantly different in evaluating acceptability (p=O.073),
purchase intention (p=O.050) and price to buy (p=O.949). This indicated that the three
129
groups perceived these attributes in the same way.
Run 5 had the highest acceptability, purchase intention and price to buy scores followed
by Run 4. Run 1 had the lowest affective scores followed by Run 2. These two
prototypes were very different in the level of components in their formulation. Run 1
contained the highest level of adhesive substance and lowest level of gel-forming agent
while Run 2 had the lowest level of adhesive substance and highest level of gel-forming
agent.
Acce ptab i l ity

score
,.5
Run
Run
Figure 6.2
R u n 4-
Run
Run 6
Acceptability of the prototypes
Run ,
-------
-----
1 30
6.7.2
Sensory Attributes
The mean scores
of
sensory attributes of glue stick prototypes obtained from the
consumer testing were calculated and are summarised in Table 6.5.
Table 6.5
Samples
Note:
Sensory attributes of the prototypes
Groups
Slip
Disintegrate Coverage
Adjust
Smooth
Stick
Univ.
5.1 (3.6)
9.3 (3.6)
7.2 (3.7)
8.0 (4. 1 )
7.9 (4.4)
10.5 (3.0)
Office
4.4 (3.7)
8.9 (4.0)
7.3 (3.7)
6.2 (4.0)
6.5 (4.2)
9.4 (4.1 )
School '
2.8 (2.5)
9.8 (3.9)
7.3 (3.8)
5.3 (3.3)
7.9 (4.0)
9.7 (4.1 )
TOTAL
4.1 (3.4)<
9.4 (3.8)"
7.2 (3.7)<
6.5 (3.9)a
7.4 (4.2)<
9.9 (3.7) b
Univ.
8.2 (2.9)
10.2 (4.0)
7.2 (3.0)
8.9 (3.5)
9.0 (4.2)
9.2 (3.2)
Office
4.9 (3.4)
10.6 (3.8)
7.0 (3.7)
7.8 (4.0)
7.9 (3.9)
9.5 (4.2)
School
8.7 (3.7)
10.0 (4.5)
6.5 (3.7)
8.0 (4.2)
8.8 (4.4)
7.3 (4.6)
TOTAL
7.2 (3.7) b
10.3 (4. 1 )a
6.9 (3.5)<
8.2 (3.9)"
8.6 (4.2)b
8.7 (4.1 j<
Univ.
12.1 (2.4)
3.0 (3.3)
1 0.6 (4.2)
9.2 (4.2)
1 2.6 ( 1 .5)
9.4 (3.2)
Office
1 2.4 (2. 1 )
4.0 (3.9)
8.9 (3.7)
1 0.3 (3.2)
1 1 .2 (3.4)
8.3 (4.3)
School
1 1 .3 (3.8)
4.4 (4.8)
8. 1 (4.4)
7.8 (4.2)
1 1 .1 (2.9)
6.0 (4.2)
TOTAL
1 1 .9 (2.9)a
3.8 (4.1 )<
9.2 (4.2)b
9.1 (4.0)a
1 1 .6 (2.8)a
7.9 (4.2)<
Univ.
1 1 .2 ( 1 .4)
3.7 (3.3)
1 1 .4 (2.2)
7.5 (4.6)
123 (2.4)
1 1 .6 (2. 1 )
Office
1 1 .0 (2.8)
5.2 (4.1 )
1 0.3 (2.8)
1 0.3 (3.5)
1 1 .3 (3. 1 )
1 1 .3 (3.3)
School
1 0.3 (3.2)
4.2 (3.7)
1 0.0 (3.8)
8.1 (4.0)
10.4 (3.9)
1 0.2 (3.6)
TOTAL
1 0.8 (2.6)a
4.3 (3.7)'
1 0.6 (3.0)a
8.7 (4.2)a
1 1 .3 (3.3)"
1 1 .0 (3.1 )a
Univ.
7.5 (3.3)
8.1 (3.3)
9.1 (2.7)
7.5 (3.9)
9.5 (3.5)
1 0.8 (3.3)
Office
7.1 (3.3)
7.5 (3.2)
9.6 (2.7)
8.9 (3.3)
9.2 (3.5)
1 0.9 (3.3)
School
7.5 (2.8)
6.7 (3.7)
8.4 (2.7)
6.6 (3.4)
9.5 (3.9)
9.5 (3.5)
TOTAL
7.4 (3. 1 )b
7.5 (3.4)b
9.0 (2.7)b
7.7 (3.6)"
9.4 (3.1 )b
1 0.4 (3.4)ab
Univ.
6.7 (3.2)
7.6 (3.9)
8.6 (3.5)
9.1 (3.5)
9.2 (3.7)
1 0.6 (3.2)
Office
6.6 (3.4)
7.5 (4. 1 )
9 . 1 (3.1 )
7.8 (3.9)
8.0 (4. 1 )
1 1 .2 (3.5)
School
7. 1 (3.8)
6.4 (3.9)
8.5 (3.8)
7.2 (3.5)
8.9 (3.4)
10.2 (3.4)
TOTAL
6.8 (3.6) b
7. 1 (4.W
8.7 (3.W
8.0 (3.7)"
8.7 (3.9)b
1 0.7 (3.3)"b
(1)
(2)
The number in the parentheses are standard deviation values

Scores are on scale 0 to 1 5
15 = slips
Slipperiness
0 = drags
15 = high
DiSintegration
0 = low
1 5 = very easy
o = very d ifficult
Adjustability
15 = very smooth
o
very wrinkly
Smoothness
15 = paper stuck very strongly
o = easy to peel
Stickability
Mean scores within the column followed by a different letter are significantly d i fferent at
p <0.05
=
(3)
131
The data in Table 6.5 show that Runs 5 and 4, which had the highest affective scores,
also had the highest scores in terms of slip, coverage, adjustability and smoothness, both
had lowest scores on disintegration. Run 5 also had the highest stickability but Run 4
had lowest stickability.
The scores for sensory attributes of the ideal glue stick obtained from the consumers are
shown in Table 6.6.
Table 6.6 Ideal scores of the sensory attributes of the prototypes (Group mean scores)
Group
Slip
Disintegrate
Coverage
Adjust
Smooth
Stick
University
12.1 (1 .5)
1 .6 (2.1 )
1 2.9 (1 .5)
10.1 (4.2)
1 3.4 ( 1 .4)
1 1 .3 (2.5)
Office
1 2.8 (2.0)
2.4 (2.7)
1 2.7 (2. 1 )
1 2. 1 (2.7)
1 3.2 ( 1 .8)
1 1 .6 (3.2)
School
1 2.3 (2.9)
1 .4 (2.0)
1 3.0 (2.4)
1 1 . 1 (3.2)
1 3.9 ( 1 .3)
1 1 .4 (3.6)
TOTAL
1 2.4 (2.2)
1 .8 (2.3)
1 2.9 (2.0)
1 1 .1 (3.5)
1 3.5 ( 1 .5)
1 1 .4 (3.1 )
Note:
The analysis of variance showed that there were n o significant d ifferences for the ideal
scores between the three consumer panels for every attribute (p>O.OS). This indicated
that consumers from the three groups had similar ideal products. The ideal scores for
adjustability and stickability had high standard deviations. This might have happened
because there were a few panelists who gave very low ideal scores for these attributes.
6.8
RESPONSE OF CONSUMERS IN DIFFERENT MARKET SEGMENTS
Relationships between consumer sensory evaluation scores from each group were
studied. Correlations between the sensory attribute scores are shown in Table 6.7.
1 32
Table 6.7
Correlations between sensory scores obtained from three groups of

consumers
Attribute
Acceptability
Correlation Coefficient(l)
University
Office
School
.927'"
.950'"
.942'"
.984....
.963'"
"
.963'"
.937'"
.921 '''
.974....
.939'"
Office
Purchase Intention
University
Office
Price to buy
.939 '
University
Office
Slipperiness
University
.814'
Office
Disintegration
.978 ....
University
Office
Coverage
.864
University
"
-.080
University
.974....
University
Office
Stickability
.809
University
'
Significant
Significant
Significant
Significant
.310
.970'"
.975.. ..
.872"
.871 "
Office
Note: ( 1 )
"
.743'
Office
Smoothness
.835
.950'"
Office
Adjustabil i ty
.936'"
.949 '"
at
at
at
at
0.1 > P 0.05

0.05 > P 0.01
0.01 > P 0.001
p < 0.001
The critical value of the Pearson product moment correlation coefficients were used to
determine the significant level of correlations. There were good correlations between the
scores obtained from the three groups in terms of affective attributes, slipperiness,
d isintegration, coverage and smoothness. There was only slight correlation for
stickability. There was no correlation for adjustability. This might have occurred because
the adjustability of the samples were not significantly different.
133
At this stage, since the resul ts from the three groups of consumers were parallel, it was
decided that the total data from the three consumer panels be combined and used in
fu rther analysis of the consumer results.
6.9
IDEAL RATIO SCORES AND LOG OF IDEAL RATIO SCORES OF THE

SENSORY ATTRIBUTES OF THE PROTOTYPES
6.9.1
Ideal Ratio Scores of the Sensory Attributes of the Prototypes
Ideal ratio scores of the glue stick attributes were calculated by dividing the attribute
score by the ideal score from each individual - floating ideal scores were used. The ideal
ratio scores from all consumer panelists are shown in Appendix 6.3. The mean ideal
ratio scores of the glue stick attribu tes were calculated and are shown in Table 6.8
Table 6.8
Samples
Ideal ratio scores of the sensory attributes of the prototypes (Original data)
Slip
Disintegrate
Coverage
Adjust
Smooth
Stick
0.3 (0.3)
25.9 (36.4)
0.6 (0.3)
2.7 (1 4.3)
0.6 (0.3)
0.9 (0.5)
0.6 (0.4)
28.4 (42.0)
0.6 (0.3)
2.5 ( 1 2.7)
0.7 (0.3)
1 .0 ( 1 .6)
1 .0 (0.4)
9.4 (25.9)
0.7 (0.3)
1 .7 (5.8)
0.9 (0.2)
1 .0 (2.8)
0.9 (0.3)
1 1 .2 (22.9)
0.8 (0.2)
1 .3 (4.8)
0.9 (0.3)
1 .6 (5.8)
0.6 (0.3)
1 9.8 (31 .4)
0.7 (0.2)
1 .6 (7.2)
0.7 (0.3)
2.0 (10. 1 )
0.6 (0.4)
1 7.0 (25.2)
0.7 (0.3)
2.9 (1 4.9)
0.7 (0.3)
2.7 ( 1 5.8)
Note:
When the data were transformed to ideal ratios, standard deviations of some attributes
were high: disintegration, adjustability and stickability. This was because the placing
of the floating ideal varied a great deal for these attributes, see Table 6.6. Hence, it was
decided to used log of ideal ratios to compare the prototypes with the ideal product and
to reduce the effect of high scoring panelists.
6.9.2
Log of Ideal Ratio Scores of the Attributes of the Prototypes
Log of ideal ratio scores of the glue stick attributes obtained from the 3 groups of
consumer panel are shown in Appendix 6.4.
1 34
Ideal ratio scores of the glue stick attributes obtained from the logarithm transformation
are shown in Table 6.9
Table 6.9
Ideal ratio scores of the sensory attributes of the prototypes (Transformed

data)
Samples
Slip
Disintegrate
Coverage
Adjust
Smooth
Stick
0.2
9.4
0.5
0.5
0.4
0.8
0.5
10.4
0.5
0.7
0.5
0.7
0.9
2.4
0.6
0.8
0.8
0.6
0.9
3.1
0.8
0.7
0.8
1 .0
0.5
7.6
0.7
0.7
0.6
0.9
0.5
6.7
0.6
0.7
0.5
1 .0
Note:
Ratios were analyzed using logarithms
Calculation of ideal ratio scores using logarithm transformation reduced the effect of
some panelists who gave ideal scores very low for disintegration and very high for
adjustability and stickability.
From the ideal ratio scores in Table 6.9, Runs 4 and 5 were closest to the ideal ( 1 .00) in
all attributes except that Run 4 was furthest in terms of in stickability. The
disintegration was too high for all samples. Generally all the other attributes needed to
be increased slightly.
6.10
RELATIONSHIPS BETWEEN ATTRIBUTES OF THE PROTOTYPES
6.10.1 Physical
A t tr i b u tes
From correlations between physical attributes in Table 6.10, wet glue and dry glue per
area were highly correlated (p<O.OOl) and were slightly correlated with open time
(p<O.lO) . Hence, open time was dependent on the amount of glue applied per area. It
was found that moisture content had negative correlation with wet glue and dry glue
per area. It can be said that glue sticks which had low moisture contents, i.e. high
amount of solid in the formulation - mainly adhesive substance - gave high glue residue
per area.
135
Table 6.10
Correlations between physical attributes of the prototypes from mixture

design experiment with three components
Wet glue/area Dry glue/area
Dry glue/ area
Moisture
H ardness
0.999'"
%Moisture
-0.813'
-0.822'
Hardness
-0.621
-0.636
0.709
Open time
0.765
0.747
-0.650
-0.413
-0.532
-0.543
0.190
0.31 6
Peel strength
Note:
Open time
0.047
'
..
...
- Significant at p
<
0.001
From Table 6.1 1, it was found that disintegration was negatively correlated with degree
of coverage, slipperiness, and smoothness of finished work. It can be said that glue stick
w i th high disintegration would give low slipperiness while applying, leave low degree
of glue covered on paper and cause wrinkly finished work.
Table 6 . 1 1 Correlations between sensory attributes of the prototypes from mixture

design experiment with three components
Acceptability Purchase
Purchase
Price
Slipperiness
Disintegrate
Coverage
Adjust
Smooth
Stick
Note:
0.996'"
0.998'"
0.906'
-0.946"
0.942"
0.761
0.954"
0.131
'
..
..
0.995'"
0.923"
-0.927"
0.917"
0.793
0.960"
0.098
Price
0.888'
-0.928"
0.951"
0.752
0.937"
0.184

, - Significant at p < 0.001
Slip
-0.867'
0.728
0.931"
0.985'"
-0.273
Disintegrate Coverage Adjust
-0.893'
-0.695
-0.920"
-0.019
0.564
0.815'
0.4 1 6
0.862'
-0.329
Smooth
-0.164
1 36
From the correlations, i t was found that acceptability, purchase intention and price to
buy were highly correlated (p<O.OOl), see Table 6 . 1 1 . This indicated that acceptability
could be used to predict purchase intention and the price consumers were prepared to
pay for this product. For further analysis, only acceptability was used in analysis of
results.
6.10.3 Acceptability and Sensory Attributes
Correlations between acceptability and sensory a ttribute scores are shown in Table 6.12.
Smoothness, d isintegration and degree of coverage were highly correlated with
acceptability. This indicated that these attributes could be used to indicate acceptability
of the product. Adjustability was correlated with acceptability at p<0. 10. Stickability was
not correlated with acceptability even though from the previous survey, stickabil ity was
found to be an important attribute for glue stick.
Table 6.12
Correlation between acceptability and sensory attributes
Attribute
Correlation coefficient{l)
0.954'"
Smoothness
Disintegration
Coverage
-0.946'"
0.942'''
Adjustability
0.906 "
0.761 '
Stickability
0.131
Slipperiness
Note:
Significant at 0.01 > P 0.05

6 . 1 0.4 Sensory Attributes and Physical Attributes
Correlations between sensory attributes and physical attributes are shown in Table 6.13.
137
Table 6.13
Correlations between sensory attributes and physical attributes
Sensory a ttributes
Physical a ttributes
Acceptability
Hardness
Slipperiness
Correlation coefficients
"
0.813'
Moisture
0.775"
Moisture
0.954. ...
0.852'"
Hardness
"
Disintegration
Hardness
-0.900'
Adjustability
Moisture
Dry glue per area
0.959 .
-0.892'"
Wet gl ue per area
-0.884'
.. .
"
Coverage
Hardness
Smoothness
Moisture
0.747'
0.877'"
Hardness
0.877' "
Significant
Significant
Significant
Significant
Note:
at
at
at
at
0.01 P 0.05
0.05 > P 0.01
0.01 > P 0.001
p < 0.001
It was found that hardness, moisture content, wet glue and dry glue per area were
correlated with most of the sensory attributes of glue stick. Hardness and moisture
content significantly correlated with acceptability, slipperiness and smoothness,
therefore they could be used to assess perception of the consumers toward these
a ttributes. I f hardness and moisture content of glue stick increased those sensory
attributes also increased. Disintegration and degree of coverage were also dependent
on hardness, the harder the glue the lower the disintegration and higher degree of
coverage. Adjustability increased with the increasing of moisture content but the
decreasing of wet glue and dry glue per area. Open time and peel strength were not
related to any sensory attributes.
6. 1 1
RELATIONSHIPS BETWEEN THE COMPONENTS AND ATTRIBUTES OF THE

PROTOTYPES
Correlations between the components and attributes of prototypes were calculated and
138
Table 6.14
Correlations between components and attributes of the prototypes
G lue stick component
Attributes
Adhesive
SENSORY AITRIBUTES
Adjustability
Solvent
Slipperiness
Correlation coefficient
Adjustability
0.950'"
0.9 1 9'"
Acceptability
0.897"
PHYSICAL
ATTRIBUTES
Open time
0.906"
Smoothness
Adhesive
Moisture
Wet glue/ area
Dry glue/area
Solvent
Moisture
Gel-forming agent
Open time
Note:
-0.896"
0.975 ...
Significant
Significant
Significant
Significant
at
at
at
at
-0.885"
"
0.879
0.873"
0.959
-0.871
'"
'
0.01 > P 0.05

0.05 > P 0.01
0.01 > P 0.001
p < 0.001
From the correlation coefficients, it appeared that the solvent had significant effects on
acceptability and sensory attributes such as slipperiness, smoothness, and adjustability;
adhesive had an effect on adjustability. This agrees with the results from the previous
experiment (see Chapter 5). Adhesive had effects on most of the physical attributes
w hich was not found in the previous experiment. Gel-forming agent affected open time,
similar to previous experiments.
E mpirical equations obtained from multiple regression between sensory and physical
attribute and glue stick components are shown in Table 6. 15.
1 39
Table 6.15
Empirical equations showing relationships between components and

attributes of the prototypes
Attributes
Regression Equations
SENSORY ATTRIBUTES
Acceptabili ty
-0. 156*Adhesive
+0.293*Solvent
-0.332*Gel
0.99
Slipperiness
-0.315*Adhesive
+0.372*Solvent
-0.296*Gel
0.99
Disintegration
0.243*Adhesive
0.89
Coverage
0.151 *Solvent
0.99
Adjustability
-0.052*Adhesive
+0. 1 66*Solvent
0.99
Smoothness
0.167*Solvent
0.99
Stickability
0.168*Adhesive
+0.086*Solvent
0.99
Wet glue/ area
2.660*Adhesive
-0.265*Solvent
0.99
Dry glue/area
1 .491 *Adhesive
-0.233*Solvent
0.99
Moisture
-0.020* Adhesive
+0.882*Solvent
+0.121 *Gel
0.99
Hardness
-0.012*Adhesive
0.083*Solvent
0.99
Open time
0.564*Adhesive
-0.005*Solvent
-0.688*Gel
0.95
Peel strength
0.043*Adhesive
+0.055*Solvent
+0.040*Gel
0.99
PHYSICAL ATTRIBUTES
From this mixture design experiment, linear equations fitted the data very well with
high R-squared values and there was no need to go to a higher level equation. The
equations obtained were used later for glue stick optimization.
140
6.12
DISCUSSION
This study showed tha t quantitative descriptive analysis can be used effectively with
consumers in measuring glue stick sensory attributes and to generate the ideal product
profile. Al though the consumer panelists did not undergo any sensory training, with
only some instruction for sensory testing, they could perform the task which would be
normally conducted by the trained sensory panel. The results from the test showed that
consumers from different groups gave similar results for the affective tests and also the
sensory attributes, except adjustability. This could have happened because the panelists
could not identify the differences between adjustability of the prototypes.
The results from this consumer testing showed that some of the prototypes were
reasonably accepted by the consumers. However, in order to decide in which direction
the product should be improved, the ideal product profile assigned by the consumers
could be used as a guideline. By using only sample scores without an ideal, the
researcher cannot decide how good the product is, while using 'ideal ratio' he/she can
tell how different the product attribute is from the ideal attribute and in what direction.
Sinthavalai (1986) compared using scores and ideal ratio scores. She concluded that
using scores only, ideals needed to be shown so that the difference of the product in
magnitude and direction from what the consumers desired could be determined. Ideal
ratio had advantage over product attribute scores in that they were quick to present,
read and interpret.
During the test, each consumer was asked to indicate his/ her own ideal glue stick
product on the scale for each attribute. It was found that some consumers gave a very
low or a very high ideal attribute score compared with other consumers in the group.
This is the problem with normally occurs in using floating ideals, consumers tend to use
d ifferent part of the scales and the ratio values vary greatly. In this study, using a mean
of the ideal ratio scores, some large ra tios affected the results markedly but when the
data were logarithmically transformed, the effects of these few 'outsider' scores were
reduced. In the sensory testing of a Nham product using floating ideals, Wiriyacharee
( 1 990) found tha t when ideal ratio scores were transformed to the logarithms of the
ideal ratio scores, i t gave a greater confidence in analysis.
141
Attrlbut..
Acceptability
,...-----r---,
Purchase Intention
Price to buy
Slipperlne88
Disintegration
Coverage
AdJustabll1ty
Smoothness
Stickablllty
10
12
14
Saneory Sco....
Run4 -- RunS
Figure 6.3
--
Ideal product
Product profiles of the prototypes compared with ideal product profile
1 42
The ideal ratios of each prototype, which were obtained from the logarithm
transformation, were compared against ' 1 .0' - the consumers' ideal product. There were
2 prototypes, Runs 4 and 5, closer to the ideal product than the others. Both were
slightly low on slipperiness, coverage, adjustability, and smoothness, however, the
diSintegration was markedly too high. Run 5 was ideal on stickability. The profiles of
Run 4 and 5 against the ideal profile are shown in Figure 6.3.
Physical attributes of Run 4 and 5 also were closer to those of the commercial products.
The wet glue per area, dry glue per area and hardness were only slightly higher and
moisture content and open time were slightly lower than the commercial products.
So the formulations from these two prototypes could be used as a guidance for further
product optimization. These prototypes had the following formulations:
Adhesive (%) Solvent (%)
Gel-forming agent (%)
Run 4
23
62
15
Run 5
28
62
10
The solvent in these formulations were higher than the other prototypes and from the
relationships between ingredients and sensory attributes, solvent increased acceptability,
slipperiness, adjustability, smoothness and stickability. This indicated that in order to
improve these attributes - move them closer to the ideal attributes - amount of solvent
in the formulation should be increased . The level of adhesive and gel-forming agent of
Runs 4 and 5 were at the lower level and middle of the ranges used in the
experimentation. It was found from the ingredient/sensory attribute relationships that
adhesive and gel-forming agent decreased acceptability and slipperiness. Adhesive also
decreased adjustabili ty and increased disintegration of glue stick. It was decided that
the level of these two components should be decreased. Although decreasing of amount
of adhesive in the formulation would result in decreasing of stickability, in order to
increase acceptability and improve other attributes the stick ability had to be traded off.
In this study, it was found that acceptability, purchase intention and price to buy were
highly correlated with each other (p<0.001). This indicated that the product which
obtained high acceptability would have high opportunity to be purchased and the
143
consumer would be prepared to pay a high price for that product.
Moskowitz (1983) stated that purchase interest ratings, often parallelled acceptability
ratings. Those products which score highest on over all acceptability usually score
highest on purchase interest ratings. However, he also mentioned that from time to
time, acceptability and purchase interest may correlate only modestly with each other,
or even correlate negatively. This will occur in the case that panelists purchase the
product for reasons other than sensory attributes or overall acceptability, e.g. breath
fresheners which often have highly unacceptable flavours but these flavours signal
efficacy.
In this study, acceptability was positively correlated w ith physical hardness, moisture
content, slipperiness, degree of coverage and smoothness and negatively correlated with
d isintegration. Consequently these attributes can be used to indicate product
acceptabi li ty. The consumers accepted the products with high slipperiness, degree of
coverage and smoothness but low disintegration.
Peel strength was not correlated with any sensory attributes which agreed with the
results from previous study. This indicated that the peel strength testing method used
in this study cannot be used to determine perceived stickability of glue stick at least for
the ranges of components used in this study. Therefore, if glue stick was made
commercially it was necessary to find another suitable method which could be used to
predict consumers' reactions towards stickability.
6.1 3
CONCLUSIONS
Consumer panels can be used as a subjective method in identifying the quanti tative
effect of ingredients on sensory attributes of non-food products, in this case glue stick
product. They can be used to identify differences between products. Consumers had no
trouble in using the line scale and most of them had no difficulty with scoring their
ideal product. There were no differences between the three consumer groups in terms
of acceptability, purchase intention, price to buy and most of the sensory attributes, i t
can be concluded that the target consumers can be regarded a s having identical needs
144
in the product. It was also found that with non-food products like glue stick, consumers
were able to test the samples up to 6 to 7 samples without any difficulty. Significant
linear models relating the components with glue stick attributes were also obtained.
The use of consumer ideal product profile as the reference to compare the developed
prototypes was found very useful since glue stick was a new product compared to
other existing glue products, no standards had been set for the physical testing and for
suitable levels of physical attributes. At the same time, the direction in which the
product should be improved was also obtained.
In the next stage in which the optimization technique was used to generate optimum
glue stick formulation, the limi ts of each ingredient would be as followed:
Lower limit (%)
Upper limit (%)
Adhesive
15
30
Solvent
60
70
Gel-forming agent 1 0
15
Consumer ideal profile would be used to set limits for sensory constraints, with the
upper and lower bounds being calculated by adding 2 and substituting 2 from the ideal
a ttribute levels. This was done in order to allow a realistic d istance that the optimum
product could be from the ideal. The physical attributes of commercial glue stick would
also be used to generate upper and lower bounds of physical attribute constraints.
1 45
CHAPTER 7
Linear programming is a mathematical technique used in product formulation problems

to determine the optimum formulation, subject to certain constraints, and to either
maximize or minimize a specified objective. At this stage, linear programming would
be applied to generate optimum glue stick formulations. Empirical equations showing
relationships between glue stick attributes and components in the formulation were
used to develop constrain ts for the linear programming model. The product ideal profile
obtained from consumers indicated the direction in which the product should be
improved and was used to generate upper and lower bound for the sensory constraints.
Physical attributes of commercial glue sticks were also used to generate upper and
lower bounds for physical constraints. Glue stick formulations were then derived from
either acceptability maximization or cost minimization.
It was found that in the initial model, there were too many constraints and some of the
sensory constraints were too tight and a feasible solution could not be obtained.
Adjustments were made to the constraints. Glue sticks using formulations from selected
feasible solutions were made and then tested by a laboratory sensory panel. The
formulation which obtained the highest acceptability was selected for a pilot scale
production.
7.1
AIM AND OBJECTIVES
The aim of this product optimization was to obtain a tapioca starch based glue stick
product which was highly acceptable to the target consumers at a low cost and could
compete w ith the products already in the market.
------ --
- --
1 46
,.. Use a ttribute/ ingredient relationships and the ideal product profile obtained
from the previous mixture design experiment to develop sensory constraints.
,.. Generate optimal formulations from the linear programming model by
acceptabili ty maxi mization or cost minimization.
,.. Select the optimum formulation obtained from the linear programming model
to be used in pilot scale production.
7.2
OPTIMIZATION PLANNING
During the optimization process, linear programming was used to generate the
optimum glue stick formulations. Formulations were selected and made in the
laboratory. Physical measurements were conducted on the samples. Selected samples
were then tested with the laboratory sensory panel . The sample which obtained highest
acceptability was chosen for further consumer testing. The optimization plan was as
follows:
,.. Develop a linear programming model from the empirical equations of

sensory attribute/ingredient relationships, acceptability /ingredient relationship,
and physical attribute/ingred ient relationships.
,.. Make the product prototypes from the optimum solutions obtained from
the linear programming model.
,.. Measure the physical attributes of the prototypes.
,.. Measure the sensory attributes of the selected prototypes using a laboratory
panel.
,.. Select the formulation which obtained highest acceptability for pilot scale
production and final consumer testing.
1 47
7.3
DEVELOPMENT OF LINEAR PROGRAMMING MODEL
Linear programming model comprised three major components; objective function,

decision variables and constraints.
7.3 . 1
Objective Functions
Although cost is the important factor in formulation of a commercial product, for a

consumer product, consumer acceptability is another important factor to be considered.
Therefore, in this study, the objective function of the l inear programming models was
either to minimize cost of the optimum glue stick formulation or to maximize consumer
acceptability. In the minimization of glue stick cost, costs of all ingredients were used
to create a cost equation which was used as an objective function to generate a low cost
formulation. In the maximization of consumer acceptability, the empirical equation
relating acceptability and ingredients was used as an objective function to generate a
product with high acceptability. The reason that these two objective functions were used
in the linear programming model was to compare the formulations and their product
a ttributes obtained from both cost minimiza tion and acceptability maximization then
the decision was made on which formulation to be used further for final consumer
testing.
The cost of each ingredient used in glue stick formulation was as shown in Table 7.1 .
These costs were ind ustrial prices obtained from the suppliers i n Bangkok.
Table 7.1
Costs of the ingredients used in glue stick formulation
Ingredients
Cost (Baht/kg)
25.00
495.00
Glycerin
50.00
Stearic acid
20.00
Sodium hydroxide
1 8.00
G lyceryl monostearate
1 20.00
1 48
The cost of each component was calcula ted based on the ratio and the costs of the
ingredients in each component. They are shown in Table 7.2. These component costs
were entered into the model as cost in Baht per kilogram.
Table 7.2
Costs of the components used in glue stick formulation
Components
Cost (Baht/kg)
Adhesive
260.00
Solvent
1 1 .65
Gel-forming agent
19.53
1 20.00
Since 3.5 grams of glyceryl monostearate was added into every 100 gram of the three
component mixture, so the objective was to minimize the cost of a 1 03.5-kg batch of the
glue stick product. Therefore, the objective function was to:
minimize
Z
260(Adhesive)
1 l .65(Solvent)
19.53(Gel)
1 20(GMS)
where the value of ingredients were the weights (in kilograms) of the components and
the coefficients were the per-unit costs. Water was assumed to have no cost.
Acceptabili ty equation was developed from the multiple regression of acceptability with
glue stick components (Table 6.14). The objective function was to:
maximize
Z
7.3.2
-0.156(Adhesive) + 0.293(Solvent) - 0.332(Gel-forming agent)
Decision Variables
The decision variables for the glue stick formulation were the 3 components used in the
previous mixture design experiment together with glyceryl monostearate (see Table 7.3.)
1 49
Table 7.3
Decision variables in glue stick formulation

Variables
Components
Adhesive
Xl
Gel-forming agent
7.3.3
Constraints
Constraints were set on the component level, batch size, sensory attributes and physical
attributes of the product. The constraints used in the glue stick linear programming
model are shown in Figure 7. 1 . The component constraints were included in order to
control the level of each component in the sui table range suggested by the previous
experiment. The physical constraints were used so that the physical attributes of the
optimum formulation would be in the ranges of the commercial products which were
assumed to be products acceptable to the consumers. The sensory constraints would
constrain the product to get closer to the consumers' ideal product.
l3atch size constraint
Constraint ( 1 ) was the batch size constraint. The batch size of 1 03.5 kg was fixed so that
the final values of the decision variables could be interpreted as either kilograms or
percent of total of three components, not including glyceryl monostearate.
X l + X2 + X3 +X4
103.5 kg
Component Constraints
Constraints (2) to (8) were glue stick component constraints. The upper and lower
bounds of glue stick components were derived from the component levels of the
previous mixture design experiment.
1 50
BATCH SIZE CONSTRA I NTS
TOTAL
(ADHESIVE+SOL VENT+GEL+GMS)
1 03.5
(1)
COMPONENT CONSTRA I NTS

A DH ESIVE
ADHESIVE 1 5
(2)
ADH ES I V E
A DH ESIVE 30
(3)
SOLVENT
SOLVENT 60
(4)
SOLVENT
SOLVENT 70
(5)
G E L-FORMING AGENT
GEL 1 0
(6)
GEL-FORMING AGENT
GEL 1 5
(7)
G LYCERYL MONOSTEARATE
GMS
SLIPPERINESS
-.3 1 S*ADHESIVE+.372*SOLVENT-.296*GEL 1 0.5
(9)
SLIPPERINESS
-.3 1 S*ADHESIVE+ .372*SOLVENT-.296*GEL 1 4.5
(10)
D ISINTEGRATION
.243*ADHESIVE 0
(11)
DISINTEGRATION
.243*ADHESIVE 4.0
( 1 2)
COVERAGE
. 1SI *SOLVENT 1 1
( 1 3)
COVERAGE
. 1 S l*SOLVENT 1 5
(14)
A DJUSTABILITY
-.OS2*ADHESIVE + . l 66*SOLVENT 9
(15)
A DJ USTABILITY
-.OS2*ADHESIVE+.1 66*SOLVENT 1 3
(16)
SMOOTHNESS
.1 67*SOLV ENT 1 1 .5
( 1 7)
SMOOTHNESS
. 1 67*SOLVENT 1 5
(18)
STICKABI LITY
.168*ADH ESIVE+ .086*SOLVENT 9.5
( 19)
STICKABI UTY
. 1 68*ADH ESIVE+.086*SOLVENT 1 3.5
(20)
ACCEPT ABILITY
-. I S6*ADHESIVE+ .293*SOLVENT-.332*GEL 1 3
(21 )
W ET GLUE/ AREA
2.66*ADHESIVE-.26S*SOLVENT 3 1
(22)
WET GLUE/ AREA
2.66*ADHESIVE-.26S*SOLVENT 46
(23)
DRY GLUE/A REA
1 .49PADH ESIVE-.233*SOLVENT 1 0
(24)
DRY GLUE/ A REA
1 .491 * ADHESrVE-.233*SOLVENT 1 5
(25)
MOISTURE CONTENT
-.02*A DHESIVE+ .882*SOLVENT+ . 1 2 1 "GEL S6
(26)
MOISTURE CONTENT
-.02*ADHESIVE+ .882*SOLVENT+ . 1 2 1 *GEL 68
(27)
H ARDN ESS
-.012*ADHESIVE+ .083*SOLVENT 2
(28)
HARDNESS
-.01 2*ADHESIVE+ .083*SOLVENT 6
(29)
OPEN TIME
.564* ADHESIV E-.00S*SOLVENT-.688*GE L 2
(30)
OPEN TIME
.564* ADH ESIVE-.00S*SOLVENT-.688*GEL 1 00
(31)
PEEL STRENGTH
.043* ADHESIVE+ .OSS*SOLVENTt- .04*GE L 3.6
(32)
PEEL STRENGTH
.043*ADHESIVE+ .OSS*SOLVENT+.04*GEL 6.3
(33)
3.5
(8)
P HYSICAL CONSTRA I NTS
Figure 7.1
Constraints for gl ue stick linear programming model
151
Sensory Constraints
Sensory constraints, constraints (9) to (20), were developed from the empirical equations
showing relationships between glue stick components and sensory attributes obtained
from regression analysis in Section 6.1 1 . In the cost minimization, acceptability
constraint, (21), could also be used in the model . The upper and lower bounds of the
sensory constraints were calculated by add ing and subtracting 2 from the mean product
ideal a ttribute scores obtained from consumer testing in Table 6.6. Those ranges were
used in order to allow enough space to obtain a feasible solution. However, these
bounds could be adjusted in order to get a feasible solution. The lower bound of
acceptability was set at 1 3 .
Physical constraints
Physical constraints, constraints (22) to (33), were developed from the empirical
equations showing relationships between glue stick components and physical attributes.
The upper and lower bounds of physical attributes were derived from the ranges of
commercial glue stick physical attributes, so that the physical attributes of the optimum
formulation would be in the ranges of commercial products.
7.4
DEVELOPMENT OF OPTIMUM FORMULATIONS FROM LINEAR

PROGRAMMING
There were two objective functions to be used in generating optimum formulation for
glue stick: cost minimization and acceptability maximization. The formulations were
developed from one of these two objective functions, together with sensory constraints
and / or physical constraints.
1 52
7.4. 1
Cost Minimization
Linear Programming Model ( 1 ) which was comprised of cost minimization objective

function, batch constraint (1) and the component constraints (2) to (8) was used as the
basic model for cost minimization. Model ( 1 ) is shown in Figure 7.2.
M IN I M IZE
Zc
260* ADHESIVE+ 1 1 .65*SOLVENT+ 1 9.53*GEL+ 1 20*GMS
SUBJECT TO
ADHESIVE+SOLVENT+GEL+GMS
1 03.5
(1)
A DH ESIVE 1 5
(2)
A DHESIVE $; 30
(3)
SOLVENT 60
(4)
SOLVENT $ 70
(5)
GEL 1 0
(6)
GEL $ 1 5
(7)
GMS
3.5
Figure 7.2
(8)
Model (1 ) for cost minimization
Minimize cost with sensory attribu tes (FI)

Sensory constraints (9) to (20) were added to Model ( 1 ) in order to generate glue stick
formulation with desired sensory attribute levels at the lowest cost. In the first running
of the model on the compu ter using LP88 computer programme, no feasible solution
was obtained. It was found that the sensory constraints were too tight. Therefore,
adjustments were made to these constraints until the feasible solution was obtained. The
constraints were changed as follows:
* Slipperiness constraint, from 1 0.5-1 4.5 to 1 0.5-15.0

* Disintegration constraint, from 0.0-4.0 to 0.0-4.5
* Coverage constraint, from 1 1 .0-15.0 to 9.0-1 5.0
* Smoothness constraint, from 1 1.5-15.0 to 1 0.5-15.0
* Stickability constraint, from 9.5-1 3.5 to 8.5-13.5
These adjusted constraints were also used in the other linear programming models.
1 53
M inimize cost with acceptability (F2)

Only acceptability constraint was added to Model ( 1 ) in order to obtain glue stick
formulation which had high acceptability at the lowest cost but did not take other
constraints of glue stick attributes into account.
Mini mize cost with sensory and acceptability (F3)

In this linear programming model, both sensory constraints, constraints (9)-(20), and
acceptability constraint, constraint (21) were added to Model (1). The model was run
on
the LP88 programme, but no feasible solution was obtained. The upper bound of
slipperiness had to be increased to 17 in order to get a feasible solution.
Minimize cost with physical attributes (F4)

Only physical attribute constraints, constraints (22)-(33), were added to model ( 1 ) in
order to obtain a glue stick formulation with desired attribute levels at a lowest cost.
Minimize cost with physical and sensory attributes (FS)

Physical constraints and sensory constraints were added to Model ( 1 ) so a glue stick
formulation with desired sensory and physical attribute levels could be obtained at a
lowest cost. The lower bound of open time had to be adjusted in order to get a feasible
solution.
Minimize cost with physical attributes and acceptability (F6)

PhYSical constraints and acceptability constraints were added to Model ( 1 ). From this
model a glue stick formulation with desired levels of acceptability and physical attribute
levels was obtained .
Minimize cost with physical and sensory attributes and acceptability (F7)
All gl ue stick attribute constraints were added to Model (1). In order to get a feasible
formulation, slipperiness upper bound constraint had to be increased to 16 and lower
bound constraint of open time had to be decreased to 1 .98.
The optimum formulations are shown in Table 7.4.
1 54
Table 7.4

obtained from cost minimization
F1
F2
F3
F4
F5
F6
F7
Ad hesive
17.73
1 5 .00
1 5.00
1 8.51
1 8.31
1 8.51
1 8.51
Solvent
67.27
70.00
70.00
68.73
66.69
68.73
68.70
Gel-forming agent
15.00
1 5.00
15.00
1 2.76
1 5.00
1 2.76
] 2.79
GMS
3.50
3.50
3.50
3.50
3.50
3.50
3.50
Slipperiness
1 5.0
(1 6.9)
] 6.9
(16.0)
1 4.6
(16.0)
1 6.0
4.3
(3.7)
3.7
(4.5)
4.5
(4.5)
4.5
Coverage
1 0.2
( 10.6)
1 0.6
(1 0.4)
10.]
( 10.4)
1 0.4
Adjustability
1 0.3
(1 0.8)
10.8
(1 0.5)
10.1
(10.5)
1 0.4
Smoothness
1 1 .2
( 1 1 .7)
1 1 .7
( 1 1 .5)
1 1 .1
( 1 1 .5)
1 1 .5
Stickability
8.8
(8.5)
8.5
(9.0)
8.8
(9.0)
9.0
Wet glue/area
(29.3)
(21 .3)
(21 . 3)
3 1 .0
31 .0
31 .0
31 .0
Dry glue/area
(1 0.8)
(6. 1 )
(6. 1 )
1 1 .6
1 1 .8
] 1 .6
1 1 .6
Moisture
(57.6)
(60.5)
(60.5)
58.4
56.9
58.4
58.4
Hardness
(5.37)
(5.63 )
(5.63)
5.48
5.32
5.48
5.48
Open time
(0.02)
(-1 .51)
(-1.51)
2.00
0.34
2.00
1 .98
Peel strength
(5.06)
(5.10)
(5. 1 0)
5.09
5.06
5.09
5.09
Costs (Baht)
6 1 06
5428
5428
6282
6249
6282
6281
Acceptability
( 1 2.0)
1 3.2
13.2
(1 3.0)
( 1 1 .7)
1 3.0
13.0
Disintegration
Note:
Number in the parentheses were the attributes which were not limited by the constraints
7.4.2
Acceptability Maximization
For product acceptability maximization, linear Programming Model (2) comprised of

acceptability maximization objective function, batch constraint ( 1 ) and the component
constraints (2) to (9) was used as the basic model. The model is shown in Figure 7.3.
--
- ------
1 55
MAXIMIZE
Z.
- 1 .1 56* ADHESIVE+.293*SOLVENT-.332*GEL
SUBJECT TO
ADH ESIVE+SOLVENT+GEL+GMS
1 03.5
(1 )
ADHESIVE 1 5
(2)
ADHESIVE :5 30
(3)
SOLVENT 60
(4)
SOL VENT :5 70
(5)
GEL ] 0
(6)
GEL :5 1 5
(7)
GMS
Figure 7.3
3.5
(8 )
Model (2) for acceptability maximization
Maximize acceptability (FS)

Model (2) was run on the computer without any other glue stick attribute constraints.
This was to get optimum glue stick formulation with highest expected acceptability.
Maximize acceptability with sensory attributes (F9. 1)

Sensory a ttribute constraints, constraint (1 0)-(20) were added to Model (2). Two other
formulations were derived from formulation F9. 1 by varying the solvent level in the
formula tion.
Formulation F9.2 was obtained by setting solvent constraint at 60-65%. In this model,
the disintegration upper bound constraint had to increased to 4.S6 in order to get a
feasible solution. Expected stickability was more than that of F9. 1 but d isintegration was
higher.
Formulation F9.3 was obtained by setting solvent at constraint 60-67% and

d isintegration upper bound constraint had to be increased to 5.6 so a feasible solution
could be obtained.
Maximize acceptability with physical attributes (Fl O)

Physical a ttribute constraints, constraints (22)-(33), were added to Model (2).
1 56
Maximize acceptability with sensory and physical attributes (FI l)

Both sensory attribute constraints and physical attribute constraints were added to
Model (2).
Optimum formulations obtained are shown in Table 7.5.

Table 7.5
Glue stick optimum formula tions and their expected attribute levels
obtained from acceptability maximization
FB
F9. 1
F9.2
F9.3
Fl O
F11
Ad hesive
20.00
1 8.52
20.00
23.00
20.00
20.00
Sulvcllt
70.00
67.29
6S.00
67.00
70.00
70.00
Gel-forming agent
1 0.00
1 4. 1 9
1 5.00
1 0.00
1 0.00
1 0.00
3.50
3.50
3.50
3.50
3.50
3.50
Slipperiness
( 16.8)
15.0
1 3.4
1 4.7
( 1 6.B)
1 6.8
Disintegrate
(4.9)
4.5
4.9
5.6
(4.9)
4.9
Coverage
( 1 0.8)
1 0.2
9.B
1 0. 1
( 10.6)
1 0.6
Adjustability
( 10.6)
1 0.2
9.B
9.9
(1 0.6)
1 0.6
Smoothness
( 1 1 .7)
1 1 .2
10.9
1 1 .2
( 1 1 .7)
1 1 .7
Stickability
(9.4)
8.9
9.0
9.6
(9.4)
9.4
Wet glue/area
(34.7)
(31. 1 )
(36.0)
(43.4)
34.7
34. 7
Dry glue/area
(1 3.5)
(11.9)
( 14.7)
(1 B.7)
1 3.5
1 3.5
Moisture
(62.6)
(57.3)
(55.1)
(55.7)
62.6
62.6
Hardness
(5.57)
(5.36)
(5. 1 6)
(5.2 9)
5.57
5.57
Open time
(4.05)
( 1 .02)
( 1 .29)
(6.43)
4.05
4.05
Peel strength
(5. 1 1 )
(5.06)
(5.04)
(5.07)
5.1 1
5.1 1
Costs (Baht)
6631
6296
6670
7376
6631
6631
Acceptabil i ty
1 4. 1
1 2. 1
1 1 .0
1 2.7
1 4. 1
14.]
GMS
Note:
Number in the parentheses were the attributes which were not limited by the constraints.
7.5
SELECTION
OF
FORMULATIONS
OBT AINED
FROM
LINEAR
PROGRAMMING
The feasible formulations obtained from linear programming were selected to be made
in the laboratory. Formulations F4, F6 and F7 were similar, so F7 which included all
constraints was chosen. Formulations Fl and F5 gave similar predicted sensory
1 57
attributes but some physical attributes of FI were out of the range of commercial glue
sticks hence, F5 was chosen. Formulations F2 and F3 had the same levels of components
in the formulations so F3 was chosen.
Since FB, FlO and F l l were the same formulation, Fl l was used. F9. 1 , F9.2 and F9.3
were selected since these formulation showed sensitivity to decreasing upper bound of
solvent constraint.
The formulations selected from linear programming and the predicted scores on product
attributes are shown in Table 7.6.
Table 7.6
Selected formulations obtained from linear programming

F3
F5
F7
F9.1
F9.2
F11
F9.3
Adhesive
1 5.00
1 8.31
1 8.51
1 8.52
20.00
20.00
23.00
Solvent
70.00
66.69
68.70
67.29
65.00
70.00
67.00
Gel-forming agent
1 5.00
1 5.0
1 2.79
14.19
1 5.00
10.00
1 0.00
Slipperiness
1 6.9
1 4.6
1 6.0
1 5.0
1 3.4
1 6.8
1 4.7
Disintegration
3.7
4.5
4.5
4.5
4.9
4.9
5.6
Coverage
10.6
1 0. 1
1 0.4
10.2
9.8
1 0.6
1 0.1
Adjustability
10.8
10.1
1 0.4
1 0.2
9.8
1 0.6
9.9
Smoothness
1 1 .7
1 1 .1
1 1 .5
1 1 .2
1 0.9
1 1 .7
1 1 .2
Stickability
8.5
8.8
9.0
8.9
9.0
9.4
9.6
Wet glue/area
2 1 .3
3 1 .0
31.0
3l.1
36.0
34.7
43.4
Dry glue/area
6.1
1 1 .8
1 1 .6
1 1 .9
1 4.7
1 3.5
1 8.7
Moisture
60.5
56.9
58.4
57.3
55. 1
62.6
55.7
H ardness
5.63
5.32
5.48
5.36
5.16
5.57
5.29
Open time
- 1 .51
0.34
1 .98
1 .02
1 .29
4.05
6.43
Peel strength
5.10
5.06
5.09
5.06
5.04
5.1 1
5.07
Costs (Baht)
5428
6249
6282
6296
6670
6631
7376
Acceptabili ty
1 3.2
1 1 .7
1 3.0
1 2. 1
1 1 .0
14.1
1 2.7
Figure 7.4 shows the positions of these formulations on the mixture space. The glue
sticks were made from these optimum formulations using the processing method
described in Section 3.5 . The samples were aged for at least 5 days before the testing
was conducted.
1 58
60% Solv e n t
7 0 % Solv e n t
1 5% G e l
1 0% G e l
1 5% Adhesive
30% Adhesive
o
[2d
Figure 7.4
7.6
Area i n t h e p re v i o u s exper i m e n t
An e x t e n t i o n
Positions of the optimum formulations on the mixture space
PHYSICAL ATTRIBUTES OF GLUE STICKS DEVELOPED FROM LINEAR

PROGRAMMING MODELS
Physical attributes of the optimum glue sticks were measured using the same methods
as described in Section 3.3.
Table 7.7
Physical attributes of the glue sticks developed from linear programming

models
Samples
Wet glue/area
Dry glue/area
(g/m2)
F3
Hardness
(Newton)
Open time
(minute)
(g/ m2)
Moisture
content
(%)
Peel
strength
(Newton)
26.4
9.8
58.9
3.86
0.00
4.05
F5
32.0
1 2.7
55.4
3.58
1 .50
4.78
F7
35.9
13.7
56.8
4.01
2.50
4.75
F9
34.9
13.1
55.6
3.62
2.00
4.60
F9.2
31 . 1
1 1 .1
59.9
4.63
0.00
4.32
F9.3
31 .5
1 2.2
57.2
4.01
3.00
4.81
F1 1
36.5
1 3.Y
57.5
3 .96
1 .50
4.52
1 59
Apart from open time, the physical attributes of most samples from the formulations
developed from linear programming models were in the commercial ranges. Only the
sample from formulations F3 and F9.2 had lower wet glue and dry glue per area than
the commercial range. The open time of all the samples were much lower than those
of the commercial glue sticks. Samples from formulations FS, F7, F9, F9.3 and F l l had
open time between 1 .50-3.00. Formulation F3 and F9.2 had no open time.
The samples were selected for sensory testing by the laboratory sensory panel. Since
formulation F3 and F9.2 had low amounts of glue residue left on paper and their open
time was 0.00, they were not tested any further.
7.7
SENSORY TESTING
7.7.1
Selection of Samples for Sensory Testing
Formula FS, F7, F9, F9.3 and Fll from linear programming were chosen to be tested
by a laboratory panel. These samples were presented to the panelists together with the
Runs 4 and 5 from the second mixture design experiment (Section 6.2), in order that the
results between optimum formulation from linear programming could be compared
with the best glue stick samples from the previous mixture design experiment.
7.7.2
Sample Preparation
The samples were allowed to age for at least 5 days before the test was conducted. They
were coded with 3 random digit numbers.
7.7.3
Method of Sensory Testing
Sensory evalua tion was carried out by 1 0 panelists: 8 post graduate students and 2 non
academic staff at the Product Development Department, Kasetsart University, all of
whom had already participated in the first consumer testing. This panel will be
mentioned as 'pseudo-consumer panel' subsequently. Samples were pesented
simultaneously to the panelists. The questionnaire used in this test was in Thai, the
1 60
same as the one used in consumer testing described in Section 6.4.3 and the method of
testing was as in Section 6.4.5.
7.7.4
Sensory Attributes
Mean scores of the sensory attributes of the glue stick samples were calculated, the
results are shown in Table 7.8.
Table 7.8
Sensory a ttributes of the glue sticks developed from linear programming .

model (mean scores)
Samples
Accept
Slip
Disintegr.
F9.3
1 1 .0'
1 1 .8a
3A
F7
i1b
1 0.2
10.9-b
Fl l
9.9'b
F9.l
F5
8A
RUN4
1 0.0-b
RUN5
9 .3 be
Note:
Coverage
Adjust
Smooth
Stick
l 1 .1 a
l O.8a
1 2 5a
1 1 .2ab
5. 9-be
10.9-'
1 0.8"
1 1 .6ab
1 2.0" b
l 1 .3,b
6.l ,b
l OA'
8.8'
1 2. O"b
1 1 .8,b
9.7ab
l OA b
e
5.7ab
1 0.0"
9.7"
1 1 .2 be
9.7
1 1 .2ab
7A'
9.9'
1 0.3'
1 0.6
4. 1
he
1 0.3a
10.0"
12.6'
10.2 be
,\
4.2 be
10.3"
9.4"
1 2.2'
1 2 .2'
9.1
7.8
12.1'
Mean scores within the column followed by a different letter are Significantly different a t p
using t-test.
<
0.05,
The sample from formu lation F9.3 obtained highest acceptability and lowest
disintegration. It also scored higher for acceptabili ty than Runs 4 and 5, the best
samples from the previous experiment. In general, F9.3 had better sensory attributes
than Run 4 and 5. Degree of coverage and adjustability were not significantly different
between samples.
Ideal ratio scores of samples were calculated by dividing the sample scores by the mean
ideal scores obtained from consumer testing in Table 6.6. The results are shown in Table
7.9.
161
Sensory attribute ideal ratio scores of the glue sticks developed from
Table 7.9
linear programming
Samples
Slip
Disintegrate
Coverage
Adjust
Smooth
Stick
Total Difference
from 1 .0
F9.3
0.95
1 .85
0.86
0.97
0.93
0.98
1 .1 6
F7
0.88
3.23
0.85
0.97
0.86
1 .05
2.72
Fl l
0.91
3.32
0.81
0.79
0.89
1 .03
2.95
F9.1
0.84
3.14
0.77
0.88
0.83
0.85
2.97
F5
0.90
4.07
0.77
0.93
0.79
1 .06
3.74
RUN4
0.74
2.25
0.80
0.90
0.93
0.90
1 .98
RUN5
0.63
2.31
0.80
0.85
0.91
1 .07
2.19
Note:
Total difference from 1 was calculated from sum of the difference of ideal ratio scores from 1 .
From the mean ideal scores, sample F9.3 was closest to the consumers' ideal glue stick.
Hence, formulation F9.3 was chosen for pilot plant scale production. F7 was not chosen
even though some of i ts attributes were close to the ideal product and the formulation
cost was lower than that of F9.3 because the disintegration was much higher and other
sensory attributes were slightly lower.
7.7.5
Opti mum Formulation
The profile of the optimum formulation (F9.3) is shown in Figure 7.5 in comparison
w ith the consumers' ideal product profile. The slipperiness, adjustabil i ty, smoothness
and stickability were quite close to the ideal product. Disintegration was higher and
degree of coverage was lower.
1 62
Attrlbut..
SlIpperlneaa
.....--O-----.---T---...--r--..,."o---r--,
Dlalntegratlon
Coverage
AdJuatablllty
Smoothneaa
10
12
14
Seneory Soor..
-- FG.3
Figure 7.5
--
Ideal
product
Product profile of optimum formulation compared with

ideal product profile
1 63
7.8
DISCUSSION
7.8.1
Comparison of Optimum Formulation with Commercial Products and

Consumers' Ideal Product
Table 7.10 shows the attributes of the optimum formulation compared with attributes
of commercial glue sticks and ideal product.
Table 7. 10
Attributes of optimum glue stick compared with commercial products

and ideal product
Attributes
Optimum Product
Commercial Ranges
Ideal Product
Slipperiness
1 1 .8
5.9-9.6
1 2.4
Disintegration
3.4
3.I:H!.1
1 .8
Coverage
1 1 .1
1 0.2-1 1 .4
1 2.9
Adjustability
10.8
8.5-1 0.4
1 1 .1
Smoothness
1 2.5
10.5- 1 2.2
13.5
Stickability
1 1 .2
9.4-10.7
1 1 .4
Wet glue/area
3 1 .5
31 .9-45.8
Dry glue/area
1 2.2
1 0.3-1 4.9
Moisture
57.2
57.0-65.3
H a rdness
4.01
2.42-3.55
Open time
3.00
12.3-99.0
Peel strength
4.81
3.65-6.30
Sensory attribute
Physical attributes
The sensory attributes of optimum product were quite close to those of the ideal
product. Disintegration of the optimum product was higher than the ideal but still
lower than the commercial product. Other sensory attributes of the optimum product
were slightly higher than the commercial ranges. Physical attributes were in the range
of commercial products except open time which was much lower.
1 64
7.8.2
Comparison between Cost minimization and Acceptability Maximization
The cost and sensory attributes of formulation from cost minimization and acceptability
maximization are compared i n Table 7.1 1
Table 7.1 1
Comparison between using cost minimization and
acceptability
maximization
Cost Minimization
Acceptability Maximization
5,428-6,282
6,296-7,376
Acceptability
8.4-10.2
9.7-1 1 .0
Slipperiness
1 0.9- 1 1 .2
10.4-1 1 .8
Disintegration
5.9-7.4
3.4-6 . 1
Coverage
9.9-10.9
10.0-1 1 . 1
Adjustability
10.3-10.8
8.8-10.8
Smoothness
1 0.6-1 1 .6
1 1 .2-12.5
Stickability
1 2.0-1 2 . 1
9.7- 1 1 .2
Wet glue/area
26.4-35.9
3 1 . 1 -36.5
Dry glue/area
9.8-13.7
1 1 .1 -1 3.9
Moisture
55.4-58.9
55.6-59.9
Hardness
3.58-4.01
3.62-4.63
Open time
0-2.50
0.00-3.00
Peel Strength
4.05-4.78
4.32-4.81
Cost (Bahts/ 103.5 kg)

Sensory attributes
P hysical attributes
With acceptability maximization the cost of product formulations were higher than
those from cost minimization. This was because there was high amount of adhesive
substance in the formulations obtained from acceptabil ity maximization. The
acceptability of the formulation obtained from acceptability maximization were higher,
sensory attributes were similar. Most physical attributes from acceptability maximization
were higher.
1 65
7.8.3
Use of Linear Programming in Glue Stick Optimization
Linear programming was found useful for the formulation of a tapioca based glue stick
with sensory constraints, cost limitation and high acceptability. It was conclusively
shown that the relationships between sensory attributes and glue stick ingredients
obtained from the consumer panel could be used in developing a linear programming .
model. Mean ideal scores was also useful for guiding the upper and lower bounds of
sensory a ttributes.
Although sensory attribute constraints had already been included in the l inear
programming model, physical attribute constraints were included as well in order to
control the physical properties of glue stick at the levels of commercial glue sticks.
Kavanagh ( 1978) successfully used physical attributes of commercial products as
standards in the optimization of paint and resin formulation. Chan and Kavanagh
(1988) also used a similar method in the formulation of l ight duty liquid detergent.
With using cost as an objective constraint, it was possible to obtain a product with high
acceptability. However, when acceptability was used, the formulations gave product
with higher acceptability but a higher cost. So it should be considered carefully which
objective function is to be used. Although the formulation chosen to be made in the
pilot scale did not give the lowest formulation cost, its sensory attributes were
considered close to the ideal product and the acceptability was highest. This glue stick
had potential to be accepted by the target consumers.
The advantage of using this method is that it allows the researcher with no experience
with the product to reach an acceptable formulation in a finite number of steps.
H owever, the disadvantage is the need for linear relationships between product
attributes and ingredients in the formulation therefore the ingredients used in the
experimentation have to be limited in a narrow compositional ranges.
1 66
7.9
CONCLUSIONS
The l inear programming model using the linear relationships between sensory attributes
and ingredients as constraints of the model was used successfully to optimize the
formulation during the product optimization process in order to obtain a highly
acceptable glue stick to consumers. Since the consumers' ideal product profile was used
to create upper and lower bounds in the linear programming model, it was possible to
generate an optimum product which possessed the attribute levels near those of the
consumers
'
ideal product.
It is suggested that this system could be used successfully for any product, in which the
relationships between the product's attributes and the ingredients are l inear. The
constraints to be used in the model could include component levels, sensory attributes
as an ideal product profile and physical properties specification. The minimum number
of constraints should be used so that a feasible solution is obtained. If the relationships
between sensory and physical attributes are correlated. There might be no need for
sensory constraints. Product optimization using linear programming as a tool to
optimize sensory attribute levels is an excellent method to obtain the optimum product
if the sensory / ingredient relations are linear and the interaction between ingredients are
negligible.
The use of linear programming was found suitable for glue stick formulations if it was
combined with mixture experimental design. The proportions of the main ingredients
in the glue system were interrelated and factorial designs coul d not be used. With linear
programming, the optimum formula tion can be obtained easily but with other methods
such as response surface methodology it is more difficult to interpret the results.
However, with the use of l inear models, the range of ingredients which could be used
are limited since moving outside the specified range the relationship may not be l inear.
The final product possessed attributes close to the consumers' ideal product in terms
of slipperiness, adjustability, smoothness, and stickability. Degree of coverage was
slightly lower and disintegration was higher than those of the ideal product. Most of
physical a ttributes were in the ranges of the commercial products except open time
which w as lower than the commercial products. However, the recommended open time
1 67
was a t least one minute so this was considered acceptable.
The glue stick optimum formulation obtained from linear programming model and i ts
cost are shown below:
%
Starch
1 1.11
1 1.11
Glycerin
14.93
Water
49.81
Stearic acid
7.39
Sodium hydroxide
2.27
3.38
100.00
Cost(Baht/ 1 00 kg)
Cost per 8 g stick (Baht)
7126.43
0.57
(estimate at no loss in the processing)
The formulation from the linear programming model which obtained highest consumer
acceptability was chosen to be made in pilot plant scale. This optimum glue stick
product obtained had potential to be accepted by the target consumers. However, this
product had to be tested for acceptability in home-use testing before launching to the
market to ascertain that it could compete with the products already in the market.
The processing method used in the laboratory scale was to be applied and used in the
pilot scale production of glue stick. Pilot scale equipment was to be developed with the
same principle as the laboratory equipment.
1 68
CHAPTER 8
PRODUCTION TRIAL
The optimum formulation from the previous study was used to make glue stick in the
pilot plant development stage. At this stage, the product was made a t the pilot plant
level which was the smal l scale of the commercial manufacturing. The product was then
tested against the product from the same formulation in the previous s tudy, and a
commercial product. The effect of storage on the product attributes was also s tudied.
8.1
AIM AND OBJECTIVES
In the previous section, an optimum glue stick was made on a laboratory scale. It was
expected that when the glue stick was produced commercially, it should have the same
properties as those of the one developed in the laboratory scale. Therefore, the aim of
this production trial study was to confirm that the formulation and process developed
in the laboratory could be transferred into industrial production, and that the product
would have the same acceptability to consumers as the product from the laboratory
experiments.

* Design a suitable system to convert the raw materials into a product with the
specified properties.
* Compare glue stick from the pilot scale with the laboratory scale and a
commercial product.
* Generate samples for final consumer testing.
169
Figure 8.1
Pilot scale equipment
1 70
8.2
PILOT SCALE PRODUCTION
8.2.1
Equipment
The equipment used in the pilot scale production is shown in Figure 8 . 1 . A mechanical
stirrer was developed at the Department of Product Development, Kasetsart University.
The stirrer was rotated by a 1 /4 HP motor. The speed was controlled by a gear box and
this could be turned on and off by a switch. The stainless steel stirrer was a double
paddle form, with rectangular blades (size 4 cm x 1 2.5 cm). The top blade and bottom
blade were set 3.5 cm apart, so that it could stir and sweep the glue mixture from the
bottom and the side of the pot. The stirrer was attached to an adjustable axle so that it
could be lifted up or lowered down as desired.
A 32 cm diameter and 20 cm deep stainless steel pot was used. The pot was covered
with a lid which had one hole in the middle for the stirrer and one hole at 2 cm from
the edge for the condenser. A spiral glass condenser was used to provide a reflux
system for the process with 17 C input water from the cooling system. A rubber gasket
was attached around the edge of the lid in order to prevent any leaking of the steam.
The temperature of the mixture during processing was measured using a thermometer.
The pot was placed in a water bath which was electrically heated.
8.2.2
Formulation
The formulation which was obtained from the product optimization was used. The
product was made in 621 0 g batches. Table 8.1 shows percentage and amount of each
ingredient used in the production.
171
Table 8 . 1
Levels o f glue stick ingredients used in the pilot scale production
Ingredient
Percentage
Quantity (grams)
1 1 .1 1
690
11.11
690
Water
49.81
3093
Glycerin
14.93
927
Stearic acid
7.39
459
Sodium hydroxide
2.27
141
3.38
210
1 00.00
621 0
Total
The ingredients used in this stage were the same as those used in the laboratory scale
except the sodium hydroxide was industrial grade instead of laboratory grade.
8.2.3
Processing Method
The processing method was similar to that employed in the laboratory scale. The
cooking time was extenqed from 60 minutes to 90 minutes in order to allow for the
heating time necessary for the large amount of raw materials.
The liquid ingredients were put in the stainless steel pot. Then the modified starch was
grad w i lly stirred into the liquid, followed by polyvinyl pyrrolidone, stearic acid,
glyceryl monostearate and sodium hydroxide. The mixture was stirred manually until
it became a homogeneous mixture. The pot was then put in the water bath at 90 C and
covered with the lid which had the stirrer attached to it. Then the stirrer was connected
to the motor, and the mixture was stirred at 15 rpm. At this stage the temperature of
the water in the water bath was increased up to about 100 C. The mixture was then
heated for a further 90 minutes.
172
The pot was taken out of the water bath and the hot mixture was poured into a 48 cm
x 38.5 cm x 5 cm stainless steel tray. This tray was placed in a bigger tray which
contained hot water, about 90c, so that the mixture temperature would not drop too
quickly. The finished glue stick mixture was allowed to cooled down at room
temperature for 24 hours and then manually cut into sticks using a 1 .5 cm dia. x 12 cm
height PVC plastic cylinder. Each stick was then put into a lip stick type plastic
container normally used for 8 gram commercial glue stick.
8.2.4
Testing of the Finished Product
Physical and sensory testing was conducted on the finished product. Glue sticks were
randomly sampled for the test. For sensory testing, the finished product was presented
to the panel together with the laboratory scale sample. The same laboratory sensory
panel, comprised of 10 panelists, which did the product testing in the final product
optimization was used in this study. The questionnaire used for the test was the same
as that described in Section 6.4.3 and the method of testing was as in Section 6.4.5.
The glue sticks were tested at 3, 5, 1 0, 1 7 and 30 days after the process finished - from
the time that the glue mixture was poured into the tray. This was done in order to see
how much the glue stick changed with time in terms of sensory and physical a ttributes
of glue sticks and to consider whether the age of glue stick would affect the
acceptability of the glue stick to the target consumers in the home-use test.
8.3
YIELD AND COSTS IN GLUE STICK PRODUCTION
Yield of the glue stick made in the pilot scale was investigated in order that the costing
of glue stick production could be established. Glue stick was made in a 6.21 kg batch,
which gave 5.25 kg finished glue stick mixture, i.e. a 84.5% yield. It might be expected
that the losses would be less in full sca le production as most was lost because of
sticking to the pan.
1 73
It was assumed that if all of the finished glue stick mixture could be made into 8 g
s ticks, 1 0,562 sticks could be obtained from 100 kg batch. Hence, the raw materials cost
for an 8 g glue stick could be calculated as follows:
Cost per 100 kg of ingredients
7126
Bahts
Hence, cost per stick
0.67
Bahts
Cost for glue stick production was calculated from raw material cost, processing cost,
factory overhead, company profit ( 10%), whole sale mark up (5%) and retail mark up
( 15%). The pricing was very approximate but i t was predicted as followed:
,.. Processing costs
- labour
(Bah t / 100kg of finished product) - operating
1500.00
5000.00
Total processing cost
6500.00
Processing cost of one glue stick (Baht)
0.52
,.. Factory overhead was estimated as 3.35 Bahts per stick

,.. Container for 8 g stick was estimated as 3.50 Bahts
The results shown in Table 8.2 are the cost in producing an 8 g glue stick at different
percent weight losses.
The predicted retail price of the new glue stick was 10.70 baht which was lower than
the average sale price of commercial glue stick, about 15 baht. This indicated that the
new glue stick could be priced competitively with the products already in the market.
174
Table 8.2
Costs incurred i n producing an 8 g glue stick

Value (Bahtl stick)
Type of Cost
15.5% loss
1 0% loss
5% loss
Raw material costs
0.67
0.63
0.60
Processing costs (1)
0.52
0.52
0.52
Packaging costs (8 g container)
3.50
3.50
3.50
Factory overheads (2)
3.35
3.35
3.35
Total production cost
8.04
8.00
7.97
Company profit
0.81
0.80
0.80
Company price
8.85
8.80
8.77
Wholesale price
9.30
9.24
9.20
1 0.70
1 0.63
1 0.58
Retail price
Note: ( 1 ) This includes all labour and operating costs.

(2) This includes foremens' wages, office staff salaries, rent repair and
maintenance, plant depreciation and other miscellaneous cost plus sales force
salaries, administration cost and distribution costs.
8.4
CHANGES OF PRODUCT ATTRIBUTES DURING 30 DAY STORAGE
The results from the sensory and physical testing during storage are shown in Tables
8.3 and 8.4.
1 75
Table 8.3
Days
Physical attributes of glue stick at di fferent storage times

Wet glue/area
Dry glue/area
(g/ m2)
(g/m2)
33.4
1 3.3
32.7
10
Hardness
Open time
(Newtons)
(Minutes)
56.9
3.91
3.1
13.1
56.9
4.04
3.0
32.6
13.0
57.0
4.16
3.1
17
32.4
1 2.7
57.6
4.12
2.8
30
32.4
12.7
57.5
4.16
2.8
Table 8.4
Moisture
(%)
Sensory attributes of glue sticks at different storage time

Accept
Slip
Disintegrate
Coverage
Adjust
Smooth
Stick
10.8
1 1 .8
2.2
1 1 .6
1 1 .0
1 3.0
1 2.2
1 1 .9
1 2.0
3.3
1 1 .3
1 1 .6
1 2.3
10.7
10
1 1 .3
1 1 .3
3.8
1 1 .8
1 1 .9
1 2.5
1 2.2
17
1 1 .5
1 1 .8
2.9
1 2.2
1 2.7
1 2.5
1 2.5
30
1 1 .6
1 2.2
3.8
1 2. 1
1 1 .6
1 2.5
1 1 .5
Days
There was some small changes of physical attributes of glue stick after processing. There
appeared to be some trends: the stick tended to harden, open time tended to decrease
a little bit. However, these changes were not statistically significant. It was also found
.
that the sensory attributes of the stored glue stick samples were not significantly
different, only for stickability was the sample stored for 5 days slightly lower from the
others, this might have resulted from differences between panel evaluations or some
variation in the sticks.
This indicates that the age of glue stick within 30 days after processing would not affect
the attributes of the glue stick and consequently should not affect the perception of
consumers toward the glue stick in the home-use tested.
1 76
8.5
PRODUCT VARIABILITY BETWEEN BATCHES
The variability of the product from different batches was investigated i n order to
i nvestigate the reproducibility of the processing method and the equipment developed
for the pilot scale production. Product made from two different batches were tested and
the results are shown in Table 8.5 and 8.6.
Table 8.5
Physical attributes of glue sticks from d ifferent batches

Batch 1
Batch 2
T-tests
Wet glue per area
33.4
33.2
0.44
0.698
Dry gl ue per area
13.3
1 3.4
0.51
0.659
Moisture
56.9
57.2
0.69
0.614
Attributes
Hardness
3.93
3.97
0.27
0.793
Open time
3.13
3.00
1 .00
0.423
Table 8.6
Attributes
Sensory attributes of glue sticks from different batches

T-tests
Batch 1
Batch 2
Acceptability
1 1 .6
12.0
1 .05
0.309
Slipperiness
1 2.2
1 2.5
1 .3 1
0.213
3.8
2.1
2.09
0.056
Coverage
12.1
] 2.4
0.91
0.378
Adjustability
1 1 .6
1 2.3
1.12
0.280
Smoothness
12.5
1 1 .7
1 .56
0.141
Stickability
1 1 .5
1 1 .4
0.39
0.700
Disintegration
Note:
Samples from Batch 2 were tested against 30 day samples from Batch 1
From the t-tests, there were no significant differences between physical and sensory
attribute of glue stick samples from the two batches.
1 77
B.6
PRODUCT VARIABILITY BETWEEN STICKS FROM DIFFERENT LOCATIONS

ON THE TRAY
To ensure that the location of the glue stick in the rectangular tray had no effect on i ts
properties, various glue sticks were sampled from different locations on the tray: middle
and the two extreme sides.
The results are shown in Table B.7 and Table B.B.
Table B.7
Physical attributes of glue sticks from different locations on the tray

Wet glue/area
Dry glue/area
Moisture
Hardness
Open time
(g/m2)
(g/m2)
(%)
(Newtons)
(Minutes)
33.4
13.3
56.9
4.03
3.1
32.7
13.1
56.9
4.04
3.0
32.6
1 3.0
57.0
4.1 1
3.1
Location
Note:
Table B.B
Width side of the tray
Length side of the tray
In the middle of the tray
Sensory attributes of glue sticks from different locations on the tray
Position
Accept
Slip
Disintegrate
Coverage
Adjust
Smooth
Stick
1 2.0
12.5
2.1
1 2.5
1 2.3
1 3.0
1 1 .4
1 1 .8
1 2.5
2.1
1 2.5
1 2.5
1 3.0
1 1 .4
1 1 .8
] 2.6
2.1
1 2.5
1 2.4
1 3.0
1 1 .5
There were no significant d ifferences between physical and sensory attributes of glue
stick samples from different location on the tray, although the wet glue per area were
slightly d ifferent. This indicated that the glue stick mixture was homogeneous.
-----
---_._-
178
8.7
PRODUCT VARIABILITY BETWEEN LABORATORY SCALE AND PILOT

SCALE SAMPLES
The glue sticks made in the pilot scale were compared with the ones made in the
laboratory scale in order to confirm that the process and equipment used in pilot scale
production could produce glue sticks which had the same attributes' intensities as those
from the laboratory. The results of the physical and sensory testing of samples made
from two pilot scale batches were compared.
Table 8.9
Physical attributes of pilot scale and laboratory scale samples
Attributes
Pilot scale
Laboratory scale
T-tests
Wet glue per area (g/m2 )
33.3
31 .5
1 .62
0.248
Dry glue per area (g/m2)
13.3
13.2
1 .95
0.192
Moisture content (%)
56.9
57.2
0.72
0.601
Hardness (Newtons)
3.93
4.01
0.28
0.800
Open time (Minutes)
3. 1 3
3.00
1 .00
0.423
Note: The results of 3 day laboratory sample were compared with the results of 3 day
sample from Batch 1
The results in Table 8.9 show that the physical attributes of the sample from pilot scale
were not significantly different from the one produced on the laboratory scale.
179
Table 8.10
Attributes
Sensory attributes of pilot scale and laboratory scale samples

Pilot scale
Laboratory Scale
T-tests
Acceptabili ty
1 1 .4
1 1 .9
4.30
0.023
Slipperiness
1 1 .7
1 2.0
2.77
0.069
3. 1
2.6
0.97
0.402
Coverage
1 1 .6
1 2.0
1 .93
0 . 1 49
Adjustability
1 1 .8
1 1 .7
0.59
0.597
Smoothness
1 2.5
1 2.8
2.00
0.140
Stickability
1 1 .9
1 1 .8
0.49
0.660
Disintegration
The glue stick made in the pilot scale production had slightly lower acceptability,
slipperiness, degree of coverage and smoothness scores than those of the ones made in
the laboratory scale. Stickability, adjustability scores were slightly higher.
From the Hest, it was found that there was a significant difference between the
laboratory scale and pilot scale samples only for acceptability at p
0.023. There were
no significant differences in terms of sensory attributes. This indicated tha t the glue
sticks made from the pilot scale and laboratory scale were only slightly different.
8.8
COMPARISON OF N EW GLU E STICK W ITH A COMM ERCIAL GLUE STICK
A sample from the pilot scale production was also tested against the commercial
product to identified how di fferent the new product was from the commercial one.
UHU glue stick was used in this study since it was the glue stick most consumers used
according to the consumer survey.
1 80
Table 8 . 1 1
Sensory attributes o f new glue stick and a commercial glue stick
Attributes
New glue stick
Commercial glue stick
T-tests
Acceptability
1 2.0
1 2.3
0.84
0.416
Slipperiness
1 2.5
] 2.9
1 .90
0.078
2.1
2.1
0.02
0.988
Coverage
12.4
] 2.5
0.56
0.587
Adjustability
12.3
1 1 .6
1 .1 8
0.257
Smoothness
1 1 .7
1 3.2
1 .07
0.303
Stickability
1 1 .4
1 2.6
3.31
0.005
Disintegration
Most of the physical attributes of the new glue stick were similar to those of 'UHU' glue
stick except the open time. Open time of the new glue stick was very much lower than
the commercial one. This indicated that the new glue stick should have less drying time.
However 3 minute open time should be enough for readjusting the paper when
consumers used glue stick.
Although the new glue stick, compared with UHU glue stick, had slightly lower
sensory scores for acceptability and some attributes; from the analysis of variance, there
were no significant di fferences between the new glue stick sample and commercial glue
stick (UHU). With the exception of stickability, it can be said that this new glue stick
had a ttributes which were comparable to that of the commercial glue stick.
8.9
DISCUSSION
The process which was used to make the glue stick on a small scale was used
successfully in the production of pilot scale samples. It was expected that similar
equipment could be used i n a commercial process for the manufacture of glue stick.
However, suitable packaging equipment should be developed so that the glue stick
could be a u tomatically mounted to the container.
The yield of the product from the process can be improved in the commercial scale
production in whicl a double jacketed kettle could be used and the finished product
- -
--
----
181
could be discharged from the bottom so the temperature of the product in the container
could be maintained until totally discharged .
Although i t was found that the attributes of glue stick stored for 30 days after processed
were not significantly changed, the shelf life of the glue stick needs to be tested for a
longer period for production and selling because it may remain on the retailers' shelves
for 6-12 months.
The new product made in the pilot scale had the same properties as the one from the
laboratory scale. The sensory attributes of these two products were not significantly
d ifferent although the acceptability of the glue stick made in the pilot plant was very
slightly lower. It was found that the new product was different from the most popular
commercial product only in terms of open time. However in the sensory testing the new
product was not significantly different from the commercial product in acceptability and
product attributes.
8.10
CONCLUSIONS
From this study, it can be concluded that there was no change in the physical and
sensory attributes of the glue stick in the scaling up from the laboratory scale to the
pilot scale production as they had similar properties.
From a comparison of the physical and sensory attributes, it was shown that the glue
stick that was developed in this study possessed attributes which were comparable to
those of the most popular commercial glue sticks so it was expected the new glue stick
would be accepted by the target consumers.
Based on the above results, glue stick samples from this pilot scale production would
be used in the final consumer testing to measure consumer acceptability of the new
product. The attributes that needed further improvement would also be identified.
182
CHAPTER 9
FINAL CONSUMER TESTING OF PRODUCT
In the previous chapter, glue stick was made in the pilot scale production and tested
with the small consumer panel. The results showed that the product was acceptable and
the product a ttributes were comparable to a commercial glue stick. Therefore, it was
decided that the product should be tested with the target consumers to see how
consumers reacted to the product and how they compared the new product w ith the
one they currently used. The product was tested with four groups of consumers:
university students, school students, office workers in government offices and in private
offices.
9.1
AIM AND OBJECTIVES
The aim of this study was to measure product acceptability of the new glue stick.

* Establish the performance of the developed product in relation to competitive
products.
* Determine consumer purchase intention towards the developed product.
* Determine the potential target market for the developed product.
* Obtain information from consumers on how the developed product should be
launched into the market.
183
9.2
METHOD OF CONSUMER TESTING
9.2.1
Selection of Consumers
Consumers used for the final glue stick testing were selected from glue stick users. who
used glue stick at least once a month, so that the new product could be compared with
the product consumers currently used. Four groups of consumer were selected. They
were 2 groups of students: university student (60), school student (60) and 2 groups of
office workers: government officers (35), private officers (35). These people were
considered as target consumers for the developed product. The total number of
consumers was 190.
The university students from Kasetsart University and the school students from
'
Kasetsart Demonstration School were selected because of the convenience to ask for
permission and conduct the test.
The government officer and private officer were chosen since they were glue stick users
and most of them used glue stick.
9.2.2
Sample Preparation
Glue stick samples were packed in a commercial glue stick container. No brand name
or information about the glue stick was printed on the container. Samples were aged
for at least 7 days before the test was conducted.
9.2.3
Questionnaire
The questionnaire was written in Thai. The questionnaire in both the Thai and English
languages is shown in Appendix 9. 1 . The questionnaire had both check list and open
ended questions relating to the new glue stick as well as a section with demographic
details of the consumers.
The questionnaire was kept as short as possible. Easy words were used in order not to
cause any misunderstanding or confusion. The questionnaire was pre-tested with 1 6
1 84
consumers, 4 from each group of consumers and the questionnaire was changed where
necessary to make it easier for the consumer to answer.
9.2.4
Testing of the Product
The questionnaire together with the glue stick were put into a brown envelope and
given to the respondents themselves or to the representatives of the group of
consumers. The respondents were allowed about one week to use the glue stick sample
in the same way as they normally used glue stick. They could use the product at home
or at school for the students and in the office for the office workers. After they had tried
the sample for at least 2-3 times then they answered the questions in the questionnaire.
9.3
The questionnaires were checked by the author before processing was conducted. The
data were coded and en tered in the VP Planner spreadsheet. The entered data were
then analyzed using the SPSS / PC+ programme where cross-tabulation and chi-square
analysis were conducted. The results are shown in Appendix 9.2.
9.4
PATTERN OF GLUE STICK USAGE OF THE CONSUMERS
Sixty questionnaires were given to each group of students and 35 questionnaires were
given to each group of office workers. There were 55, 53, 34 and 30 questionnaires
returned from university students, school students, government office workers and
private office workers respectively, i .e. a total of 1 72.
9.4.1
Consumers were asked how often they used glue stick. The results are shown in Table
9.1.
1 85
Table 9.1
Frequency
Total
University
School
Gov.&Private
(1 .9%)
12 (18.8%)
( 1 .8%)
2 (3.8%)
8 ( 1 2.5%)
3-4 times a day
1 3 (7.6%)
once a day
1 1 (6.4%)
2-3 times a week
56 (32.6%)
1 1 (20.0%)
27 (50.9%)
18 (28.1%)
2-3 times a month
63 (36.6%)
24 (43.6%)
18 (34.0%)
21 (32.8%)
once a month
29 (1 6.9%)
1 9 (34.5%)
5 (9.4%)
5 (7.8%)
It was found that there were differences in frequency of glue stick usage between the
four groups of consumers. Office workers used glue stick more often than students.
About 31 % of office workers used glue stick daily. School students used glue stick
more often than university students, 34.5 % of university students and 9.4 % of school
students only used glue stick once a month.
9.4.2
Source of the Glue Stick Consumers Used
Consumers were asked how they received the glue stick they normally used. The results
Table 9.2
Source of Glue Stick Consumers Used
Source
Total
U n iversity
School
Gov.&Private
Buy
1 1 1 (64.5%)
51 (92.7%)
50 (94.3%)
10 (15.6%)
Provided by the office
57 (33.1%)
4 (7.3%)
3 (5.7%)
53 (82.8%)
Other
4 (2.3%)
3 (5.7%)
1 ( 1 .6%)
Note:
Some consumers selected more than one answer.
Most students bought the glue stick themselves, some were provided by their institute.
Some school students were provided with glue stick by their parents. Most office
workers obtained it from the office where they worked.
1 86
9.4.3
Brand Name of Glue Stick Consumers Used
Consumers were asked to give the brand name of glue stick they normally used. The
results are shown in Table 9.3
Table 9.3
Brand Name of Glue Stick Consumer Normally Used
Brand name
UHU
Total
University
School
Gov.&Private
1 55 (90 . 1 %)
54 (98.2%)
45 (84.9%)
56 (87.5%)
P RIlT
8 (4.7%)
PELIFIX
4 (2.3%)
Others
5 (2.9%)
( 1 .8%)
7 ( 13.2%)
4 (6.3%)
1 ( 1 .9%)
4 (6.3%)
It was found that UHU was the glue stick used by most consumers (90. 1 %), especially
university students. A few office workers used PELIFIX and other brands as well and
some school students used PRITT. Hence, i t can be said that when the consumers used
the new glue stick, most of them compared it with UHU.
9.4.4
Table 9.4
Size
Small
Medium
Large
Univ&School&Priv
Gov. office
1 1 5 (66.9%)
1 02 (73.9%)
1 3 (38.2%)
55 (32.0%)
36 (26 . 1 %)
19 (55.9%)
Total
2 (1 .2%)
2 (5.9%)
Most students and private office workers used a small size glue stick followed by a
medium size glue stick. The government officers tended to use medium size rather than
the small size. This was because they were heavy users so the medium size tended to
suit their usage.
187
9.5
CONSUMER ACCEPTABILITY OF THE NEW PRODUCT
Consumer acceptability of the new glue stick was measured. Consumers were asked to
make the decision whether the sample was acceptable or not. They could respond only
yes or no.
About 68 percent of consumers accepted the new glue stick. Long drying time and a not
very strong bond were the reasons that some of them did not accept the new glue stick.
It was found that there were no significant differences (p=O.91 5) in acceptability pattern
of consumers from different groups.
Consumers were asked to compare the developed product with the glue stick they
normally used. About 45 % of the consumer thought that the new product was the
same or better than the one they were using. The results are shown in Table 9.5. From
chi-square analysis, there were no significant differences (p=O.863) between consumers
from different groups.
Table 9.5
Comparison between the New Product and the Product Consumer

Currently Used
Comparison
Very much better
Number of consumers
4 (2.3%)
Slightly better
25 ( 14.5%)
The same
49 (28.5%)
Slightly worse
72 (41 .9%)
Very much worse
22 ( 12.8%)
The consumers gave their reasons why they thought the new product better or worse
than the one they were using.
1 88
The reasons they thought that the new product was better were:
>I-
Long drying time allowed the time for adjusting the paper so the paper can
be repositioned without any damage to the paper.
>I-
Bond not too strong
>I-
Not messy
>I-
Stronger bond strength
>I-
No bad smell, contained pleasant odour
>I-
Glue more cohesive not messy when used
>I-
Not make the paper wrinkle
>I-
No wrinkle when used with thin paper
>I-
Thinner layer of glue coated on the surface
>I-
Not stringy
>I-
Not stick to hand when used
>I-
Glue not come off in a lump
The reasons they thought that the new product was worse were:
>I-
Weaker bond strength
>I-
Glue not clear
>I-
Need more strength to rub glue on the working surface
>I-
Too slippery
>I-
Dries slower
>I-
Easy to peel the paper apart
It was found that abou t 66 % of the consumers who accepted the product thought that
it had weaker bond strength than their glue stick (see Appendix 9.3).
9.6
CONSUMER BUYING INTENTION
There were no significance differences (p=O.697) in consumer buying intention pattern

between consumers from different groups. About 57 lr, (98 respondents) of the
consumers said that they were going to buy the new product. Amongst these
consumers, 73.5 % thought that the developed product was better than or the same as
their glue stick (see Table 9.6. and Appendix 9.4).
1 89
Table 9.6
Comparison between the developed glue stick and the glue stick
consumers normally used (% of the consumers who said they would buy
the product)
Source
Number of Consumers
Very much /Slightly better and The same
72 (73.5%)
Slightly worse
25 (25.5%)
Very much worse
9.7
1 (0.01 %)
PRICE OF THE NEW GLUE STICK
The consumers who said they would buy the product were asked to recommend the
price that the new glue stick should be sold. They were asked to give the price
compared with the price of the glue stick they normally used. The results are shown in
Table 9.7.
Table 9.7
Price that consumers who said they would buy the product
recommended for the developed glue stick compared with their glue stick
Source
Number of consumers
Slightly higher and the same
26 (26.5%)
Slightly lower
68 (69.4%)
Very much lower
4 (4.1%)
There were no significance differences (p=0.340) between consumers from different

groups in pattern of price that consumers thought this developed glue should be sold.
About 26.5 % of the consumers that were willing to buy the developed product want
the product to be sold at the price slightly higher or the same as the glue stick they
were currently using. About 73.5 % thought that the developed glue should be sold at
a price lower than the price of glue stick they normally used.
190
9.8
THE CHARACTERISTICS OF THE DEVELOPED GLUE STICK WHICH

A TIRACTED CONSUMERS TO BUY
Consumers were given the list of the attributes which were supposed to be the
attributes of the developed glue stick. They were asked to rate each attribute according
to the degree of attraction that the attribute had toward their decision to buy or use the
product.
Table 9.8
Characteristics which could a ttract consumers to buy or try the developed

glue stick product
Characteristic
The most attractive
Second attractive
Third attractive
Not attractive
Safe
26 (26.5%)
20 (20.4%)
7 (7.1 %)
45 (45.9%)
Environmental
friendly
30 (30.6%)
18 ( 1 8.4%)
12 ( 12.2%)
38 (38.8%)
Made from Thai

tapioca flour
21 (21 .4%)
24 (24.5%)
8 (8 .2%)
45 (45.9%)
Cheaper than the

other glue
sticks
1 8 ( 1 8.4%)
13 (1 3.3%)
17 (17.3%)
50 (51 .0%)
Made in Thailand
10 (10.2%)
2] (21 .4%)
17 (1 7.3%)
50 (51 .0%)
New product
1 5 ( 1 5.3%)
1 4 ( 14.3%)
10 (10.2%)
59 (60.2%)
N ote:
The results were from consumers who were going to buy the developed glue stick. They could select
more than one characteristics
Safe, environmental friendly, and 'made from Thai tapioca flour' were the images of the
glue stick that attracted 52-60 % of them to buy or try it. This indicated that a
significant proportion of consumers were interested in a product that would not do any
harm to themselves or the environment and was made from raw material available in
Thailand.
For 'environmental friendly', there were differences in the ranking pattern of consumers
from d ifferent groups (see Table 9.9).
191
Table 9.9
Environmental friendly ranking by consumers
Source
The most attractive
Second attractive
Third attractive
Not attractive
School students
1 6 (53.3%)
6 (20.0%)
5 (1 6.7%)
3 (1 0.0%)
University student
14 (30.4%)
8 ( 1 7.4%)
4 (8.7%)
20 (43.5%)
4 ( 1 8.2%)
3 (13.6%)
1 5 (68.2%)
and Private
company
Government
officers
Note:
The results were from consumers who were going to buy the developed glue stick.
9.9
DISCUSSION
9.9.1
Reaction of the Target Consumers towards the Developed Glue Stick
Reaction of different groups of consumers towards the developed glue stick were quite
similar. This agreed with the results from the prototype testing in which consumers
from these target groups evaluated glue stick attributes in the same way. The product
was accepted by 68 % of the consumers and 57 % said that they were going to buy the
product. It was indicated from the survey that the consumers compared the product
with 'DHU' which was a commercial gl ue stick which most consumers used normally.
Therefore, the new product had a very good potential that if i t was launched into the
market about half of the target consumers would buy it.
However, according to the survey, only 26.5 % of the consumers were going to buy the
product at the price slightly higher or the same as their glue stick (about 15 Bahts).
About 68 % and 4 % were going to buy the product at prices slight lower and very
much lower respectively than their present glue stick. Maybe the reason for this was
that they thought the price of glue stick on the market was too high compared with
other glue products so they wanted the price to be decreased.
192
9.9.2
Selection of Market Segment
Since there were no differences in product acceptability and purchase intention amongst
the consumers from different groups, it can be said that the market segment would
include both students and office workers. From the consumer testing, office workers
used glue stick more often than students, therefore office workers can be considered as
heavy users. However, the number of students are higher than office workers so these
consumers can be included as target market as well. And most of the students buy glue
stick to use themselves so they can make their decision about which brand of glue stick
they use. Since the new glue stick could be sold at the price lower than the ones
currently in the market this would urge the office especially the government office to
buy the product and also the students who had a limi ted amount of money to spend
on sta tionery.
9.9.3
Improvement of the Developed Product
Although the developed product was reasonable accepted by the target consumers, i t
is possible to increase the acceptability b y improving o r modi fying some attributes.
From the information obtained from consumer testing, the new product should be
improved in terms of colour of the container, colour of glue stick, and odour / perfume.
Since the containers of the samples used in glue stick testing were the same as that of
UHU glue stick, in the real situation this could be easily avoided in order that
consumers could d ifferentiate between the products. The container could be redesigned
in terms of method of winding glue up-down, lid, colour and design on the outside of
the container. The container is the factor that make the price of the product very high,
a refill-stick should be considered.
Some consumers liked the glue stick that gave colour on the paper when applied on the
paper and the colour disappeared after the glue dried. This could also d isguise the
white colour of the new glue stick which consumers did not l ike because they were
used to the transparent glue stick.
Perfume could be added into the glue stick mixture to give a glue stick with a pleasant
odour. Flower perfume should be used in order to emphasize that the glue stick is
1 93
made from natural ingredients.
9.9.4
Future Research
A market test on the product would have to be conducted before the product is
launched on to the market in order that other factors (price, packaging, brand name and
advertising) which affectconsumers' buying decision apart from the sensory attributes
could be tested. Also in the market test, if the product was put side by side with the
competitors, i t should be possible to assess the influence of those products on consumer
buying decision and the market share could be estimated.
9.4.5
Position of the Developed Product in the Market
The developed product differed from the glue sticks already in the market in terms of
price, bond strength, raw materials. Therefore, it can be positioned as a low price glue
stick w ith adjustable bond - the paper can be readjusted after a period of time without
any damage but sticks strongly after drying. The product is considered more safe and
environmental friendly since i t has replaced 50 % of a synthetic polymer with a natural.
Since UHU glue stick is a very strong competitor, most of the glue stick users used this
brand, the size of the market is difficult to predicted. However, with the cheaper price
and other attributes which differentiate the developed product from the existing
products: odour, bond strength, not stringy after rubbing and cleanliness of work, these
should attract the consumers to buy the developed product. With the low price, it
should also encourage the non-glue stick user who does not use glue stick because of
the high price to buy the new glue stick as well.
9.10
CONCLUSIONS
A new glue stick product acceptable to the target consumers was developed. This glue
stick had the general properties comparable to the commercial product except that the
product had adjustable bond strength in which the paper could be reposi tioned after
a specific of period and not damage the paper. From the preliminary costing, the
194
product could be sold at a lower price than the products in the market. It is possible
that the price of packaging can be reduced in the long run since the original price
included the design and developing the mould. This product was found worthwhile to
be developed into a commercial product.
1 95
CHAPTER 10
D ISCUSSION AND CONCLUSIONS
10.1
INTRODUCTION
The research in this thesis aimed at optimizing the acceptability and a ttributes of a glue
stick product for Thai consumers. Consumers were directly involved in the product
optimization process and the final glue stick product developed was considered
acceptable by the consumers. The developed product was also comparable to the most
popular commercial glue stick on the Thai market in terms of acceptability and sensory
attributes. The developed product had a slightly higher adjustability and slightly lower
stickability than the commercial glue stick tested so it could be considered as an
adjustable bonding glue stick in which the paper can be repositioned without any
damage.
10.2
OPTIMIZATJON PROCESS
The product optimization process used in this study is similar to the processes that
many product developers have successfully used. This process had been used mainly
in food product development (Lagrange and Norback, 1987; Beausire et al., 1988) and
there were few applications in the development of other products (Rabino and
Moskowitz, 1 980; Moskowitz, 1984). This study was the first time it had been used in
the development of a glue product as recorded in the literature but of course it could
have been used in a commercial company. An optimum product was obtained in only
three steps, although the researcher had no experience in the product area being
studied. This method could be even more useful for a researcher working in an area
where the effects of ingredients on product attributes are well established. The optimum
process used in this study can be summarized as shown in Figure 10.1
196
INITIAL DEVELOPMENT
Consumer Survey
Literature Search
Identification of
Selection of
Important Attributes
Preliminary Formulation
Establishment of Test Methods
Selection of
Basic Formulation
PRODUCT FORMULATION
DEVELOPMENT
Experimental Design
Prototypes Development
Testing of the Prototypes
Constraint Generation
FORMAL OPTIMISATION
LP Model Development
STUDY
Optimum Formulations Development
Testing of Optimum Formulations
Pilot Scale Production

FINAL CONSUMER
TESTING
Consumer Testing
:PI;; : , ',', FiIlal,pioouct'f6r ,':'

;';';;":':':':;';'i,: osation..
Figure 10.1
,.,
Steps in Glue Stick Optimization
197
AttrI but..
Acceptability
..----.-----..---.-..--...
Sllppertfl888
Disintegration
Coverage
AdJuatab1l1ty
Smoothness
10
12
14
Sensory Scor..
Optimum product -- Ideal product
Figure 1 0.2
Comparison of prototypes from d ifferent steps during

product optimization
198
With the use of this method, only 3 steps were needed to get the optimum formulation.
It can be seen from Figure 10.2 that the product prototypes became closer to the
consumers' ideal product after the second step. The optimum formulation obtained from
applying the optimization technique, linear programming, gave a product with sensory
a ttributes which were quite close to the ideal product.
This method would be beneficial for the si tuation in Thailand as there is the need to
promote the use of agricultural products made in the country. Research needs to be
carried out in order to increase the use of the available raw materials, e.g. use as a
substitute for synthetic raw materials in certain products. With the use of this
optimization technique, products could be developed more efficiently and more
acceptable to consumers.
1 0.3
THE USE OF CONSUMER INPUT IN THE OPTIMIZATION PROCESS
In the optimization of a new product or an existing product, the main task is to

i nvestigate consumers' attitudes toward the product and the attributes of the product.
In the area of glue products or similar products, although the researchers have
recognised tha t the products have to meet the consumers' needs (Ando and Yamazaki,
1974; Palm, 1989), no research on consumers' reactions toward the products had been
reported. In this study, consumer input was used at al most all stages: in ini tial
development study, product prototypes, product formulation development, formal
optimization study and final product testing, and the study showed that consumers
could provide very valuable information to assist in the development of the product.
10.3.1 Benefits of Using Consumers throughout the Optimization Process
At the beginning of the study, consumers gave information in terms of the problems of
glue usage particularly of glue sticks. They could provide guidance on how the existing
products should be improved as well as generate the important attributes of the
product. Sinthavalai (1 986) also used information obtained from the target consumers
to identify the attributes of a nutritionally-balanced snack product. The important
attributes of glue sticks were used later in the sensory evaluation during the prototype
199
development and the final product optimization. This confirmed that the consumers can
be used as an i mportant tool in generation of the important attributes of products for
which no information is available.
With the new type of product, temporary bonding glue stick, which they had never
seen or used, a related product, self-adhering note pad, which had similar performance
had to be used as an example to provide a general concept of the product. This
technique was also used in the development of a facial scrub product (Moskowitz,
1 984). However, when the self-adhering note pad was used as a reference product, the
consumers recognised this product as an expensive product and therefore tended to
find the new product not very acceptable because they thought i t would be expensive.
Hence, in using a reference product one must be careful that no characteristics are
transferred to the new product concept which are not true.
Sensory scores for product attributes obtained from testing of the product prototypes
with consumer panels were used to create empirical models relating ingredients with
product sensory attributes. The se models were employed as constraints in the product
optimization stage using linear programming. Ideal product profile was used to
generate upper and lower limits of sensory constraints for the linear programming.
Although, the statistical accuracy of the results of a consumer panel is not as brilliant
as that of the trained sensory panel, a consumer panel can help the product developer
to measure consumer perception of prod uct attributes and the overall acceptability and
this can be related to product formulation. It was found that by using product profile
with line scale, quantitative data could be obtained. The model relating the ingredients
and the sensory attributes can be used in the formal optimization with reasonable
accuracy.
In final prod uct testing, Thai consumers evaluated the developed product in terms of
acceptability in the home-use situation and compared the product with the commercial
product they normally used. This was very important in considering how well the
product performed compared with their usual brand. Consumers had opportunity to
use the developed product in the same way as they normally used glue stick, as many
times as they wanted, before they evaluated the product. Purchase intention and price
which consumers were prepared to pay for the product were also obtained. This
200
information allowed the product developer to estimate the success of the product before
the product was actually launched into the market.
It can be concluded tha t the consumer study and the consumer panels provided several
advantages. Firstly, prediction of product success is based on the opinion of the
individuals whose buying activities will determine the actual sales achieved directly or
i ndirectly. Secondly, the consumer panel can be used during the product optimization
process to get the potential consumers' preference and reaction to products so that the
product can be improved before the final product is launched. There is also more
confidence in launching a product that is acceptable to the consumer because of the
continuous testing with consumers. However, in spite of the many advantages, there
are also some d isadvantages. Firstly, the primary research and data collection is time
and cost consuming. Secondly, competitors may learn about the company's strategies
from the public testing of the product and plan competitive actions.
10.3.2 Differences between Thai and New Zealand Consumers
No major d ifferences in glue stick usage or important product attributes were found
between Thai and New Zealand consumers. Although in generating the product
a ttributes, New Zealanders appeared to generate more attributes than the Thais, some
of these attributes were similar and could be grouped into the same category and some
of them were not important. This showed a difference in the consumers' tendency to
use descriptive terms to describe their perception of a particular product and product
usage. Therefore, in conducting a consumer survey or testing of a product in Thailand,
this factor have to be taken into account in the designing of the questionnaire. Thai
consumers felt more easy with multi-choice questions or questions requiring brief
answers. The questions have to be designed carefully in order to elicit as much
information as possible from the consumers. Indirect questions may need to be used
when the subject is directly involved with personal status such as price of the product.
-----
------
201
10.3.3 Selection of the Important Attributes
The attribu tes emphasized in the study of the gl ue stick optimization were the usage
attributes which included the attributes consumers perceived while applying glue stick
as well as the effect of glue on the finished work and the glue performance. Although
other attributes also had great impact on product acceptability such as: glue appearance,
colour, aroma, packaging, price, and brand name, they were not included in the study.
This was because this study was aimed at generating the optimum glue stick
formulation using tapioca starch as substitu te for synthetic adhesive substance polyvinyl pyrrolidone, therefore only the effects of major ingredients on product
a ttributes were focused. The effects of other ingredients: perfume and colour could be
evaluated in future study and the effect of packaging, price and brand name should
a lso be studied before the product is launched into the market.
It was found that the elicitation method was useful in generating the list of important
attributes. This list not only helps the product developer in establishing the set of
a ttributes for product testing but also helps in planning of marketing policy. The
measurement of attribute importance by the elicitation method and direct-rating method
gave similar results for the most important attributes, but slightly d ifferent for the less
important ones. Price, which did not appear to be an important attribute by the direct
rating method, was considered important by the elicitation method. This suggests that
the selection of the importance measuring technique is vital and the type of product and
type of consumers have to be taken into account. Although the consumers in general
wanted the price of the glue stick product to be decreased, this attribute should not be
directly mentioned in the advertising of the product as consumers might perceive the
product as a low quality product.
1 0.3.4 Relationship of Acceptability to Purchase Intention and Price to Buy
It was found from the study that acceptability of the product was correlated highly with
purchase intention and the price that consumers were prepared to pay for the product.
Generally researchers have used acceptability, purchase intention and price to buy in
the same manner. In testing of hand lotion, Moskowitz (1983) found tha t purchase
intent varied with both liking and stated i tem price. Increases in liking or product
202
acceptability generated a relatively slowly increasing purchase intent. Changes in the
formulation to increase purchase interest by increasing liking would show less of an
effect than would pricing change. Moskowitz and Jacob (1988) found that price
moderately influenced purchase intent in frankfurter development.
Although the product was tested in 'horne use test' where the consumers were able to
judge the performance of the product in an uncontrolled environment, a market
simulation test should be conducted in order to obtain information on initial triat repeat
purchases and sales volume.
1 0.3.5 Comparison between Consumer Panel and Trained Panel
Use of the trained panel at the beginning of the development of the basic formulation
helps to define the sui table area of the ingredients to be used in the design of the
experiment. However, with the use of the consumer panel in the prototype testing, the
perception of consumers toward the prototypes can be assessed. This provides the faster
method to detect consumers' reaction and generates an ideal product profile which
cannot be obtained by using a trained sensory panel. With the use of the ideal product
profile, not only the prototypes can be compared with the ideal product but the
d irection in which the product should be developed is also identified. This profile can
be used to set up the sensory constraints in the formal optimization study in which the
optimization technique is used.
It was found that the Thai consumers were able to evaluate up to 10 product attributes
using line scales. This confirms the results of the study by Lawless and Malone (1986)
and Lawless ( 1 988) that consumers could efficiently identify the difference between
products by using line scales.
Although it was not possible to directly compare the results obtained from the
consumer panel and the trained panel in this study since the set of samples used in the
tests were different, it was found from correlations between attributes, that both panels
evaluated most attributes in the same way.
203
1 0.4
USE OF EXPERIMENTAL DESIGNS IN PRODUCT OPTIMIZATION
Experimental design is an essential tool for product formulation development since i t

refers t o the statistical design of experiments b y which the researcher develops a matrix
of systematically varied stimuli (Box et al., 1978) . The statistical design l ays out specific
combination of the formulation variables so that a limited combination is generated.
Basic glue stick formu lation is a mixture formulation, a change in the level of one
ingredient in the formulati on will also result in changes of the levels the other
ingredients. Therefore, mixture designs were used during the product optimization
process to study the effects of ingredients on physical and sensory attributes of starch
based glue stick. This enabled the researcher to generate the subset of all possible
samples which could be tested. The use of an experimental design to generate multi
products to be tested is considered more efficient than the traditional method of back
and-forth testing. The multi-product approach tests many prototypes and develops a
model relating input variables and output variables.
The relationships between ingredients and attributes of the product obtained from the
experimentation were used for directing the experimentation and were assigned as
constraints in the linear programming model. In comparison with the sequential
procedure followed by some optimization techniques, this method included all
Significant a ttributes and significant raw materials in the equations. Hence, i t was very
valuable for product optimization in which the product developer was not familiar with
the product and could not predict which ingredients should have significant effects on
the product.
Since the relationships between attributes and ingredients must be linear to be used in
linear programming this may limit the use of this method in product optimization.
1 0.4.1 Selection of the Preliminary Formulation
Before the product prototypes could be developed, the preliminary formulation had to
be chosen. In this study, the preliminary formulation was selected from the commercial
formulations found in the literature. The formulation and the method of processing can
204
be used as a guidance only, the modification has to be done according to the purpose
of the study and the resources: raw materials and equipment available. Normally simple
formulation should be used at the beginning in the case where the relationships
between the ingredients and product properties are not established, so the product
developer can manage to identify those relationships.
1 0.4.2 Use of I ngredient/Product Attribute Relationship Models in Final Product

Optimization
For the glue sticks, usage attributes were the important attributes, but some of these
attributes acted counter to each other. This meant that to increase acceptabil i ty of one
a ttribute, one needed to sacrifice the acceptability of other attributes.
Disintegration was one of the most important attributes of glue stick. It was correlated
highly with product acceptability. Disintegration played an important role in glue stick
usage in terms of ease of use, which included slipperiness, degree of coverage as well
as the effect on paper. If disintegration was too high, it was likely that too much glue
would be applied which was wasteful and could be messy and could cause wrinkling
and curling of paper. On the other hand if the stick was too hard, inadequate adhesive
was transferred to the surface to be adhered to allow a firm bond to be achieved. It was
likely also that the surface was damaged by the hard force applied. It was found that
disintegration could be decreased by increase of gel forming agent and reduction of
adhesive substance in the formulation. However, this resulted in a reduction of
stickability of glue stick. Therefore in order to obtain an optimum product, there had
to be a trade-off between these attributes.
Fishken ( 1988) using response surface analysis in reformulation of pizza topping also
found that it could be difficult to optimize two ingredients. He found that the optimal
formulation did not seem to meet the consumer desire for more meat and cheese, since
at the highest level of cheese, a level dictated by the physical l imits of the crust, the
optimal formulation included the lowest meat level tested. Therefore, with the inter
related attributes it may not be possible to maximize the l iking of all attributes in the
formulation of the product.
205
Moskowitz (1982) stated that normally the variation of two or more attributes
simultaneously i mpacts on acceptance of a product. Sometimes these sensory attributes
interact with each other. Variation in one attribute alone does not suffice to show the
full range of acceptance levels. Furthermore, only by relating acceptance to the key
attributes in combination can the product developer be sure of having generated the
correct combination of sensory attributes.
Linear programming was used to generate the optimum formulation using these
sensory a ttributes as constraints. This work extended the work done by Beausire et al.
( 1 988), Kavanagh (1978), Chan and Kavanagh (1 988) by adding to the model the
empirical equations relating ingredients and attributes (physical and sensory attributes)
of the product as well as including consumers' ideal product profile as the limit of
sensory constraints. Also with the use of both consumer acceptability and sensory
attributes during the product optimization, it is possible to know which attributes have
to be i mproved and by how much in order to increase acceptabil ity.
1 0.5
GLUE STICK FORMULATION AND ATTRIBUTES
1 0.5.1 Use of Tapioca Starch in Glue Stick
It was found that although modi fied tapioca starch used in this study (cross-linked and
stabilized starch) could not totally replace polyvinyl pyrrolidone in glue stick
formulation, it could effectively replace half of the polyvinyl pyrrolidone. In the past
only 2-6 percent by weight of a carbohydrate or modified carbohydrate polymer was
recommended in glue stick formula tion (Werke H,u.M. Fischer, G.m.b.H., 1 974). With
use of modified starch in the formulation, it was also possible to shorten the processing
time from 6-7 hours to only one hour at 90 C.
Research on starch specially modified for use as adhesive substance in glue stick should
be carried out in order that more starch can be used in the formulation. This could
reduce the formulation cost as well as increase the use of natural raw material in the
product.
206
1 0.5.2 Sensory Properties of Glue Stick
The sensory attributes of glue sticks generated in this study were:

* Rub-out
- Slipperiness
- Disintegration of stick
* Glue residue on paper
- Degree of coverage
* Performance of glue stick - Adjustability

- Stickability
>I-
Effect on paper
- Smoothness
Although there were some other sensory attribu tes the consumer perceived when using
glue stick, some attributes were related to the attributes mentioned above and some
were not considered as important attribu tes, so it is not necessary to include these
a ttributes in sensory testing. Aesthetic attributes such as appearance, colour, and aroma
of the glue stick were not included since the aim of the thesis was only to study the
basic formula tion of a glue stick using tapioca starch.
1 0.5.3 Relationships of Sensory Properties and Physical Tests
Physical testing can be used to assess some of the sensory attributes of glue stick. It was
found tha t wet glue and dry glue per area, hardness and moisture content of glue stick
could be used to estimate consumer reaction to product attributes. Since peel strength
testing used in this study was not correla ted with consumer reaction, in order to assess
the perceived stickabiJ ity, a suitable physical measure has to be developed. The methods
which should be considered are shear strength used by Ando and Yamazaki (1974) or
peel resistance of adhesives (T-peel test) ASTM standard testing method 01876-72
(ASTM, 1991 c).
10.5.4 Properties of Developed Glue Stick Compared with Commercial Products
In the development of existing products, the prototypes are normally compared with
products already in the market. Rabino and Moskowitz (1980) compared sensory
a ttributes of the product prototypes with the commercial products during the
207
development of skin lotion.
The prototypes developed during this study were compared with the commercially
available glue sticks in order that the developed product possessed the physical
a ttributes in the suitable ranges. This was to make sure that the product could perform
the task it was supposed to do. However, some of the physical attributes of the
commercial glue sticks were in a very wide range, and as a result some of the sensory
attributes were far from the consumers' desired product. Therefore, in the optimization
study, the consumer ideal product profile was used so as to develop a product which
possessed the attributes close to those of the ideal product. The outcome of the
optimization process gave a glue stick which possessed general properties similar to the
commercial glue sticks, however the strength of the bond was slightly lower than that
of the commercial products. With this type of bonding, there is enough time for paper
to be readjusted or taken off from the receiving surface without any damage, but the
bond will strengthen after a period of time.
1 0.6
RELA TIONSHIP
BETWEEN
MARKET POSITIONING
AND
PRODUCT
ATTRIBUTES
From both the consumer study at the beginning of the product optimization process and
the consumer testing of the developed product, it was found that consumers wanted
the glue stick to be sold at a price lower than the prices of the products already on the
market. One reason was that consumers considered the prices of the commercial glue
sticks were too high compared with other glue products in the market. Hence, if the
developed product was sold at a lower price there was a high opportunity that the
product could compete with the competitors in term of market share and it may be
possible to make non-users to become glue stick users and increase the total market.
In terms of stickabili ty, the developed product had lower bond strength than the
commercial glue stick according to the consumers so it is likely that the product should
be positioned as a weaker bonding glue stick.
208
1 0.7
RECOMMENDATIONS FOR FUTURE WORK
Although the final product was successfully developed and was reasonably acceptable,
there were some points to be suggested for future work. Since the aim of the project
was to study consumer inpu t in the optimization of starch based glue stick, the type of
starch which is most suitable for this type of product was not examined. Some more
work could be done on the selection of suitable starch. Tn order to obtain a cheaper
product formula tion, a study is needed on reducing the amount of total adhesive
substance or increasing the percentage of starch products which are cheaper than
synthetic polymers as adhesive substance in the formulation.
Colour and fragrance can be added in the product formulation in order to improve the
acceptability of the product. These attributes could be used to build the image of the
product as a natural product by using sweet fragrance or flower l ike perfume and soft
colour. These would also differentiate the product from the existing products which
possess harsh colour and rather unpleasant aroma. As the product is aimed for the Thai
market the packaging can be designed with Thai classical style which not only adds a
value-for-money image but also attracts the consumers who prefer the classical Thai
product. The packaging of the product including lid and the application method should
be improved for more ease of use.
Although, there was no problem on deterioration of the product from microbiological

spoilage during the study, no formal storage test was conducted. Since the developed
glue stick is made from a starch product, therefore a storage test should be carried out
to make sure that the product can be kept at room temperature without spoilage from
microorganisms. If there is a need for use of preservatives, a study on sui table type of
preservatives for the system should be carried out.
209
10.8
CONCLUSIONS
This study demonstrated the successful use of consumer inputs in the product
optimiza tion (product design) of a glue stick product for the target consumers in
Thailand - school and university students, office workers. The consumers identified the
important attributes of glue stick products as: effect on paper, cleanliness of work, ease
of use and stickability. These attributes were used to assess the performance of the
developed product prototypes. Consumers quantitatively evaluated the sensory
a ttributes of the prototypes using line scales. In this study, consumer acceptabil ity of
glue sticks was correlated with physical hardness, moisture content, ease of use
(disintegration, slipperiness, degree of coverage) and effect on paper (smoothness) of
the products. The levels of these attributes in the glue sticks could be used to predict
consumer acceptability.
The empirical equations showing relationships between each group of ingredients in the
formulation: adhesive (modified starch and polyvinyl pyrrolidone), solvent (water and
glycerin), gel-forming agent (stearic acid and sodium hydroxide) and the consumers'
acceptability and perceived sensory attributes were established. A l inear programming
model developed from these relationships together with cost and component constraints
was found useful in helping to generate the optimum formulation in terms of cost
minimization and acceptability maximization . It was found that with the use of the
consumer ideal profile to set limits for sensory constraints, it was possible to generate
products which had high acceptability.
The study showed that physical attributes and sensory attributes of the products should
both be considered in order to obtain optimum formulations. It is recommended that
physical testing is concurrently used with the sensory testing during the optimization
of the product formulation. If a suitable physical test is correlated with a sensory
a ttribute, then the sensory test can be replaced by the physical test in the routine work.
In this study, physical hardness, moisture content, amount of glue applied per area
were correlated with the sensory attributes.
The final product testing by the target consumers in Bangkok showed that the product
could be a success in the market and the project was believed to be worth continuing
210
for commercialization by the private sector. However, it was believed that further
improvements could be made to the product i tself, in terms of colour and aroma, as
well as packaging, and a market test should be conducted in order to measure the
potential success of the product against the competitors.
With the use of representative consumers as navigators and also for objective
measurement throughout the process, the researcher can obtain the optimum product
in a minimum number of steps. This research confirm that consumer testing with
representatives of the target population remains critical to the product optimization
process and cannot be substituted .
211
REFERENCES
Adhesive Tapes Limited. 1974. Improvements in and Relating to Adhesives. New

Zealand Patent No. 1 67,516.
Aldridge, D.N., Shore, A.J., and Wiley, M. 1983. Finding the cube route. In
Demonstrating the Contribution of Research. ESOMAR Congress (36th), Barcelona.
Almeida-Dominguez, N.G, Valencia, M.E., and Higuera-Ciapara, 1. 1 990. Formulation
of corn-based snacks with high nutritive value: Biological and sensory evaluation.
Alpert, M.l. 1 971. Identification of determinant attributes: a comparison of methods.

Journal of Marketing Research, 8(5), 1 84-191 .
Alreck, P.L. and Settle, R.B. 1 985. The Survey Research Handbook. Richard D. Irwin
Ltd., Homewood.
Ando, T. and Yamazaki, H. 1974. Adhesive Crayon Composition Containing Sorbitol

Benzaldehyde Reaction Product as Additive. US Patent No. 3,846,363.
A.S.T.M. 1 986. Manual on consumer sensory evaluation. ASTM Specia l Technical

Publication No.682. ASTM Philadelphia, PA.
A.S.T.M. 1991a. Standard test method for appli ed weight per unit area of dry adhesive
(0898-90). Annual Book of ASTM Standards. Vol. 15.06: 18-19.
A.s.T.M. 1 991b. Standard test method for peel or strippi ng strength of adhesive bonds
(D903-49). Annual Book of ASTM Standards. Vol. 15.06: 22-24.
A .S.T.M. 1 991c. Standard test method for peel resistance of adhesives (T-peel test)
(D1876-72). Annual Book of ASTM Standards. Vol. 15.06: 122-124.
Baker, M.J. 1991. Research for Marketing. Macmillan Education Ltd., Hampshire.
212
Beausire, RL.W., and Earle, M.D. 1986. Optimum location profiles: a profile technique
used in product design. Food Technology in Australia, 38(7): 298-300.
Beausire, RL.W., Norback, J.P. and Maurer, AJ. 1988. Development of an acceptability
constraint for a linear programming model in food formulation. Journal of Sensory
Studies, 3: 137-149.
Bender, F.E., Kramer, A and Kahan, G. 1982. Linear programming and its appli cations
in the food industry. Food Technology, 36(7): 94-96.
Bennett, H. 1963. Industrial Waxes. Chemical Publishing Company, Inc., New York.
Birn, R 1 990. The Effective Use of Market Research: A Guide for Management. Kogan
Page, London.
Box, B.E.P., Hunter, W.B., and Hunter, J.S. 1978. Statistics for Experimenters. John Wiley
& Sons, New York.
Brand t, M.A, Skinner, E., and Coleman, J. 1963. Texture profile method. Journal of Food
Science, 28: 404-410.
Brennan, G.A 1966. Adhesive Appl i cator Crayon. US Patent No. 3,267,052.
Cairncross, W.E., and Sjostrom, L.B. 1 950. Flavor profile - a new appr o ach to flavour
problems. Food Technology 4: 308-331 .
Chan, KY. and Kavanagh, P.E. 1988. The application of linear programming and
regression analysis of light duty liquid detergent formulation. Journal of American
Oil
Chemist Society 65(12): 1971-76.
Civille, G.V., and Liska, I.H. 1975. Modifications and applic ations to foods of the
General Foods sensory texture profile technique. Journal Texture Studies, 6: 1 9-23
-----
-------
213
Civille, G.V., and Szczesniak, AS. 1973. Guidelines to training a texture profile panel.
Journal of Texture Studies, 4: 204-223.
Cooper, H . R., Earle, M.D., and Triggs, CM. 1989. Ratios of ideals - A new twist to an
old idea. In Product Testing with Consumers for Research Guidance. Edited by E.
Wu. ASTM, Philadelphia.
Dano, S. 1 972. Li near Programming in Industry. Springer-Verlag/Wein, New York.
Dethmers, A E. and Boomsma, J.C 1 989. Appl i cations of sensory science within the
home care business. Journal of Sensory Studies 3: 193-204.
Durgee, J. 1 990. Qualitative methods for developing advertising that makes consumers
feel, 'Hey, that's right for me'. The Journal of Consumer Marketing 7(1 ) : 1 5-2 1 .
Earle, M D. 1 989. Introduction t o Product Development. Product Development Sub

Department, Technology Faculty, Massey University, New Zealand.
Fishken, D. 1983. Consumer oriented product optimization. Food Technology, 37( 1 1 ) :

49-52.
Fishken, D. 1 988. Marketing and cost factors in product optimization. Food Technology,
42( 1 1 ) : 1 38-140.
Fleche, G. 1985. Chemical modification and degradation of starch. In Starch Conversion

Technology. Edited by G.M.A Van Beynum and J.A Roels. Marcel Dekker, Inc.
New York.
Fuller, L. 1 984. Use of panels for qualitative research. Journal of the Market Research
Society, 26(3): 21 1-220.
Gibson, I.M. 1973. The evaluation of hand-care preparations. Journal of the Society of
Cosmetic Chemists 24: 31-41 .
214
Giovanni, M . 1 983. Response surface methodology and product optimization. Food
Technology, 37(1 1): 41-45, 83.
Gollub, H. J., Hechenberger, D.A., and Moermann, R 1987. Adhesive Stick Composition.
European Patent No. 233,685.
Gordon, N.M. 1985. A Product Development, Positioning and Optimization Process

Guided by Organizational Objectives. Master of Science Thesis. University of
Wisconsin, Madison.
Gordon, N.M. and Norback, J.P. 1 985. Choosing objective measures when using sensory
methods for optimization and product position. Food Technology, 39( 1 1 ) : 96-10 1 .
Griffin, R and Stauffer, L . 1990. Product optimization in central-location testing and

subsequent validation and calibration in home-use testing. Journal of Sensory
Studies 5: 321 -240.
Hague, P.N. and Jackson, P. 1990. How to Do Marketing Research. Kogan Page Ltd.,
London.
Hart, S.J. 1 987. The use of the survey in Industrial Market research. Journal of
Marketing Management, 3(1): 25-38.
Heeler, RM., Okechuku, C. and Reid, S. 1979. Attribute importance: contrasting

measurements. Journal of Marketing Research, 1 6(2): 60-63.
Hoggan, J. 1975. New product development. Master Brewers Association of America

Technical Quarterly, 1 2(2): 81-86.
Hovenden, J.E. 1 979. Variation and repeatability of an untrained beef sensory panel.
Journal of Food Science 44: 1598-1601.
IBM. 1 966. Linear programming - Meat blending. IBM, Tech. Publications Dept., White
Plains, New York.
-------
215
Jaccard, J., Brinberg, D. and Ackerman, L.J. 1 986. Assessing attribute importance: a
comparison of six methods. Journal of Consumer Research 1 2(3), 463-468.
Kavanagh, P.E. 1978. The application of linear programming to paint and resin
formulation. J. Oil Col. Chern. Assoc., 61 : 146-150.
Lai, P.W. 1987. Development of a Bakery Snack for Export from New Zealand to
Malaysia. PhD. Thesis in Product Development, Massey University, Palmerston
North, New Zealand.
Lagrange, V. and Norback, J.P. 1987. Product optimization and the acceptor set size.
Journal of Sensory Studies 2: 1 19-136.
Lawless, H.T. 1 989. Logarithmic transformation of magnitude estimation data and
comparisons of scaling methods. Journal of Sensory Studies 4: 75-86.
Lawless, H .T. and Malone, G.J. 1 986a. The discrimination efficiency of common scaling
methods. Journal of Sensory Studies 1 : 85-98.
Lawless, H .T. and Malone, G.J. 1986b. A comparison of rating scales: sensitivity,
replicates and relative measurement. Journal of Sensory Studies 1 : 1 55-174.
Meilgaard, M., Civille, G.V., and Carr, B.T. 1987a. Sensory Eval uation Techniques. VoLI.
C RC Press, Inc., Florida.
Meilgaard, M., Civ ill e, G .V., and Carr, B.T. 1987b. Sensory Evaluation Techniques.
Vol.lI. CRC Press, Inc., Florida.
Moskowitz, H.R. 1 982. Sensory analysis of thickness. Cosmetics and Toiletries 97(3): 3545.
Moskowi tz, H.R. 1983. Product Testing and Sensory Evaluation of Foods: Marketing
and R&D
Approaches.
Food & Nutrition Press Inc., Connecticut.
216
Moskowitz, H.R. 1984. Cosmetic Product Testing : A Modern Psychophysical Appro ach.
Marcel Dekker, Inc., New York and Basel.
Moskowitz, H.R. 1 985. New Directions for Product Testing and Sensory Analysis of
Foods.
Food & Nutrition Press Inc., Connecticut.
Moskowitz, HR. 1987. Optimizing consumer acceptance and perceived product quality.
In Objective Methods i n Food Quality Assessment. Edited by I.G. Kapsalis. CRe
Press, Boca Raton, Florida.
Moskowitz, H.R. 1988. Applied Sensory Analysis of Foods. Vol.1I e.R.e. Press, Boca
Raton, Florida.
Moskowitz, H .R., and Chandler, J.W. 1 978. Consumer perceptions: a ttitudes and trade
offs regarding flavour and other product characteristics. Food Technology 32(1 1 ),
34.
Moskowitz, H.R. and Jacobs, B.E. 1988. Simultaneous optimization of products and
concepts for foods. In Applied Sensory Analysis of Foods. Vol II. Edited by H
Moskowitz. CRC Press, Inc., Florida.
Moskowitz, HR. and Rabino, S. 1983. Alternative strategies for product optimization.
Advances in Strategic Management 2: 99-123.
Muszik, J.A. and Dierichs, W. 1971 . Adhesive Appl i cator Crayons. US Patent No.
3,576,776.
Nicklin, S.H. 1 979. The use of linear programming in food product formulations. Food
Technology in New Zealand 33(6): 2-7.
Norback J.P. and Evans, S.R. 1 983. Optimization and food formulation. Food
Technology 37(4): 73-80.
Palm, P. 1989. Adhesive Stick. European Patent No. 326,530.
217
Pangborn, RM., Guinard, J. and Meiselman, H.L. 1 989. Evaluation of bitterness of
caffeine in hot chocolate drink by category, graphic, and ratio scaling. Journal
0f
Sensory Studies 4: 31-53.

Penny, rc., Hunt, LM. and Twyman, W.A. 1972. Product testing methodology in
relation to marketing problems - a review. TournaI of the Market Research
Society 1 4(1): 1 -29
Pletcher, W.A. and Wong, R. 1978. Friction-Activatable Adhesive and Articles Thereof.
US Patent No. 4,066,600.
Rabino, S. and Moskowitz, H.R 1 980. Optimization the product development process:
strategical implications for new entrants. Sloan Management Review (Spring), 4551.
Radley, J.A. 1 976. Industrial Uses of Starch and Its Derivatives. Applied Science
Publishers Ltd ., London.
Rust, R.E. 1 976. Sausage and Processed Meats Manufacturing. AMI Center for
Continuing Education, Am. Meat Inst., Washington, DC.
Rutenberg, M.W. and Solarek, D. 1984. Starch derivatives: Production and Uses. In
Starch Chemistry and Technology. Edited by RL. Whistler, J.N. BeMiller and E.F.
Paschall. Academic Press, Inc., Orlando.
Schutz, H.G. 1 983. Multiple regression appro ach to optimization research. Food
Technology, 37(1 1 ): 46-62.
Schwartz, N.O. 1975. Adaptation of the sensory texture profile method to skin care
product. Journal of Texture Studies 6: 33-42.
Side!, J.L., and Stone, H. 1983. An introduction to optimization research. Food
Technology, 37(1 1): 36-38.
218
Sidel, J.L. and Stone, H. 1985. Sensory Evaluation Practices. Academic Press. New York.
Sinthavalai, S. 1 986. Development of Nu tritionally-Balanced Snack Product for Urban

School-Age Thais. PhD. Thesis in Product Development, Massey University.
Sne'7Rand Marquard i). 1 974. Extreme vertices Designs for linear mixture models.
Technometrics, 1 6(3): 399-408.
Stone, H., Sidel, J., Oliver, S., Woolsey, A., and Singleton, R.c. 1 974. Sensory evaluation
by quantitative descriptive analysis. Food Technology 28( 1 1): 24,26,28,29,32,34.
Swinkels, J.J.M. 1 985. Sources of starch, i ts chemistry and physics. In Starch Conversion
Technology. Edited by C.M.A. Van Beynum and J.A. Roels. Marcel Dekker, Inc.
New York.
Szczesniak, A.S. 1963. Classification of textural characteristics. Journal Food Science, 23:
385-389.
Szczesniak, A.S. 1987. Correlating sensory with instrumental texture measurements - An

overview of recent developments. Journal of Texture Studies 18: 1 -15.
Szczesniak, A.s., Loew, B.J., and Skinner, E.Z. 1975. Consumer texture profile
techniques. Journal of Food Science,
40 :
1 253 - 12 5 b.
TTTA. 1 990. Tapioca: The Answer to Lower Feed Cost. The Thai Tapioca Trade
Association. Year Book Publisher Co., Ltd., Bangkok.
TuB, D.S. and Albaum, C.s. 1973. Survey Research. International Textbook Co., Ltd.,
Aylesbury.
Uaphithak, H. 1990. Linear Programming and Consumers' Ideal Sensory Attributes in

Product Optimization. Master Degree Thesis in Product Development,' Massey
University, Palmerston North, New Zealand.
219
Werke H.u.M. Fischer, G.m.h.H. 1974. Stick for the Applic ation of Adhesive to
Substrates. UK Patent No. 1,365,147.
Williams, A.A., Whittlestone, D.J., and Martin, D.C. 1 992. The role of fragrances in
product development: Turning images into fragrances. Marketing and Research
Today (May): 95-106.
Wiriyacharee, P. 1 990. The Systematic Development of a Controlled Fermentation

Process Using Mixed Bacterial Starter Cultures for Nham, a Thai Semi-Dry
Sausage. Ph.D Thesis in Food Technology, Massey University, Palmers ton North,
New Zealand.
Wurzburg, O.B. 1986. Converted starches. In Modified Starches: Properties and

Uses. Edi ted by O.B. Wurzburg. CRC Press, Inc., Florida.
SCALING EXERCISES
APPENDIX 3.1
lNSTRUCTIONS:
. Mark on the line at the right to indic"te the proportion of the area
that is shaded.
PRESCREENING QUESTIONNAlRE FOR A GLUE STICK PANEL

NAME:
PHONE (HOME)--TIME
What time during the week will you be available on a regular basis (between
9-6 pm weekdays)
Are you using glue stick regularly?

No
___ Yes
(If no, have you ever used glue stick?
Yes
___
I.
NON
__
_
_
_
_
_
_
_
_
_
_
_
_
_
_
_
_
_
2.
NONE
_
_
_
_
_
_
_
_
_
_
_
_
_
_
_
_
_
_
3.
NONE
4.
NONE
5.
NONE
6.
NONE
7.
NONE
8.
NONE
9.
NONE
10.
NONE
ALL
ALL
No)
Aniwer each question in your own words, as best as you can.

1. When you rub an oily film on your skin, how do your fingers move?
or
____ Slip
___ Drag (check one)
2. What is the difference between lumpy and gritty?
3. How do you describe the feeling of your fingers
--------ALL
_
_
_
_
_
_
_
_
_
ALL
_
__
__
_
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
ALL
__
_
_
_
_
__
__
__
__
__
__
_
_
__
__
__
__
__
3.1 when you touch sellotape on the side coated with adhesive
3.2 when you move your fingers along the length of sellotape on the side coated
with adhesive
4. Please describe your ideal adhesive
5.
lease describe your ideal
glue stick
____,A
___ LL
--------------------
ALL
_
_
_
__
__
__
-J
ALL
_
_
_
_
_
__
__
__
__
__
__
__
__
__
__
__
__
__
-J
ALL
__
_
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
ALL
__
__
__
__
__
__
N
N
o
APPENDIX 3.2
GLUE RESIDUE O N PAPER
The following attributes are evaluated by visual

inspection
Descriptive Terms Developed by the Trained Panel

Evenness
- Evenness of adhesive coated on paper after applying adhesive

on paper. Rated as ' uneven - very even'.
Degree of coverage
- Degree of adhesive cover the area of paper after applying one

coat. Rated as 'none - total'.
Thickness
- Amount of adhesive left on paper after appl}ing one coat. Rated

as 'very thin - very thick'.
TERM DEFINITIONS
APPEARANCE
Wind up the stick and evaluate the appearance of the stick
Whiteness
- Whiteness of the adhesive stick. Rated as 'coloured - white'.
Opacity
- Opacity of the stick. Rated as 'transparent - opaque'.
Evenness
- Homogeneity of the stick. Rated as ' uneven - very even'.
Visibility of glue trail - Ease of seeing the glue trail on paper. Rated as 'invisible - very
visible'.
Smoothness
- Smoothness of the stick surface. Rated as 'very rough - very

smooth'.
EFFECT ON PAPER Place another piece of paper on top of the coated paper and rub
Molslns
- Moi!lness of the stick surface. Rated as 'very dry - very wet'.
Hardness
- Perceived hardness by visual inspection. Rated as ' very soft very hard'.
repeatedly with fingers then evaluate effect on paper by feeling

the surface of the papers with fingers and using visual
inspection.
Wemess
- The amount of moistness left on the paper after applying

adhesive. Rated as 'very dry - very wet'.
Smoothness
- Unevenness of paper surface may cause by moisture or lump of

adhesive. Rated as 'very wrinkly - very smooth' .
Cleanliness of work
- Adhesive residue left on top surface of (degree of stain) paper

which makes paper dirty and sticky. Rated as 'very dirty - very
clean ' .
S EN SORY ATTRIBUTES WHILE APPLYING G LU E STICK
Ease of applying
Rub the stick along the surface of paper and spread adhesive
cover the area to be bonded.
to
Slipperiness
- The feeling of force between working surface of the stick and

paper while rubbing the stick along the paper. Rated as 'drags
slips'.
SpreadabiJity
- Ease of spreading adhesive on the paper. Rated as 'difficult to

spread - easy to spread' .
Perceive attribute of the stick while applying
Visibility o f glue
trail
- The degree of glue trail which can be seen through the paper.
Rated as paper ' invisible - very visible'.
STICKABILITY
While using the stick the following a ttributes are evaluated:

Adjustability
- Ease of repositioning the paper. Rated as 'very difficul t - very

easy'.
Bond strength
- Strength of adhesive bond between two

surfaces. Rated as 'very weak - very strong'.
Hardness
- Perceive hardness of the stick. Rated as 'very soft - very hard'.
Deformation
- Tendency to deform when apply with hard force. Rated as 'low high'.
Disintegration
- The tendency to disintegrate when apply with hard force. Rated

as 'low - high'.
Before drying - Evaluate by peeling bonded paper immediately after two

surfaces are placed to contact each other.
Stickiness
- Ease of taking the stick away from the paper at the end of
rubbing. Rated as 'not sticky - very sticky'.
After drying
Shininess
- Shininess of the stick after use. Rated as 'very duJl - very shiny' .
Delamination of
paper
- Evaluate by peeling after adhesive dried out (30 minutes after

placing two surfaces to contact each other).
- Degree of delamination of paper after peeling the bonded
papers. Rated as 'none - all',
tv
N
......
APPENDIX 3.3
GLUE STICK SENSORY TESTING
NAME
__
__
__
__
__
__
__
__
__
__
__
__
__
_
DATE
A. Wind up the stick and rub the stick along the surface of paper and spread adhesive to
cover the area to be bonded. Then place the coated paper on top of another piece of paper
and rub repeatedly with fingers. Do this for all samples to be tested and keep these papers
for the final test.
__
__
_
__
_
samples according to the following order:
B. Rub the stick along the surface of paper and spread adhesive to cover the area to be
bonded. Score the product for 'sensory attributes while applying glue stick' on line scale
in the questionnaire. After application score attributes for 'glue residue on paper' attributes.
PRODUCT SAMPLES
SENSORY ATIRIBurES WHILE APPLYING GLUE STICK
INSTRUCTIONS:
You will receive a number of glue stick samples. Please evaluate the
Please use the sample as the following instructions and evaluate each sample for the specified
attributes then place a mark (X) on the scale at the point representing the perceived intensity
of the attribute along with the sample number.
Slipperiness
drags
slips
Hardness
EXAMPLE
very soft
Th ic\mess
very hard
Deformation
very thin
very thick
low
high
Disintegration
low
high
GLUE RESIDUE ON PAPER
Degree of coverage
none
total
Thickness
very thin
very thick
Visibility of glue trail
invisible
very visible
C. Use a new piece of paper. Rub the stick along the surface of paper and spread adhesive
to cover the area to be bonded. Place this coated paper on top of another piece of paper and
rub repeatedly with fingers. Score the product for 'effect on paper' attributes.
EFFECT ON PAPER
Smoothness of paper
very smooth
very wrinkly
Cleanliness of work
very clean
very dirty
D. Use a new piece of paper. Rub the stick along the surface of paper and spread adhesive
to cover the area to be bonded. Place the coated paper on top of another piece of paper, try
adjusting the position of paper then rub many times with fingers. Score the attribute for
'stickability
attributes.
STICKABILITY
Adjustability
very easy
very difficult
Bond strength
peel the bonded papers as soon as the rubbing finish
very strong
very weak
Please come back to do the folowing part after you finished evaluation every sample for
the above attributes in order to allow about 30 minutes drying time
F. Use the bonded papers prepared at the beginning of the test and evaluate the following
attributes.
Bond strength
peel the prepared bonded paper
very weak
very strong
Delamination of paper
none
all
APPENDIX 4.1
4. What are the characteristics of those products you
think should be improved?
QUESTIONNAIRE
I am a Ph.D student studying product development at Massey UniverSity. I am

gathering information to use in the development of a glue stick product.
Could you please answer the following questions?
NAME
DATE
PART n
_
_
_
_
_
_
_
The products which you received are a sample of commercial glue stick products and
a sample of self-adhering note pads available on the market. Please use them in the
same way as you usually use these products then answer the following questions.
PART I
1 . Which of the following products have you ever used to stick paper together or to
other materials or for other purposes?
(
(
(
(
(
(
(
(
)
)
)
)
)
)
)
)
Glue stick
Liquid glue
PYA glue
Paste glue
Cellotape
2-sided tape
Self-adhering note pads (e.g. 3M's 'Post-it' note pads)
Other glue products (please specify)
2. Which products in Question 1 do you use most often and how often?
Please select from the following frequency:
Every day
3-4 times a week
Twice a week
Once a week
Once a month
2.3
Product
__
__
__
__
__
__
_
__
__
__
__
__
__
__
__
__
__
_
3. What do you use those products for?

Product 2.1
Product 2.2
Product 2.3
glue stick to
buy.
1.
2.
3.
4.
5.
6.
7.
8.
9.
10 .
11 .
12.
13.
1 4.
15 .
___
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
_
___
__
_
__
_
__
_
__
_
__
_
__
__
__
__
__
__
__
__
_
__
___
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
___
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
___
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
___
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
___
__
__
_
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
___
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
___
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
-------------------
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
2. Please list the characteristics that are important to you in evaluating a self-adhering
note pads to buy.
l.
Other (please specify)
2.1
2.2
1. Please list the characteristics that are important to you in evaluating
Frequency
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
1 4.
1 5.
IV
IV
PART m
1 . The following phrases are attributes of glue sticks.
Please rate the importance of these product attributes, use the score from 0 to 10;
o '" not important to 10 = very important.
___
___
Size
Colour of glue
Nice perfume
No awful smell
The thickness of adhesive film left on the paper when applying each
coat of the glue
Drying time
No curling or wrinking of paper after applying glue
Brand name
Price
Ease of use
Versatility
Keeping quality
2. The following phrases are attributes of self-adhering note pads. Please rate the
importance of these product attributes, use the score form 0 to 10;
o '" not important to 10 '" very important.
Colour of paper
Uniformity of glue coated on the paper
The force needed to pull the paper after it is stuck on other paper or
other materials
Damage on other surface after pulling the paper away from that surface
Brand name
Price
Ease of use
Versatility
Keeping quality
Thank you very much for your cooperation.
tv
tv
<.11
APPENDIX 4.2
Questionnaire for Consumer Survey in Thailand
flO I ' iu I Ru ' nunliHanwnl' 1flU I nl : nl'UnJ

L ; UU
fl eUUUuUOI
, i a nw: afl e Haflnwnl'
uunuolnIUiUn L UUa'U e ifl nl ; U L U ' nU " H aflnnl ' n "
flOI I u ' nunl, u a : uUU J H a flnWn l ' U n J nfl O I '
i n l n U H afl iuU : L nfiinu ifl n l ; i un n'uunAnlnlfl; IUIHafln
J H a fl " n u u uuu : a J ( Ha flnw n )
I n u l ui U: L nfin,aU H. ; U n n l n I U e. a L U ' nUnliHaflnWnl'
flOI I u ' nunlIHa flnwnl'UnJnfli ' fl I , 4 H aflvln
a : eft L U L fi u 'nUHaflnWnl,un = i n n l lu UU u i nl ; Un
U U uuu: a ( H aflnw )
flOI l fi u ' n Uflnuuuueuol
*
, u , nIU'Uft 'il n uuUUanuo ln l uufi,e' J e J n l ' U n nflflOII

n'il U ' Un'il J fl a l fl n 1wla n Jinl,n l 'ilU : n 'ilUUnl fl'ilUUUU'ilUOI
1) .
L rl u e n l 1 u a : Ua J L ri ui J IU I i u , 1uun n l ' n nWUflnlnl ,nU
ft ei 'n u i I . ' ' L n U : l nU : 'ni n : fllU ; U J U ftnflfl, ' J , u H aflh. :
'unl UUUJ inW u n , n l ' U n J n U U I Un e J fl a l fl u a : . : n l fll n n 1 1
1.
HaflnWUI J J 'il iuQnlu l fluiun l fln ' : fllnun : I lli 'il n u;u
u a : nlu i H a flnW J na l ' U'il u u fiiu
L du " ni;flquH afl'UU : l nfiUn l H a fl
2) .
Haflnwnv : n ft nW: I neun l , u n J" U U . : nWU ' J U U niufiu l ri ui
flfln : fll I ,'Unu i u fln : fllnu; u U ' 1I o U U n ' flfliiunnlu
u U J l ilu n
fl J n l ' U n n : nW Ufl l llnUUnUnl l l flft eUUUn : I iu fl n n l ' U' l iu n : flliufl

a L U " ift-n" ( Post- i t ) U 3M i J IIouiuflUU I Un l i: ue{ ; U
n 'ilfl l nU
HUJ iflU L ri u uun 'n{, l nflfl" L aUI U n UUH 'U n : fl l iu uUn" nQH aflhw .
L Ylun l 1 n 2 '\'Iih
U I I O iftfl1 UUnl n I U JIU O,u l un , 1 n 1 J I I O u n : D e n u a ' flflUU J i
U ( iu : l! )
fl' fl n n, i fl U ' nn : flllfli e l uu u u

naiUfl n . : u,uiun l n " WI I H aflnwi 2 nfl' v : 'unl
nlu a : uiuu1 U' J ' ' : l uuu : iu u u ' J lna U J lu ; u uau u iu ; U J
n.
y!n;u
fliefl,,i, nuvlnnlu1 W ij,u
aUflll1a :
fl .
aUflla: 2 fli J
J.
aUflla: 1 fl i J
I 'ilua : 1 fli J
U ( iU fl : 1! )
..
...
nU"H1U
: .... ..
I'
L 'il 'ilnnn
3-4 fl J
.
.
2.
n,uiaHaflnw na,,ua ( 1 ) iu
3.
5.
5,n,u11ioHanwna" 1Ufl'n,u l o iUflaUfiO'uO ( 3 l
n 1 w'U1 : I Uufi1,unD'. D n'UOHftnW0 1ud 'lJanw: nu

( o u I n' : H anwn,u l fiu'i )
( )
;'U'U l fl a l uu
( )
Hun'U l fii a l uuU' ffnaufl'
1 dJuu,no.
;,U'UO"IS
flouii' 1 l nUU'no'.
"l!U I UO-llJ,-l l nfl
I nUu,no.u,un,
tu
<J
( iu : lJ )
n,u1;UHanwna" lJ'Ou' 1
..
flOU' l nUu,no'.
I nuu,,,, o hlJ,n
n'u { 'unmn,u'ta;u,u'uftn 'u )

u
4.
<J
( iu ; lJ )
__
__
__
__
__
'unl l a on'tnI1 1 Dn1; I anW: Dlunfi1IlJa'UnI1aU" DnIU

au' 1
1.
oau
2.
fi1'lJall l auaan,,
[ ]
111fff\'Y
fla\lii, 111fff\'Y
afi'Yu,umn
ti()\I' fff\'Y
4.
1 ''fl o o'n"
[ ]
af\'YlJ,n
5.
Hn : nufion : ' l 'lIU
[ ]
[ )
[ )
[ )
( ]
I flaouuun : '
3.
UlJ'Wn" fla'i'un'
[ )
[ )
[ )
O : ' I ,,unu
n: 'uu l' O O
( ]
nau
6.
fl1,U,u, o'un'1
[ ]
[ )
n: '
( ]
u;u'wn" flo'tun'n : ' l ii,'unu
( ]
1 ''flo o'n"
7.
fl1'lJa: o,'un,'t
8.
1 ,fl'
[ )
9.
fi1'lJff: 1n'un,'t
[ )
( ]
10 .
fi1'Uff'lJ' O'\In,'t
[ .]
[ )
nlJ1aI,u"lSil
( ]
( ]
( ]
11.
wauuft'un' l nu11
1u'\Iiuill l ouwnln
[ ]
[ ]
[ ]
[ )
6.
n;; ,
ilJ;;
7.
n1WVUfin,v (7)
ft nw n . n 1un'Yfln 1 W ft n u U uau:
n ilJ;; , n1WflVUfi n lV (10)
n 1W nUfin, n iU L nU iuun , u u : ui uU1 ft nW n .
1.
nu;;n ' 1 u n rl n n : i
un L a n n i n n 1 un n,ufl,nw: n iUn 1 1afl n n , u n nu

n u 1 n1W fi : UUU U 1 O 5 u n [ ] nu
1lJl!"1fl
nlnu , n1W'flVUn nu 2
t\nu, ilJafl
n1lJnu , n1w'vun ln (9)
a#ity1J\In '
t\ nu'ih a'1fl
afltylJ'n
8.
nu
9.
[ ]
"
11ununnI1
1 alJ L alJ n n n ' 1 L a nUUUn :
,
( n1w, nunnU2)
lO .
1 1 u n n ' 1nuun : ' L nn, n' 1 u a : i
n1W' L n'1nuilJ;;n1Un
1J;lwn1 n ;;un'n : ,
L 1,n nn'1U
1 1 a : nluunnnn'u : u
1 1 a : n ' n 'u
n lnlufl I nw: uu afl i1J : nnUfi : UUUfi" lJfl n U fl a :
nw:
IV
IV
00
u u : uwnwUnuuD nu u 1
' : DUUDU/ ' : U : UuuuU
U : U / U : U : U
1u.
U : 1 D / u : 1u u : u
1DUUuu/1u: uUUDU
*
nU L a D n uu1uuuuuu / 1 u u : uUUUU n 1W L W ft l 1uuun

uuli 3
A
4.
. n 1 u nu;W flnW n . l 'l' jju L n

u A
( \lun L U!) n l
9 n7 )
u B
( \lun L UU n l
18 n 7 )
u C
( \lun L UD n l 9 n 7 )
u '1 ( 1u'l''l' :: 1J )
5.
ii "' W a fln
n.
_
_
_
_
jju fl L n .. nlJ;;1D! h "il n i1 ( \lun L U D n 1 9 n7 ) nufl;
"il : i1uu L n n DUnUfl D !)U U L U U u n
U'l' : tII 6 L tiDu

U '1 ( 1U'l'fl 'l' :: 1J )
_
_
_
__
_
6.
O ' H 8nw n .
O, \ n,nU'DU' U vn ( Un \ dDn,. 9 n i u ) n,fl,
' ' I ': iuHanwdfl ' I U L n,

Dun, 1 0
U,l)
[ )
fl l , a'lJ' On,, i1'na'Ufl
10
U,ll
[ )
u ' a in,' n ' : 'H8anv'nuI'1n : , I\J
12
U,l)
[ )
fll'lJ L UMU L UHUMi a ' D flqMa v,n n : Dan
15
U,ll
[ )
fl l ' lJa : l nn,';
18
U,l)
[ ]
fll,a," on,inU1 aM8'UU
lJ , n n ,
m)
fl" lJ'lJ on ,n ' : ,
1 8 U,ll
O ' fl , n ft nHW: afity 1U' : n ij lJ nU fl : Ufll ,'1fitVD U a : ftnw:
H a nN . n" 4n U'fl " H a , nuUHU: ft

H anW .
L UHanWu 1Ou'UnD a, n, .dOWalJUfl L O D
n11 fl i ftM uuun" 1'uftDnl1U' , U n ' t fl ''Uft ' M6 a H'na- Dn ij 3M n "
[ J
d l lJ, : aiui L nn' ' fl "
[ )
L i n : ,'Gu L ijna, ,: uij4 H U
,a,
n W'DUflO'lJ ij 1Ufi L ijU : 1u uun,n' H anW .
3.
1.
1';
2.
n'fl , : a' lJ " o';Hanw

1
iJD )
n' L D n '; n , . u n u ' fl "
L D 1 qU : fl D : 11'U' ( L a D n DU1'lJ,nn,
n' : '1Gu L Dna, U D

n' : ,1G ij n,' UUUij,i ijHU
1\Jun' : ,
( O ' H anwd1iun, n, u a : Ovu, u
n D a , n'fl ,anw: ij IUfiOfl" U afl ij n , ftv D n' D u 1
n : ' L .a' D D
n1W' fl : Ui u 1 5 Ui ij [
i)HUm ! D ';j l1lJ,U
] n
( 'U ' ' : )
__
__
__
__
__
__
__
1i1afi
flDih 1i1afity
a'1fitV\J,na,
v : i au UD/ v : U : '1UUD
flDih a'1fity
Dlv v ; i D / a,v v : u : u'1
a'1fltylJ,n
1i1uuv
4.
O'HaflnUl
'\i;,un" nmn u a : nv'1l1U'UnD a, n,u , ';j : i ijHafliiwfi l1 a
u:Ha flnwdnu1l'1 'D n'UM ij 1 i1
D ' v v ; 1 i1 i a / ij , v v ; 1u ; u'1
1i1iDUUij/ 1u ; '1uUij
..
n ,n,u L a anDu1iDuUD/ Ii1u: '1uua
N
W
o
5.
O,afln . U' \ n'nUfl'nU'n ( uUn \ nn"
n ) n,ufi;,
fln\!
: iu,u \ n' nnun,un in i n \ uRuun
u : lnlll
in
_______
U,ft
1.
'I1iV
,;n-h 15
2.
u
6.
( 1u : )
19-25
-----
O'afln . U' \ n'nU'nu'u n ( u'I1Un \ nn"
26-40 il
40
un \ ; uu 1 ,JI ; u ____
3.
unn, ftOIU
(- )
''1'
U,Yl
12
UYl
( )
''nl n
18
U'Yl
( )
Ylun TUU%Yl \ lln Yin "I U
_____
: nu';nHafln na" llIU"In
U
1" \ uultJ1 \ au
fl" \ ulul9\
fl" \ 1tJltJ'\ln "I
fl" \ ululfln\l '
'''1 \ IU1 'n

1
( )
__
__
__
__
_
: U
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
___
__
__
__
__
__
__
__
__
__
__
__
_
fl'11
n"lU;'" \ IUIHaflnlll
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
___
10
'nn; 18 U,Yl
7.
15- 1 8 0
'nn-l'
n ) n'ufi;,
,, \ ': ftnHaflnlll' \ U \ n'
------
1 tJ: )
__
__
__
__
__
__
__
__
__
__
__
__
__
__
__
_
__
__
__
__
__
__
__
__
__
__
__
__
_
__
APPENDIX 4.2
(continued)
QUESTIONNAIRE
(Translation from Thai)
PART !
1. Which of the following products have you ever used to stick paper together or to
other materials and how ofu!n?
Frequency
The questionnaire you have received is part of the research project on 'The
Development of Glue Stick Product from Thai Tapioca Starch'. This project is conducted
o Glue stick
o Liquid glue
o Latex glue
o Paste glue
o Cellotape
o 2 sided tape
o Other glue products (please specify)
by a Ph.D. student in Consumer Technology Department, Massey University, New

Zealand. The information on glue product usage and your opinion regarding glue stick
products will be used in further development of glue stick products.
The product to be studied are:
1)
Product A
(see Note below)
Permanent bonding giue stick made from tapioca lour
nus glue product is in stick form and comes in a lip stick like container. The glue can
be wound up before using and wound down after finished. The advantages of the glue
stick over other glue products are it is easier to use, dries faster, not messy, does not
cause wrinkling or curling of paper, easy to carry. The product will be developea to
overcome the defect of the existing giue stick and the price will be cheaper because it
will be made from raw materials available in Thailand.
Note:
Please select
frequency from
the following choices:
a) Every day
b) 3 times
c) Twice a week
d) Once a week
e) Once
month
f)
week
Less than once a month
g) Not used
2)
Product B
Temporary bonding glue stick made from tapioca flour
Product B is a new type oi glue stick which is nor available in the market This glue
product will have the general properties similar to those of the conventional glue stick,
except that after the glue is used to stick paper together or to other materials, it enables
the paper to be peeled off from the receiving surface for a number of times without
damaging or leaving glue residues on the receiving surfaces. This glue has the same
property as the glue coated on self-adhering note pad such as 3M's 'Post-It' which can
be attached to note, table, board or wall. This product can also be used for reportS and
photocopying etc. in which the paper
can
Your participation in this survey is vital to the decision on the development of these
Thank you very much for your cooperation.
are
(Ii you
o
o
o
o
o
did not buy the product yourself, please go to Question 3)

Stationery shops
Stationery department in department stores
Comer shop or little shop
Supermarkets
Other (please
specify)
be repositioned without any damage.
two products which will be very valuable for this project.
The questions in this questionnaire
2. Where did you buy those products?
divided into 4 parts:
Relate to glue products usage especially glue sticks as well as

important attnbutes of glue products
Relate to the development and improvement of permanent
bonding glue stick product based on tapioca starch
Relate to the development of temporary bonding glue stick
product based on tapioca flour
Relate to personal information
Go to Question (4)
3. From whom did you get those products?
o Your office
o Other (please
specify)
4. When you ust;! a glue 'to stick pieces of paper together, how important are the
following to you:
Note: Please use score from 1 to 5; 1
not important to 5
very important
Odour
___
___
___
not satisfied.
Drying time
Effect on paper after applying the glue
Stickability
Go to Part II
Price
___
are
Cleanliness
___
If yes, please go to Part II

If no, please go to Question (9)
D Yes.
o No.
9. Please specify any reason why you
:== Uniformity of coating
___
8. Are you satisfied with current glue stick(s) .thaI' you use?
Ease of use
Versatility
Keeping quality
10. If you do not use glue stick, please state reasons.
5. PieaJe rate the following products on the attributes provided

below:
Note: Please use score' from 1 to 5 ;
Odour
to stick paper together
Drying time
Effect on paper after
applying the glue
Stickability
Cleanliness
Price
Ease of use
Versatility
Keeping quality
unacceptable to 5
Glue
Liquid
stick
glue
o No
very acceptable
PVA glue
Cellotape
0
0
0
0
0
0
0
0
0
0
0
0
PART II
0
0
0
0
0
0
0
0
The product which you received is a sample of a commercial glue stick available on the
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
6. Do you use a glue stick regularly?

D Yes
If yes, go to Question
Product A:
Permanent bonding glue stick based on tapioca flour
market. Please try it now to stick some paper together as you would usually Use a glue
stick and answer the following questions.
1. The following are attributes of glue stick products. Please rate the importance oi these
product attributes, use the score from 1 to 5; 1 = Not important to 5 = Very important
SCORE
Thickness of adhesive film left on the paper when
(7)
if no, go to Question (10)
7. What brand(s) of glue stick do you use? Please specify:
applying each coat of glue stick

Drying time
Effect on paper after applying the glue
Stickability
Cleanliness of container
Cleanliness of work
Please specify if there are any other important attributes for glue stick products and
also assi gn score for importance of each attribute.
5. How often are you going to replace this product, if you use the small size glue stick
8 grams? (Please estimate)
a once
a Once
a once
a once
SCORE
a month
every two months
every 3-4 months
every 6 months
a Other (please specify)
2. What attributes do you think should be improved in this product and specify how
you would like those attributes to be improved.
_
_
_
_
_
_
_
_
_
_
_
_
_
_
_
_
_
_
6. What do you consider a reasonable price for the small

size product (8 grams)?
a less than 10 baht
a 10 - 12 baht
a 13 - 15 baht
a 16 - 18 baht
a more than 18 baht
PART III
3. Would you buy this product or recorrunend it to your office if it was improved
suggested?
a Probably
a Might or might not
If not, please state reasons then go
Product B:
Temporary bonding glue stick based on tapioca flour
This product can be used as glue for temporary note, the same purpose as a self
adhering note pad e.g. 'Post-It' note pad, which can be adhered to a receiving surface
such as paper, notice board, des\<-top etc.
a Definitely
o Probably not
a Definitely not
as
1. Do you use self-adhering note pads e.g. Post-it note pads regularly?
to
PART ill :
a Yes If yes, what brand do you use?

a No
_
_
_
_
_
_
_
_
_
_
_
_
_
_
_
2. The following are some attributes of this temporary bonded glue stick product.
Please rate the importance of these product attributes, use the score from 1 to 5,
1
4. What is the size of the product that you prefer? (Please see the Figure provided
a Size A ( 8 grams) - as per sample provided.
a Size B (18 grams)
a Size C ( 8 grams)
a Other (please specify)
Not important to 5
Very important.
Stickability
Reattachability
Force needed to pull paper from the receiving surface
Damage on the receiving surface after pulling the paper awa
from that surface
Ease of use
Versatility
3. Please specify if there
are any either important attributes for such a product and also
assign score for importance of each of the attributes provided.
8. What do you think of the possibility that this product will replace the following
products:
1 .. No possibility
2 .. Low possibility
3 Moderate possibility
4
High possibility
5 .. Very high possibility
SCORE
4. For what purpose would you use this product?
1
o
o
o
o
o
o
o
Glue stick
Liquid glue
PVA glue
Paste glue
Cellotape
2 sided tape
Self-adhering note pads
o To stick note pad on paper or other materials

o To stick paper or card on notice boards or walls
o To stick paper to paper
o To stick paper for parcel wrapping
o To seal envelope
o Other (please specify)
2
o
o
o
o
o
o
o
PART IV
5. Would you buy this product or recommend it to your office?

o Definitely
o Probably
o Might or might not
o Probably not
If not, please state the reason then go to PART IV:
o Definitely not
6. How often are you going to replace this product. if you use the small size glue stick 8 grams? (Please estimate)
o Once a month
o Once every two months
o Once every 3-4 months
o Once every 6 months
o Other (please specify)
_
_
_
_
_
_
_
_
_
_
_
_
_
_
_
_
_
7. What do you consider a reasonable price for the small size product (8 grams)?
o less than 10 baht
0 10 - 12 baht
0 13 - 15 baht
0 16 - 18 baht
o more than 18 baht
Sex
Age group
Occupation
o Male
,0 Female
o 15-18 years
o 26-40 years
0 19-25 years
o more than 40 years
o
o
o
Undergraduate Student
Post-graduate Student
Governmentofficer
Thank you very much for your cooperation
3
o
o
o
o
o
o
o
4
o
o
o
o
o
o
o
5
o
o
o
o
o
o
o
APPENDIX 5.1
APPENDIX 5.2
Correlations between Ingredients and Physical A ttributes of Glue Stick
Physical attributes of glue sticks from casein experiment
from Dextrin Experiment

Wet glue per
Sample
Attributes
Ingredient
Hardness
Starch
PVP
Peel strength
Significant at p
<
Hardness
(Newton)
Open time
(minutes)
Peel strength
(Newton)
35
Cl
39.5
21.2
44.6
5.6
5.5
0 .227
C2
36.8
20.7
40.5
6.2
55
3.2
-0.546
C3
26.2
15.3
45.8
6.7
53
3.8
C4
38.9
20.9
46.0
6.4
23
3.7
44.0
5.6
3.8
3.0
455
4.6
3.5
3.5
Dextrin
0.391
C5
39.8
21.6
Starch
0.450
C6
37.0
19.8
PVP
0.293
Dextrin
Note:
Dry glue per

area (g/m')
%Moistwe
area (g/m')
Sensory Attributes Mean Scores of the Glue Stick Samples from Casein Experiment
-0.910'
(Scores Varied from 0-15)
0.05
SAmple
CA I
CA 2
CA 3
CA .
CA S.!
APPENDIX 5.3
Correlation between ingredients and physical and sensory attributes of
glue stick
CA 5.2
U>YU
'Thla
Vib&e
50-'> CleM
""I-
SUp
.....
_CRI\Ctftl
...
H.,d
DeIonn
DWntesr.8ond
tldoa.......,.n2
10.1
10.9
10.1
11.4
1.3
3.7
7.7
10.5
10.0
73
(1.1)
(2.2)
(UI
(I."
(10,
(11 1
(Ul,
(2.',
(I")
(13)
( 1.9,
5.9
10.3
10.3
lo.s
11.1
',2
.3
15
...
1 1.0
11.4
(1',
(1.1)
(1.9)
IUl)
(1')
(I."
(1.2,
(1..5)
113,
(0.7)
(U)
6..
(1')
(1.7)
6.
' 8. 1
..5
10.1
7.
5.1
...
9.1
9..5
8.9
.2
(2.11
(U)
(I.lI)
(2.2)
0.7)
(U)
(U)
(:I-II
(2.2)
(1.1,
(1.6,
(2.2)
11.
(1.1)
U
113)
83
(2.2)
(1:6)
U
(1.11)
7.2
.7
.8
10.2
(13)
(1..5)
(14)
1111
9.2
8.'
(2."
(I.,)
11.1
3..5
9.2
Note:
Output Variable
Glycerin
Perceived hardness
AdjustabiJity
%Moisture
Delamination
-0.8340. 737
-0.731'
0.730"
Sorbitol
Perceived hardness
Adjustability
%Moisture
Delamination
0.834-0.737"
0.731"
-0.730"
Note:
- Significant at 0.10 t! P <: 0.05

- Significant at p < 0.05
(2.2)
27
(1.61
2.J
10.3
10.3
'.7
(1.9)
(1.1)
'(1 .,
8:1
..
112)
(2.1)
9.6
(l.2)
5.3
).4
(13'
(11)
(1..5)
( 1 .6)
8.9
11.0
Ill.
(2.11
(1..5)
(I."
(1.3)
'.9
(13'
'-'
(1<)
'.7
3..5
.3
.0
7.3
5.0
(I."
(1.1)
(1.1)
(Ul)
(1.9,
(2.'1
(1.1)
8.1
83
8.2
10.7
11.2
(1-"
(1.1)
(1.6)
(1.9)
(1.9)
7:1
from casein experiment

Ingredients
IU.O
(1.91
The numbers in the parentheses are standard deviations
237
5.4
A P P E ND I X
C omme r c i a l
PEL I FI X
BOST I K
AMO S
UHU
B l ue
St ic
st ick
( 8g )
( 9g )
E S SELTE
( 8g )
St icks
g l ue
Glu- s t ik
Glue
P RI TT
G l ue
st ick
( 9g )
P e l i kan ,
( 8g )
Bost ik
( 8g )
Amo s ,
UHU
Ge r m a n y
Pty
Ltd . ,
Korea
Gmb H ,
H e n ke l ,
E s s e lt e ,
Ge r ma n y
Germany
Spa in
Korea
APPENDIX 6.1
Questionnaire for Glue Stick Testing by Consumer Panel

UUUflnounI " U
;;oijflnou . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
;\JY1 . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1.
nTWOIJ7U
'\IJuolJiu
2.
fl1'IJ91 '\U;;o I u\J:\J" '\;;Y1Yl''''" '\J'lJtl'\J;;tl
;;tlUUUtlU
1 .
.
uun ' = fl'Y1Uijn'\;; l tlti; U'\;;UY1'91fl'IJtl U n " ' ; U U , nlJtl\J
3.
UHU
2.
mltl'\n" lflu'n' ,u\Jn' :fl'U \J I n'\;;Yi'ui I ']UlY1Vi tl n')'
u ii ']\J" '\tii'lflU\Jf\, = n
10
3.
ti, : I uu ii,u'iUlWnlJ;;nll= n" HUfl'\;; l U;jfl l d\Jm (
)iifll'i U l d\Junfl ;;IlUI: n" Ufl
12
14
16
18
20
III
" uHUY1m liUI1 1 ' lJ i ffn'IJtl ';nu "; tlIJYi 'l1rt;h o i n
256
6 5-9
385
IV
UJ
00
n'!llnnflt'lOUn'! = fl,\jnn,n T l l1 t1 :n'hlnT l l1 l1 U 1 )i , ,, ; UlJ'\l1 lflUnl ' tlonn, :fl1\jYiIfO

2 .
uln
\HlU
'1lU
3 .
lJJflfl ' tlU

.
4 .
n,'!tli u , iitlUn H fl l\jnn11l1111 .i1'\l1, lnflllJ;; tl n 1 '!
tliuu, n
l JJ , iuu
.im:'!II : Onl'L,f1mnU;; O I1I '! :;;0
UU L
;ruu t'li\ "n=U""j)umnno
(continued)
APPENDIX 6.1
Questionnaire for Glue Slick Testing by Consumer Panel

(Translation irom Thai)
Part II
Please test the glue stick samples then rate the product on the following attribues.
QUESTIONNAIRE FOR GLUE STICK TESTING
1. Slipperiness
Date
Name
_
_
_
_
_
_
drags
Method of testing
2. Deformation
2.
Wind up the stick about 5mm above the top edge of the container. Try rubbing glue
stick on the paper before testing in order to smooth the glue stick surface.
Rub the glue stick along the surface of the small paper then stick this paper on the
3.
Evaluate the samples according to the given attributes and rate the sample by
1.
slips
big paper at the specified positi9n.
low
high
3. Degree of coverage after one coating
marking on the scale at the point represent your perception of the attribute.
none
4. Adjustability of coated paper to the specified position
Example
Overall liking
dislike
265
very difficult
385
r
659
like very much
Part I
very smooth
After testing every sample, please mark 'I' at the position represent your ideal product on
the scale' for every attribute.
1. Acceptability
Part m
very acceptable
not acceptable
Please test the prepared bonded paper by peeling the paper apart
2. Purchase intention
certainly buy
never buy
8 baht
very easy
5. Smoothness of paper after sticking the two piece of paper together
very wrinkled
3. Price to buy
total
easy to peel
UHU
6. Stickability
20 bath
stick very strongly
After testing every sample, please mark 'I' at the position represent your ideal product on
the scale.
Thank you very much.
APPENDIX 0.2
Note,
742
557
428
RUN 1
RUN 4
RUN 7
Sensory Attributes Scores of the Prototypes Obtained from Consumer Testing
243
181
RUN
Run
314
887
RUN
Accept.
S l ip
Run
Adjust
Acceptabi 1 i ty
Smoo ch
SAMPLE
PANEL
Purchase
O i s i n tegr
S l i pperiness
Adjustabi l i t y
purchase Intention
Oi s integration
Price
Price to Buy
Degree o t Coverage
Coverage
Stick
Smoo thness
S t ickabi lity
KABet.art University students

SAMPLE
PANEL
742
742
742
742
742
742
742
742
742
742
742
742
742
742
742
742
H
742
2.2
2.5
2.8
7.2
11 . 9
8.2
2.1
1.3
1.8
9.1
8.4
6.7
4.7
10.4
10.5
5.4
.8
3.2
.8
2.4
11. 0
6.4
1.8
1.1
.5
9.6
6.2
5.9
3.3
9.9
11.1
6.2
IS
11. 9
1.2
19
20
21
22
23
24
25
26
27
28
29
30
MEAN
PANEL
243
243
243
243
243
243
243
243
243
243
243
243
2(3
243
243
243
243
243
243
243
243
243
243
243
243
241
243
243
243
243
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
IB
19
20
21
22
23
24
25
26
27
28
29
30
MEAN
so
6.8
8.6
2.4
8.1
5.0
2.2
5.8
4.5
.2
1.5
3.8
1.8
7.5
5.51
3 . 2'
SO
SAMPLE
PURCIlASB PRICE
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
742
742
742
742
742
742
742
742
7 42
7.2
742
742
ACCEPT
ACCEPT
5.6
9.3
13.1
2.1
13.3
10.7
2.5
9.1
3.6
6.8
4.5
5.7
3.4
6.1
7.3
6.2
3.5
12 . 6
12.9
1.0
12 . 5
6.3
6.9
5.1
3.1
1.4
4.1
5.5
4.5
11.6
6 . 68
3 . 7'6
6.8
10 . 1
2.7
7.8
3.0
.7
6.2
2.8
. 1
1.7
3.0
.8
4.1
4 . 57
3. 4 5
PURCHASE
2.6
11.5
10 . 6
1.4
12 . 7
10.9
2.2
9.1
1.6
6.9
),3
5.2
2.3
4.5
6.8
10 . 1
1.7
13 . B
10.7
1.0
11 . 9
3.8
7.0
5.1
1.2
1.3
4.7
4.2
3.1
5.9
5 . 90
4 . 00
SLIP
OIS INTEGR
COVERAGE ADJUST
2.4
.1
0.0
0.0
5.1
4.0
0.0
0.0
.4
6.2
4.7
4.8
1.2
5.0
8.5
3.3
1.7
1.2
7.5
4.6
12 . 8
9.8
2.3
.3
7.0
6.8
8.5
3.5
3.8
10 . 4
9.5
2.7
14.0
13.0
12 . 3
8.8
11.4
4.8
13.0
12 . 4
6.8
8.1
4.1
8.6
6.8
3.8
2.2
11.9
.9
10 . 9
8.4
7.4
13 . 1
8.9
2.8
2.9
12 . 3
3.6
8.9
8.9
6.8
9.7
9.8
11. 3
2.4
9.3
2.7
3.9
2.9
10 . 2
5 . 1
1.8
3.3
.8
.9
1.3
2.0
6.2
14 . 0
8.8
8.5
6.0
6.'
4.4
12 . 7
5.3
13 . 5
13 .0
13 . 1
13.3
8.9
8.1
10 . '
10 . 7
6.4
8.5
5.3
3.4
4.9
0.0
2.4
.9
3 . 5
4.9
0.0
0.0
0.0
0.0
.1
2.3
2 . 35
2 . 38
PRICE
4.9
2.4
4.9
0,0
7.2
6.4
0.0
7.4
1.7
6.8
2.6
4.8
1.2
2.4
6.2
5.3
.B
4.8
5.6
0.0
7.5
1.6
4.9
4.9
0.0
0.0
2.5
0.0
.1
2.3
3.31
2 . 62
9.J
5 . 07
3 . 57
SLIP
5.0
10.8
10.2
8.4
12 . 2
10 . 2
3.2
9.7
10 . 2
12.9
6.3
10 . 3
5.7
5.4
11.0
3.0
.3
10 . 3
6.2
1.6
12 . 1
8.6
11.5
7.4
7.3
9.1
8.9
8.1
9.3
6.8
8 . 20
2 . 94
9.1
9 . 30
3 . 61
OI S I NTEGR
7 . 8
13.8
8.2
1.7
ll . 9
7.8
10.5
13.8
4.B
10.7
4.7
9.9
9.4
U.S
2.0
13 . 6
11.5
14.9
13 . 6
11.9
1.2
9.6
12 . 9
9.6
13.2
12. 1
14 . 7
12.2
14 . 9
10 . 5
10.23
3 . 95
6.7
1.3
4.4
3.2
2.9
3 . 3
3 . 3
13.4
7 . 15
3 . 62
12 . 9
12 . 8
10 . 7
9.8
12 . 9
8.9
5.4
1.1
3.5
11.4
12 . 6
9.9
4.2
12 . 5
13 . 1
7.
7.9
2.3
8.1
11 . 3
2.3
7 . 1
2.1
1.4
9.8
12 . 3
1.8
9.3
9.2
4.7
SHOOT!!
1.4
11.8
10 . 1
2.0
11.3
12 . 3
2.1
.5
4.6
9.9
11.2
10.4
8.8
11.1
13 . 3
3.8
9.8
15.0
13.7
12 . 6
7.
11.8
8.7
1.3
7.9
.3
4.0
3.1
9.2
8.3
7 . 96
4 . 08
7 . 92
4 . 44
COVERAGE .DJU ST
SMOOTH
4 . 1
8.5
9.4
6.5
12 . 7
8.6
3.7
9.8
12 . 8
6.5
7.5
9.6
5.7
B.3
10 . 5
3.9
5.1
5.8
2.8
3.0
13 . 9
6.0
7.1
B.8
5.7
9 . 1
4.5
9.2
2 ..2
5.9
7 . 24
3 . 03
. 5
12 . 2
5.3
8.5
12 . 9
9.4
2.4
11.2
3.7
7.1
12 . 0
12 . 7
3.2
12 . 5
12 . 1
11.2
7.9
11. 1
5.1
11.3
9.7
8.1
10.4
9.0
10 . 8
9.0
11. 6
5.6
13.6
5.7
8 . 90
3 . 53
7.7
11.0
6.1
10 . 2
11.0
13 . 5
3.1
5.9
5.3
9.3
10 . 7
12 . 2
12 . 0
4.6
13.3
2.6
9.B
13 .9
14 . 5
12 . 6
1' . 2
3.4
13 . 3
9.7
3.2
8.6
1.3
14 . 0
2.8
10.2
9 . 00
4 . 16
STICK
9.6
11. 6
12 . 2
6.6
4.2
14 . 5
9.6
2.9
ll.6
12 . 7
14 . 3
7 . 3
12 . 7
13.7
9.3
13. 8
U.S
4.9
9.6
10.6
13. 1
1 0.
10.0
7.6
12 . 5
13.2
11.6
9.7
10. 5
12 . 8
>lEAN
1 0 . 49
2 . 99
STICK
10.8
8.9
13 . 3
4.3
4.5
13 . 6
ll.7
4.9
11.3
12 . 0
12 . 0
9.9
13 . 7
11 . 3
9.0
13 .1
11 . 5
6.9
11.5
B.4
7.9
7.8
1.5
4.5
9.7
6.4
9.0
11.0
7.8
7.0
9 . 17
3 . 16
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
3 14
314
3 14
314
314
314
314
3H
3 14
31'
314
3 14
314
314
314
314
3 1.
314
3 1.
31'
314
3 1.
314
31'
31.
314
314
3 14
314
314
SO
SAMPLE
PANEL
557
557
557
557
557
557
557
557
5 57
557
557
557
557
557
557
557
557
557
557
557
557
557
557
557
557
557
557
557
557
557
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
2.
25
26
27
28
29
30
MEAN
SO
ACCEPT
0.0
6.1
1.2
1.2
1.8
4.4
0.0
.3
U.O
1.1
7.0
2.7
0.0
1.3
.3
3.0
.2
0.0
.5
.1
.2
.7
. 1
.8
.5
PURCHASE
0.0
4.4
.8
1.0
1.6
2.5
0.0
.3
0.0
1.4
3.9
2.7
0.0
1 . 1
0.0
2.0
.2
0.
.
0.
0.
1.
0.
.
0
3
0
0
7
0
.4
0.0
0.0
0.0
.8
.5
0.0
0.0
0.0
0.0
1 . 26
1 . 85
. 84
1. 21
ACCEPT
1 3 .9
1.4
4.5
7.9
13.0
13 . 4
5.5
12 . 2
11.8
13.3
5.9
B.O
12 . 3
12 . 1
9.4
10 . 3
11. 4
12 . 3
14 . 0
9.3
4.1
11 . 1
13 . 8
6. B
5.4
13 . 6
1.7
11 . B
13 . 1
12 . 6
9 . B6
3 . 81
.2
PRICE
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
.3
.4
.3
0.0
0.0
0.0
1.5
.2
0.0
0.0
0.0
0.0
0.0
3.5
0.0
0.0
0.0
0.0
0.0
0.0
.1
.28
.78
PURCHASE PRICE
14 . 2
1.3
2.0
6.2
12 . 5
14 . 0
5.4
12 . B
11.a
10.9
3.9
7.7
10 . 9
12 . 0
9.9
10 . 0
11. 6
12 . 4
13 . B
10 . 1
.7
12.9
14 . 1
7 . 1
3.B
14 . 0
1.5
9 . 3
13.1
10 . 7
9 . 37
4 . 33
10 . 0
.1
0.0
3.7
7.2
7.7
2.5
8.8
8.4
B.7
2.6
7.5
10 . 1
8.8
B.2
7.9
4.9
4 .B
B.8
4.8
.3
7.4
11 . 0
4.9
0.0
2.3
2.' 5
2.3
4.B
2.3
5 . 44
3 . 38
O I SINTEGR
SLI P
14 . 8
14.1
13 . 5
15.0
2.4
12 . 2
13.7
14 . 6
14 . 1
14 . 5
6.4
4.0
15 . 0
12 . 6
15.0
12 . 5
13.5
15. 0
14.2
15.0
13.6
12 . 5
15 . 0
14.8
12 . 3
15.0
13.0
12 . 1
14. 7
13 . 9
.2
1.2
.5
1.3
1.9
5.5
.7
.3
.8
3.3
4.7
1.2
. 1
1.2
2.5
1.5
2.2
.9
.8
. 1
.
.
.1
.2
.1
. 1
10 . 0
. 1
. 1
.3
.
.
1 2 . 97
3 . 15
1. 44
2 . 10
SLIP
OIS INTEGR
14 . 2
13 . 8
12 . 6
11. 8
13.5
12 . 2
10 . 1
12 . 7
13 . 1
11. 7
10 . 4
10 . 9
13 . 5
12 . 1
12 . 5
12 . 5
10 . 6
10 . 3
13 . 0
15.0
14 . 4
14 . 1
14 . 4
9.3
1.9
1 2 . 914.2
11.3
13.B
11.2
12 . 13
2 . 44
.9
1.0
2.0
1.1
10 . 8
1.9
8.6
.7
.6
6.2
2.1
2.7
3.7
2.0
3.0
1.B
),2
1.1
11.4
.5
.1
1.9
. 1
2.4
11.
.2
.1
2.1
1.7
3.7
2 . 97
3 . 32
CDVSRAGB
ADJUST
8.4
10 . 6
2.4
2.6
6.7
8.6
2.2
1.7
6.2
2.4
9.9
8.4
.1
1.4
6.7
3.2
1.7
.8
4.2
.5
2.2
.5
3.8
.3
3.2
1.4
. 1
. 1
.1
.8
.1
12 . 8
4.0
3.9
4.5
6.0
6.0
1.1
3.5
10.9
10 . 6
4.8
7.7
14 . 5
14 . 0
a.6
2.9
.1
1.2
13.
6.8
5 . 1
12 . 1
.3
12 . 0
15 . 0
15 . 0
15 . 0
15 . 0
1.0
SK)O'I'II
.1
10.3
3.4
9.5
7.7
ll.2
.7
1.5
4.6
4.7
10.7
10 .0
13.1
a.o
7.9
1.3
1.1
.1
12.6
12.
5.7
1 . 3
.2
.2
1.4
.1
10.8
15.0
15.0
.7
7 . 60
5.31
;; . 0 5
5 . 18
COVERAGE
ADJUST
SMOOTH
14.7
11.1
12 . 5
.2
13.3
12.2
10 . 3
14 . 3
13.7
12 . 2
7.5
1 1 .3
11.6
12 . 9
9.1
1 3. 1
9.7
9.6
3.3
15 .0
14 . 7
10.2
1.3
10.5
11.9
14 . B
14.4
4.4
15 . 0
'.2
.1
14 . 0
12 . 1
13 . 6
12 . 9
11.0
9.7
12 . B
3.8
12 . 1
13.7
13 . 2
5.6
12 . 5
1 1 .9
7.7
10. 1
2.3
10 . 5
11. 3
13 . 4
10 . 5
4.6
11.4
B.l
.7
4.B
B.5
2 . 1
9.5
12.9
14 . 0
12 . 7
12 . 5
10.5
13 .
10.6
12 . 4
13.7
11.4
12.5
12.2
12 . 7
13.3
13.3
12 . 7
11.9
15.0
1 3. 1
' 12 . 6
15.0
13 . 6
11.B
11.0
9.9
15 . 0
12.8
8.3
14 . 2
12.5
3 . 37
3 . 24
1 0 . 63
4 . 15
9 . 15
4 . 21
12 . 59
1 . 51
STICK
11.9
2.8
11 . 1
2.0
2.2
9.0
4.0
1.0
1.2
1.3
10 . 7
10 . 5
11 . 0
7.9
8.7
12 . 0
11.5
6.9
9.9
7.8
3.0
6.7
.5
8.6
4.0
.2
1.5
7 .
.2
5.8
6 . 0.
4 . 10
STICK
10 . 4
10 . 6
10 . 0
1 . 6
4.8
10 . 0
10 . 6
10 . 6
11 . 6
12 . 3
12 . 0
6.6
11.2
11 . 3
10.2
14 . 1
11.5
6.2
8.6
10 . 6
5.8
6.7
14 . 3
5.5
6.6
4.3
6.8
11.9
14 . 9
8.9
9 .35
3. 21
IV
.;..
......
Univers i t y
K l'I s e t s & r t
P...."!EL
Sl\HPLE
1
2
3
4
5
6
7
181
181
181
181
181
181
181
181
181
IBI
1 81
181
181
)'81
1 81
181
181
181
181
181
181
181
181
181
181
181
181
181
181
181
8
9
' 10
11
12
13
14
15
16
17
IB
19
20
21
22
23
24
25
26
27
28
29
30
SO
PANEL
847
847
847
847
847
847
847
B47
847
847
847
847
847
847
847
847
847
847
847
847
847
847
847
847
847
847
847
847
847
847
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
HEAN
SO
14 . 4
14.0
11.2
12 . 9
12 . 7
13.8
4.4
7.8
5.1
'12 . 0
9.2
11.1
11.9
11.2
9.9
11 . 9
11.4
13 . 3
12 . 9
7 .4
14 . 4
9.0
14 . B
7.5
6.9
13.6
12 . 2
13.3
11.3
13 . 7
l i . 17
2 . 83
MEAN
Sl\MPLE
studp-nt8
l\CCE PT
ACCEPT
12 . 2
13 . 2
6.7
3.1
11.6
8.7
3.0
2.3
4.9
4.2
9.2
7.6
5.9
5.4
8.8
11.7
7.7
12 . 3
10.5
2.4
6.2
8.2
1.0
2.1
10 . 4
12 . 4
13 . 8
2.2
3.6
5.5
7 . 23
3 . 88
PURCHl\SE
14 . 2
13 . 8
9.6
10.0
12.3
14 . 3
4.1
3.9
2.3
10 . 4
9.1
10 . 9
10.3
11.0
10.0
12 . 3
11.6
15 . 0
13 . 0
7.6
14.3
10.9
15 . 0
7.9
7.1
14 . 0
12 . 2
10.7
12.3
13 . 0
10. "
3 . 28
PURCHASE
12 . 9
13 . 0
3.1
1.9
10.7
7.1
2.8
1.9
2.2
3.3
7.4
7.2
4.4
3.8
7.5
11.9
6.8
12 . 4
11 . 1
2.7
3.4
6.2
.3
2.3
10.3
9.9
13 . 6
1.4
2.3
2 . 5.
6 . 21
4 . 20
PRICE
10 . 0
7.3
2.3
6 . 2
7.0
8.2
2.5
4.8
2.4
7.B
6.7
8.8
10.1
6.2
B.2
B.8
5.9
8.8
7.5
2.3
11.2
6.1
12.5
7.6
2.3
2.3
6.2
4.7
2.5
6.2
6 . 45
2 . 85
PRICE
7.S
7.3
0.0
0.0
4 . 6
4 . 6
0.0
2.2
2.2
4.8
6.7
7.5
2.9
2.4
7.0
8.4
3.4
4.8
6.2
0.0
1.1
3.5
3 .5
2.4
6.2
. 1
7.4
0.0
.1
1.2
. 3 . 60
2 . 82
SLIP
13 . 5
12 . 1
11.3
10. 1
11 . 3
10.7
7.1
11.3
12.2
10 . 5
10.4
10 . 9
12.8
12.1
8.5
12.9
10.6
10 . 3
11.5
11.0
10 . 9
14 . 1
13.1
10 . 1
10 . 6
11 . 8
11.8
9.8
11.8
10 . 7
1 1 . 19
1 . 42
SLIP
10. 9
5.5
6.0
2.2
11.9
9 . 8
6.2
8.4
10 . 8
4.0
14 . 3
8.8
4 . 6
10.4
9.7
11.9
'9 . 3
9' . 9
8.6
3.9
6.8
7.7
.8
6.9
9.5
8.2
6.2
2.7
3.0
5.4
7 . 48
3 . 26
O I S I NTEGR
. 3
2.0
9.4
3.0
13.0
4.8
5.7
1.5
.9
8.1
2.1
2.4
.7
3.8
2.5
1.4
3.2
10 . 5
3.1
2.9
1.2
.9
1.8
1.6
4.5
8.5
1.7
.5
.3
7.1
3 . 65
3 . 33
OIS IlITEGR
4.3
9.0
4.6
10.8
12 . 5
9.1
11.5
4.6
3.2
5.1
8.5
5.0
6.8
10 . 2
4.1
2.9
5.1
14 . 0
10 . 9
8.5
7.9
7.5
14 . 0
6.9
3.8
7.0
11.5
10.9
11.4
11.8
8.11
3 . 34
COVERJlGE
13 . 9
11.6
11.6
8.3
12.1
10 . 7
6.8
13.3
13 . 1
10.7
8.9
11 . 3
12.2
12 . 3
9.4
13 . B
10 . 4
11 . 5
11 . 4
12 . 0
13 . 0
11 . 0
4.4
11.6
12.6
14.8
10 . 8
13.8
13 . 8
12 . 1
11 . 44
. 2 . 20
COVERAGE
9.8
11 . 1
6.5
11.8
12.4
8.3
5.8
11 . 3
12 . 8
5.
4.1
10 . 1
6.8
12 . 3
9.6
12 . 3
8.1
5.3
4.8
10 . 7
12.3
6.8
12 . 3
7.5
11.2
10 . 0
9.0
5.4
8.0
11 . 1
9 . 08
2 . 14
l\(AJUST
.1
2.6
4 . 1
1.4
12 . 7
9.8
9.3
9.7
4.1
9.9
10 . 6
13 . 2
11 . 7
11 . 7
13 . 3
3.9
10 . 1
13.2
12 . 5
11.3
5.1
9.0
3.2
12 . 1
6 . 1
.7
. 1
3. 8
.1
8.5
SHOOn!
STICK
14 . 0
14 . 0
11 . 0
7.3
9.2
13 . 2
9.2
13.0
13 . 1
10.9
12 . 5
11.4
1 3.1
13 . 3
13.3
13.2
11 . 9
3.4
14 . 2
12 . 6
13 . 6
13.6
14 . 8
12 . 2
11.9
14.3
14.7
12 . 8
15.0
12 . 0
12 . 6
13 . 9
13 . 8
9.2
9.5
12 . 3
13 . 0
12 . 9
1 1 .3
11.5
12 . 0
11.3
10 . 9
11.3
12 . 1
13 . B
11 .5
5.8
12 .0
12 . 5
9.6
12 . 3
6.0
11.4
11.8
15.0
13 . 0
12 . 9
13 . a
10 . 1
7 . 46
4 . 58
1 2 . 29
2 . 42
1 1 . 64
2 . 07
ADJUST
SMOOn!
.5
3.4
2.6
. 6
12 . 7
12 . 1
3.0
11.2
4 . 1
9.2
10 . 5
12 . 0
1.9
12 . 5
12 . 9
4.4
7.9
12 . 1
11 . 0
11.3
3.9
6.2
8.3
7.7
7.6
8.1
6.9
7.7
5.0
7.1
7.48
3 . 85
5.2
12 . 8
11.6
3.4
9.6
1 3.1
3.7
8.1
6 . 5
e.6
10 . 6
10.S
6.0
11 . 1
13 . 3
12 . 5
9.8
13 . 4
12 . 6
12 . 6
12 . 9
11.0
2.5
7.7
10 . 8
13 . 6
3.0
4.9
12.2
9.6
9 . 45
3 . 52
STICK
11 . 9
13 . 9
14 . 2
11.6
9.8
14 . 2
13 . 3
1.9
12 . 3
6.5
10.7
11.7
12 . 7
11.9
9.5
11.7
11.5
5.8
9.1
12 . 5
14 .8
11.6
2.8
6.8
8.8
15 . 0
14. 7
8.9
12 . 1
11.5
10.79
3 . 32
Sl\HPLF.
I'l\N:L
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
428
428
428
428
428
428
428
42B
428
428
428
428
428
42B
428
42B
428
42B
428
428
42B
428
428
428
428
428
428
428
428
428
l\CrF. ..
r
.8
7.4
9.0
.2
11.6
9.2
3.6
6.3
5.1
1.9
8.4
7.6
11 . 0
11 . 2
B.2
9.B
6.B
13.7
B.6
7.4
4.7
6.9
12 . 2
B.3
3.8
2.2
2.7
10 . 0
2.6
3.2
6 . 81
3 . 61
ME....N
SO
f'URr""sr:
I'RICE
2.4
2.4
0.0
0.0
4 . 8
4 . 6
2.5
2.2
2.8
2.4
4.7
7.5
6.1
6.2
6.6
6.3
3.4
8.B
4.9
2,3
1.8
2.4
9.8
7.6
0.0
0.0
2.5
1.2
.1
1.2
.4
1 1 .5
4.2
.6
10.9
7.9
3.4
2.8
2.6
2.3
7.4
7.2
9.4
11.0
7.0
10.3
6.B
15 . 0
10 . 1
7.6
1.5
4.3
11 . 0
9.1
2.0
2.4
2.4
8.1
1.4
1.8
3 . 58
2 . 77
6 . 08
4 . 01
IDEAL
12.9
13.9
10.8
12.5
12.5
10.9
12.0
13 . 1
11.2
10.4
10.9
12.8
12 . 1
n.7
12 . 7
11.6
10.3
13.3
15.0
9.9
11.3
14.4
10.1
11.9
12 . 9
14.2
9.8
13 . 8
10.7
3
4
5
6
7
IDEAL
IDEAL
IDEAL
IDEAL
IDEAL
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
IDEAL
IDEAL
IDEAL
IDEAL
IDEAL
IDEAL
IDEAL
IDEAL
IDEAL
IDEAL
IDEAL
IDEAL
IDEAL
IDEAL
IDEAL
IDEAL
IDEAL
IDEAL
I DEAL
IDEAL
IDEAL
IDEAL
IDEAL
6 . 73
3 . 61
14 . 2
IDEAL
MEAN
HElIN
SO
SO
D I S I NTEGR
.5
8.1
9.0
6.7
11.0
11.2
6.4
1.3
11.7
4.6
7.3
9.4
10.4
10.4
10.0
8.
6.2
10.I
3.8
11.0
2.9
12 . 3
3.0
3.2
2.0
1.5
3.5
4.4
7.2
3 .B
SLI P
PANEL
Sl\MPLE
S1.11'
1 2 . 13
1 . 46
14 . 6
10 . 8
10 . 9
4.2
13.5
3.4
9.8
9.4
.9
9.2
4.7
2.7
1 . 5
10.2
10.7
4.5
9.1
.6
7.2
2.9
10 . 2
5.4
5.4
13.6
8.1
10 . 4
6.2
9.3
6.4
11 . 8
7 . 59
3 . 91
OISINTEGR
CQVERM';F.
l\DJIIST
SHOOTIl
. 1
9.4
10.3
5.5
1 1 .8
10 I
6.2
5.2
13.1
1.1
10 . 5
10.7
11.0
9.7
10.I
10.9
5.8
B.l
11. 9
12.0
10 . 7
7.6
14 . 7
2.4
5.0
7.1
11.9
7.0
6.7
10.2
13 . 8
10 . 4
11.3
5.2
12.7
11 . 8
8.7
5.9
4.1
12.B
11.2
12.2
10.1
12.5
12 . 2
8.1
7.9
11.1
12 . 5
11 . 3
11.9
9.0
6.2
2.6
8.6
9.8
1.8
10 . 9
4 . 0
2.0
.4
12.4
8.7
7.6
10.2
13.2
8.2
3.7
5.3
7 .3
B.O
11.4
7.2
13 . 3
13.3
11.5
9.8
15.0
14 . 5
12.6
12.2
12.5
10.5
2.5
6.5
7.8
7.8
11 . 3
8.2
3.5
8 . 56
3 . 53
COVERAGE
9 . 09
3 . 54
9 . 21
3 . 70
ADJUST
SMOOTH
STICK
11.6
9.9
12 . 8
5.8
5 :2
14 . 2
12.0
12 . 9
11.6
10.7
14 . 3
11.7
13.7
11 . 3
11.6
11 . 5
11.5
4.9
11.0
8.4
14 .8
9.6
12 . 3
1.9
13.5
12.4
5.1
e.o
8.6
13 .7
1 0 . 55
3 . 24
STICK
1.0
14 . 7
1 .6
13.7
13.4
12 . 6
13.3
12.8
15. 0
13 . 7
13 . 2
11. 3
11.7
13 . 0
12.9
12.8
13 . 4
11. 3
11.5
12 . 3
15 . 0
9.2
9.3
13 . 4
13.6
12.6
14 . 8
10 . 8
13 . 8
15.0
14.5
7.1
9.2
13 .1
12.7
12 . 5
12 . 2
12 . 1
13.5
11.2
13 .2
12.4
12 . 5
12 . 6
6 . 2
11.2
13 . 2
12.0
15.0
10 . 7
9.7
15 . 0
13.7
13 . 4
3.4
.1
3.8
3.4
9.5
13 . 8
13 . 6
9.8
15 . 0
12 . 7
15.0
13.7
13 . 2
12.5
12 . 2
13 . 5
13 . 3
1 3. 9
12.9
11.9
15.0
14 . 5
15 . 0
11.3
10.0
14.0
12 . 2
13 . 7
15.0
14.7
12.8
15.0
13.5
9.5
12 .8
6.5
15.0
11 . 3
9.1
11.8
12.5
12.2
11.3
10.9
12 . 7
12.5
12.8
11.5
4.7
12.0
7.4
10.9
7.2
11. 3
11.4
8.3
13 . 7
14.7
12 . 9
14,9
11.5
1 . 62
2 . 13
12 . 91
1 . 51
1 0 . 09
4 . 22
1 3 . 39
1 . 39
1 1 . 27
2 . 49
,3
1 . 0
1.1
11.0
.8
1.7
.3
.6
.3
2.1
1 . 6
.7
2.0
1.3
4 . 0
2.0
.1
2.6
.1
.5
4.3
. 1
.6
3.8
. 2
1 . 7
.5
.1
2.2
,1
7.9
H,O
14 , 0
B.O
13.9
or f i ce wockecs
SAMPLE
Pl'INEL
742
742
742
H2
742
142
742
142
142
742
742
742
742
742
742
742
742
H2
742
I
2
3
4
5
6
7
8
9
10
II
12
13
14
15
16
17
18
19
742
742
20
21
22
23
24
25
26
27
28
29
30
142
742
742
742
742
142
142
142
142
MEAtI
PANEL
1
2
3
4
5
6
7
8
9
10
II
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
243
243
243
243
243
243
243
243
243
243
243
20
243
243
243
243
243
243
243
24)
243
243
243
243
243
243
243
243
243
243
HEl'IIl
SO
PURC ..... SE
P R I CE
0.0
4.8
0.0
4.8
0.0
.1
0.0
0.0
0.0
0.0
6. I
0.0
5 . 0
4.8
8.8
10.0
.3
7.5
.2
5.)
1.6
10.0
0.0
7.6
.7
2.0
0.0
1.8
.6
5.3
3.2
0.0
3 . 1
2.5
7 .8
13.0
2.6
10 . 2
. 5
4 . )
3.5
3.9
14 . 1
12 . 6
8.6
.5
O.0
2.2
8. 4
0.0
0.0
14.4
.3
10.8
.6
0.0
5.3
3 . 5
0.0
0.0
5.5
0.0
'-8
0.0
2.3
.2
0 . 0
1.7
8.9
0.0
11 . 2
3.1
2.7
.2
.6
1.0
8.5
5.3
.2
3 . 1
7.0
7.8
10.9
1.2
12.5
.7
8.1
5 . 12
4 . 40
SO
Sl'IMPLE
l'ICCEP"r
l'ICCEPT
12.2
8.9
3.5
8.2
3.1
10.1
11.6
'4 . 5
5.4
5.6
3.3
10.;
4.9
1.8
2.1
.7
7.5
.4
3.2
2.9
e.1
2.4
. 6
.8
2.1
8.2
7.4
5.5
.6
12.6
'5 . 3 0
3 . 78
S.l
3 . 99
4 . 30
PURCI.... SE
12.2
10.0
0.0
7.6
.7
8.6
10.1
7.6
.6
.5
.5
8.4
4.2
.9
2.3
1.9
6.7
.6
1.4
1.)
e.3
0.0
0.0
.8
.3
9.9
14 .4
0.0
12.0
11.3
. 90
4 . 65
2.3
L7
2 41
3 . 09
PRICE
7.
'-8
0.0
4 . 8
O . o
4.9
5.9
2.5
0.0
0.0
2.3
7.4
5.0
3.7
4.8
.2
9.9
.5
1.3
.1
4.1
0.0
0.0
2.3
0.0
2.2
7.5
2.3
.2
3 .6
2 . 94
2 . 84
SLIP
.8
.0
. 9
. 1
.0
.2
.2
.9
2 . 1
5.6
3 . 3
. 2
3 .4
4.8
1 .9
3 . 7
1 . 0
7 . 5
.8
7.4
9
3
2
6
14
1
6.1
6.4
9.7
.1
2 . 3
10 . 5
. 1
10.7
1 .8
3 . 2
. 36
. 71
SL I P
2.
9.3
6.3
3.4
6.0
6.5
6.9
7.5
7.5
4.9
1 . 7
7.1
4.7
1.8
7.2
.9
9 . 4
.4
] .7
.8
11.0
7.7
1 . 8
2.4
.2
9.0
.2
10.7
1 .8
4.7
4 . 86
3 . 37
O I S I IITEGR
7.6
10 . 8
1 3. 8
11 . 4
11.0
4.5
1.9
7.2
13 . 9
5.8
4 . 3
14. 7
10.5
9.1
7.0
11 . 4
12.3
9 . 3
12 . 4
6.8
7.6
3 . 4
8.4
15.0
14 . 5
6.2
. 1
12.5
10.2
4.4
8 . 93
3 . 99
O I S I NTEGR
14.9
10 . 8
11.9
11.4
13.4
12.6
6.7
8.7
11 . 1
11.7
12 . 7
2.9
9.5
11 . 5
3 8
14.3
13 . 5
I!.7
13 . 9
13 . 8
3.4
8.0
9.4
II . 9
14 . 5
9.9
1.4
14.8
14 . 7
8.2
1 0 . 57
3 . 77
COVERl'IGE
7.6
7.4
7. I
11.2
10.5
8.8
13.5
6.4
7 . 1
12.9
4 . 1
2 . 1
6.1
8.2
.7
12 . 4
1l.2
5 . 3
3.3
B.7
11 . 5
3
4
7
6
12
.4
.7
. 1
.6
.0
. 3
4.8
3 . 8
8.1
7 . 26
3 . 64
COVEAAGE
1.6
7.4
10.3
11.2
10.5
13.0
9.6
7.7
7.8
12.4
3.5
8.3
1 . 9
2.6
8.1
8.4
5. I
9. I
6.3
1.4
12.4
6.6
2.0
6.0
3.6
14. 0
2.6
8.5
3.8
4.2
7 . 00
3 . 72
>.DJUST
5.6
8. 4
3.8
14.2
7.2
5.2
.7
6.3
1 . 3
9.9
6.2
. 9
5.8
5.0
14 .1
6.9
4.6
7 . 9
10.5
6.4
1.6
3.3
12.2
9.6
.6
11 .1
. 6
6.2
1 .4
8 .9
SHOO" '.
3.7
8.1
2.6
14.2
13.3
6 . 0
2.7
6.1
.9
6.1
7 . 3
.7
6.4
8.2
14 . 2
7.2
1 . 7
13.0
.9
7. 2
7.6
3 .6
8.6
9.6
12. 1
9.0
.8
. 4
.9
.9
. 21
. 96
6 .50
4 . 15
ADJUST
SMOO'I1I
1. 2
7 .8
9.0
14 .2
7 .2
6.0
7.1
8.2
4.1
10.3
4.9
8.7
3.8
2.3
7.1
5.0
12.3
11.1
14.0
13.7
4 . 1
11.2
14.0
7.1
2.6
4 . 4
1.9
1l.5
13.3
6.7
7 . 83
3 . 99
2.7
8.1
5.6
14 .2
13.3
3.3
12 . 2
12. 8
1.8
5.3
12.2
8.7
5.4
8.6
8.9
3.7
9.0
4. I
4 . 1
12.1
7.6
7 . 3
10.7
10.5
12 . 1
1 4. 1
2.5
9 . 3
2 . 2
4.5
7 . 90
3 . 93
STICK
14.1
10.7
10.5
2.3
12.1
14.2
15.0
9.2
10.0
13 . 0
5.0
6.5
7.3
7.0
9.9
.4
5.9
5.8
12.8
1 1 .2
2.9
1 3 '. 5
10.2
10.0
14.6
13.4
3.3
12.2
13.3
7.0
SAMPLE
1.7
5.4
7.9
13.7
12 . 1
13.7
9.6
7 . 8
2.2
14.0
9.8
12.1
3.9
6.5
13.9
3.4
10.2
3.5
6.6
8.9
2.9
12.3
11 . 3
11.2
14 . 6
14.9
13.9
12,2
11.5
12.7
. 48
. 16
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
HEAN
.44
4 . 06
STICK
PHEL
557
557
5 57
557
557
557
5 57
557
557
557
557
557
557
557
5 57
557
557
557
557
557
557
557
5 57
557
557
557
557
557
557
557
PAtlEL
181
)
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
lSI
181
181
181
181
181
181
181
181
181
181
181
18]
1 81
181
18J
181
18J
181
181
1 81
181
18J
1 81
181
181
1 81
1 81
181
ilEAl<
SD
12.8
12 . 2
12 . 5
2.9
10.5
11.7
3.3
8.8
12 . 6
10 . 2
10.2
9.9
9.7
9.3
5 . 8
6.7
13 .2
6.9
12.3
7 . 7
12.3
10.0
1.7
11 . 8
6.5
11 . 0
2 . 9
14.9
9 . 8
13.3
9 . 45
3 . 4B
SO
SAMPLE
ACCEP"r
ACCEPT
8.5
12.1
14 . 0
11 . 2
7 .3
11.4
6 . 1
13.3
12 . 0
10.8
12.3
11.7
10 . 6
7 . 0
4 . 2
13 . 9
3 . 3
8.2
13 . 4
13.1
13.4
11.8
3 . 1
n.o
14.2
10.6
10.0
11.8
8 . 5
11.8
1 0 . 44
3 . 19
PURClIl'ISE
12.8
11.7
0.0
7.6
9.6
13 . 5
0.0
11.8
10.7
7.4
4.7
7.9
8.8
3.3
6.0
.1
10.8
6.1
10.8
6.6
12.2
0.0
0.0
12.1
.3
13. 7
.6
13. 3
' 13.8
14 . 1
7 . 6B
5 . 07
PURCI.... SE
8.4
12.7
0.0
7.6
6.2
12.2
10 . I
14.2
9.9
9.5
9.7
9.5
9.8
1 .6
4 . 8
14.7
4 . 7
14.1
12 . 2
10.9
13 . 3
5.4
0.0
13.3
14 . 1
12.7
14.4
9.8
7 . 7
8.7
9. 41
4 . 18
PR I CE
7.4
8.8
0.0
4.8
4.9
8:8
0.0
12.5
4.8
0.0
7 . 3
7.4
7.6
4 . 8
7.4
.2
12.5
5.0
8.8
4.9
SLIP
O I S I NTEGR
0.0
0.0
6.9
0.0
8.8
0.0
2.3
4.9
4.9
14.2
10.4
13.3
13.6
13 . 1
13.6
14 . 0
II.0
U.8
12 . 8
1 3. 1
10.9
8.3
10.4
13.3
12.9
13 . 5
12 . 6
14.5
4.9
12.3
12.2
12 . 4
8.9
14.6
14.4
13.2
12.6
13.5
14 . 7
5 . 01
3 . 75
12.40
2 . 1 3
4.7
P R IC E
4.B
S.8
0.0
4.8
2.3
7.4
5.9
12.5
2.4
4.9
12.4
9.9
10.1
4.8
4 . 8
2.4
2.3
10.0
10.1
7.3
4.7
2.3
0.0
7 . 4
9.8
8.8
7.5
2.3
2.3
2 . 3
5 . 84
3 . 59
SLIP
e.7
12.6
11.1
10.2
6.0
e.4
9.
12. 6
8.0
12 . 1
11.7
12.2
9.3
4.2
14.8
14.6
4 . 0
J4.2
I!. 3
13.9
11.0
11.8
11.5
11.7
13.5
13 . 6
10.2
13 . 1
10 . 3
12 . 7
10 . 98
2 . 80
1.0
4.8
3.2
.9
1.3
.3
13.3
3.9
2.1
6.1
6.3
2.5
7.0
3.4
.3
1.8
1 .8
2.2
2.0
9.4
2.2
10.4
7.4
.3
.3
. 1
13.2
10.1
1.4
1.9
4 . 03
3 . 91
OISINTEGR
6 . 1i
l.S
8.6
. 2
11.0
3.0
9.0
.9
3.0
2 . 3
7.1
2.1
6.0
7.1
.3
'. 8
4 . 1
7.5
.5
1.2
3.4
1 . 9
12.6
13.4
. .3
1.8
14 .2
8.1
7.8
.7
5.11
4 . 14
COVERl'IGE
ADJUST
SHCXl'n1
lL3
10 . 3
15.0
2.8
9.7
13.0
3.3
11.2
10.2
5.3
10.8
8.9
3.2
9.3
4.3
7.2
6.5
7.4
14 0
12.7
10.0
15.0
14 . 2
9.0
3 .7
14.7
11 . 1
13 . 1
ILl
10.9
10.9
9.7
6.2
10.0
2.9
13 . 7
9.5
4.7
8.7
7.7
12 . 8
11.1
12 . 0
5.4
11.1
8.7
10.4
13 . 3
14.4
13 . 8
10.8
15.0
14.2
13.3
14 . 0
4.6
11.0
13.4
11.8
10.9
9.9
8.0
8.1
12.5
1 .7
10.2
5.6
14.0
9.9
1 0 . 29
3 . 24
1 1 . 22
3 . 35
COVEAAGE
A[)JUST
SHOOTII
7,D
Il.l
7.)
10.1
J),J
11.S
'-1
13.8
9.6
8.5
9.0
7.3
14.6
9.7
11.8
11.7
2.3
8 . 94
3 . 72
12.0
11.2
10.5
7.0
5.9
12 . 5
9.5
11.4
12.3
9.7
8.0
8.7
5.
13.8
9.8
1 3. 8
J2 . 5
13 . 5
12.4
10 . 3
6.1
4 .9
7,3
14.6
13.6
1 3,6
9.9
10 . 8
10.8
14 . 2
7.2
6.0
12.8
12.7
11.1
10.7
12.5
10.2
8.1
3.5
7.1
1 0 . 31
2 . 81
10 . H
3 . 54
14.3
8.7
15.0
13 . 0
1.2
10 . 5
13.5
8.9
12.8
14 .
13 . 0
13 . 7 '
9.5
4.0
12 . 8
14.3
11.7
12.0
12.5
14.5
15.0
8.8
7.0
13.6
14.6
9.0
'14.2
13.3
13.2
14 . 2
12.4
12.1
12 . 4
13 .0
9.9
9.3
8.8
8.9
14.1
3.7
9.4
10.9
1.4
12.1
12 . 4
13.0
14.0
7.0
15.0
12.5
13.8
13.6
13.4
1 1 . 32
3 . 12
STICK
5.8
10.7
12 . 1
.7
11.6
13.5
6.9
10 . 5
3.1
10.3
7.3
7.9
4.7
11.0
11.0
1.6
2.7
9.9
14 . 1
.7
6.9
10.0
8.4
12.4
14.6
13.4
2.1
14 . 7
7.0
4.0
8 . 32
4 . 34
STICK
),9
10.1
7.9
13.7
12 . 1
14 . 5
11 . 5
13.0
3.e
13.8
10. 9
12 . 1
8.7
12.8
8.5
13. 4
7.3
14.1
11.1
13.5
6.9
13 . 5
5.5
13 . 5
14.6
14 . 5
10 . 8
14 . 0
13.3
14.2
11. 26
3 . 2.6
/'V
VJ
O f f ice workrs
PANEL
SAMPLE
1
2
3
4
5
6
1
8
9
10
II
12
13
14
IS
16
17
18
19
20
21
22
23
24
25
26
21
28
29
30
841
841
841
847
841
841
841
841
841
841
841
841
841
841
841
841
847
841
841
841
847
841
841
847
847
847
847
841
847
847
MEAN
SO
IICCEPT
2.8
10.9
5.4
11.2
1.3
8.1
13 . 1
6.1
1.1
9.9
4.5
8.0
5.7
1.0
11 . 9
12.1
4.3
2.8
8.5
11.2
6.8
3.5
4.9
15 . 0
14.2
8.6
11.0
8.9
2.2
9.5
8 . 14
3 . 48
PURCHASE
2.8
11 . 1
0.0
1.6
6.2
5.2
10 . 1
9.9
2.4
8.1
2.3
5.9
5.1
2.1
11.6
13.8
5.1
3.5
1.5
8.1
1.1
0.0
0.0
15.0
14.1
10.8
14 . 4
3.5
5.3
6.4
6 . 87
4 . 39
PRICE
0.0
1.4
0.0
4.8
2.3
2.5
5.9
10.0
0.0
2.3
4.9
1.4
5.0
4.8
9.9
10 . 0
4.8
2.5
4.9
1.3
4.7
0.0
0.0
6.2
9.8
4.8
1 . 5
2.3
.2
.1
4 . 43
3 . 29
SLIP
D I S I IrTEGR
3.1
11.4
9.4
1.6
9.5
8.4
6.9
2.9
8.9
6.1
10.3
3.0
2.2
5.0
9.2
4.8
1.2
2.1
1.2
9.3
6.1
3.8
10 . 5
10 . 8
13 . 5
11 . 0
2.1
5.4
3.3
11.1
1 . 11
3 . 25
6.2
6.3
10 . 1
4.2
6.4
5.1
6.1
5.1
3.8
5.5
11.5
12 . 1
11.9
8.3
3.8
10 . 4
9.0
5.5
10 . 3
11.3
9.6
9.5
11.3
2.3
.3
8.6
10 . 5
6.0
10.2
3.2
1 . 54
3 . 19
COVERIIGE
8.5
10.3
14 . 1
11.2
10 . 5
6.2
11 . 3
5.2
10.9
10.8
8.2
4.4
9.3
1.5
1.0
11 . 1
11.1
10 . 6
10.6
11.1
10.3
1.5
3.5
8.1
14 . 4
12 . 0
12 . 2
10.3
1.5
12 . 1
9 . 61
2 . 66
IIDJUST
6.5
9.3
13.6
14.2
9.0
5.1
2.6
4.2
5.5
9.5
7.5
5.3
6.9
7.3
10 . 0
11.1
10 . 7
12 . 5
1.8
11.7
10 . 5
10 . 0
12.9
10.8
7.3
5.8
12 . 3
8.5
11.9
11 . 1
8 . 85
3 . 25
SMOOTH
5.3
9.8
12 . 5
14 . 2
13 . 3
10 . 1
10 . 6
8.2
1.2
10 . 5
4.6
5.1
4.4
8.5
11.4
11 . 3
5.4
11 . 1
2.5
14.2
3.1
9.9
9.5
11.5
14 . 5
11.2
11 . 2
3.0
11.2
11.4
9 . 24
3 . 54
STICK
13 . 6
10.1
1.9
13 . 1
12 . 1
14 . 0
12 . 8
6.3
6.8
13 . 3
3.1
9.8
9.9
12 . 8
12 . 2
14 . 4
8.1
12 . 0
9.1
7.1
8.8
4.3
12 . 3
15.0
14 . 6
14 . 0
12 . 1
13 . 0
15 . 0
5.6
1 0 . 85
3 . 31
SIIMPLE
PIINEL
SLIP
I DEAL
I DEII L
I DEIIL
I DEII L
I DEIIL
I DEIIL
I DEIIL
I DEIIL
WEIlL
I DEIIL '
I DEIIL
I DEIIL
I DEIIL
IDEAL
10
11
12
13
14
I DEII L
IS
I DEIIL
I DEIIL
16
17
18
19
20
21
22
23
24
25
26
27
28
I DEIIL
I DEAL
30
I DEAL
I DEII L
I DEAL
IDEIIL
I DEIIL
I DEIIL
I DEIIL
I DEIIL
I DElI L
I DEII L
I DEAL
29
MEAN
SO
PANEL
SJ\MPLE
418
U8
428
428
428
428
428
428
428
428
428
428
428
428
428
428
428
428
428
428
428
428
428
428
428
428
4 2 8428
428
428
1
2
3
4
5
6
7
8
9
10
11
12
13
14
IS
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
MEAN
SO
ACCEPT
PURCHASE
PRICE
SLI P
. 80
8 . 90
2 . 00
1 1 . 20
3 . 10
8 . 70
8 . SO
2 . 80
1 . SO
8 . 90
1 1 . 30
6 . 00
11. 60
7 . 00
1 1 . 90
1 . 80
7 . 50
1 0 . 20
4 . 20
13 . 10
6 . 80
6 . 10
1 1 . 10
10. 10
9 . 10
1 0 . 20
9 . 00
2 . 20
2 . 20
10 . 4 0
7 . 27
3 . 14
. 70
1 0 . 00
0 . 00
7 . 60
.10
5 . 20
0 . 00
3 . 40
. 60
6 . 60
8 . 10
4 . 70
] 0 . 90
2 . 70
1 1 . 60
2 . 40
8 . 10
8 . 70
2 . 50
1 2 . 70
1 . 10
0 . 00
1 3 . 00
1 0 . 50
. 30
12 . 1 0
1 4 . 40
0 . 00
5 . 30
4 . 10
5 . 84
4 . 66
0 . 00
4 . 80
0 . 00
4 . 80
0 . 00
2 . 50
0 . 00
0 . 00
0 . 00
0 . 00
9 . 90
2 . 30
1 2 . 50
4 . 80
9 . 90
. 20
7 . 30
1 . 50
2 . 50
8 . 80
4 . 70
0 . 00
8 . 80
4 . 80
0 . 00
8 . 80
7 . SO
2 . 30
.20
1 . 60
3 . 88
3 . 89
. 30
9 . 60
6 . 30
1 . 60
9 . 50
1 0 . 00
5 . 30
9 . 90
6 . 50
7 . 10
9 . 40
7 . 80
1 0 . 90
4 . 40
4 . 50
2 . 00
7 . 20
5 . 60
3 . 20
1 2 . 00
6 . 10
9 . 30
8.40
10 . 10
1 . 40
1 2 . 90
1 . 10
3 . 40
4 . 80
1 . 90
6 . 64
3 . 42
DISINTEGR
1 4 . 20
1 0 . 80
1 3 . 80
10 . 2 0
6 . 40
3 . 00
1 . 90
2 . 50
4 . 70
5 . 20
5 . 40
6 . 40
5 . 30
7 . 80
7 . 00
12 . 5 0
7 . 20
3 . 90
8 . 30
4 . 00
10 . 50
8 . 80
1 3 . 10
1 4 . 30
. 30
1 . 80
12 . 00
3 . 90
1 2 . 60
5 . 60
1 . 47
4 . 14
COVER!\GE
IIDJUST
SMOOTH
STICK
2 . 20
7 . 40
8 . 80
1 1 . 20
1 0 . SO
9 . 00
7 . 90
1 0 . 00
11 . 70
12 . 1 0
4 . 90
5 . 50
4 . 60
1 0 . SO
2. SO
9.10
8 . 30
1 1 . 80
8 . 20
6 . 80
1 1 . SO
8 . 80
10 . 10
1 1 . 10
1 4 . 40
12. SO
1 1 . 20
6 . 60
7 . 50
1 4 . 30
9 . 05
3 . 09
1 . 90
7 . 80
8 . 00
14 . 2 0
9 . 00
4 . 20
1 2 . 80
9 . 90
4 . 60
10.10
9 . 20
6 . 20
2 . 60
4 . 50
1 4 . 10
1. 30
7 . 10
1 3 . 90
3 . 20
3 . 50
1 . 60
12 . 40
9 . 10
8 . 50
13 . 2 0
1 . 20
10 . 40
1 . 20
10 . 20
5 . 00
7 .18
3 . 94
1 . 80
8 . 10
7 . 50
1 4 . 20
1 3 . 30
6 . 90
13 . 30
1 0 . 00
3 . 00
1 1 . 30
6 . 10
4 . 00
1 0 . 50
8 . 50
1 4 . 20
. 30
7 . 50
1 . 40
2 . 40
4 . 90
1 2 . 10
1 1 . 00
1 . 10
8 . 60
1 4 . 50
1 4 . 70
9 . 80
.10
3 . 90
2 . 10
8 . 01
4 . 41
2. SO
10. 10
1 3 . 80
13 . 70
12. 10
1 4 . 00
8.70
1 1 . 90
5 . 50
13 . 1 0
8 . 00
7 . 90
5 . 70
1 1 . 70
1 3 . 90
1 1 . 90
1 2 . 00
1 . 80
13.40
3 . 20
1 1 . 80
11.30
13 . 60
1 4 . 30
1 4 . 60
14 . SO
13 . 00
1 4 . 40
1 5 . 00
1 1 . 30
11 . 18
3.46
DISINTEGR
13 . 2
12 . 6
13.3
14 . 1
14 . 4
13 . 8
6.9
8.9
12 . 8
12 . 3
11.3
12.8
13.4
11 . 9
11 . 0
14. 6
13 . 5
14.2
14 . 5
13 . 9
12 . 3
10 . 1
12 . 4
14 . 1
14 . 6
8.1
14 . 4
13 . 7
14 . 6
14.7
2.1
4.5
1.0
.2
.1
.1
6.1
1.6
2.1
1.1
3.3
.9
2.9
.1
2.4
.8
4.1
1.5
.5
1.2
2.2
8.4
1 . 4
.8
.3
1.0
.1
8.1
12 .15
2 . 01
2 . 41
2 . 65
.1
.1
COVER!\GE
ADJUST
SMOOTH
12.6
12.0
15.0
14.4
13 . 4
13 . 4
11.3
13.3
10 . 2
12 . 6
10.1
12.3
13 . 0
11.9
13.3
11.1
11.1
13.8
14 . 0
13 . 5
14 . 6
8.1
6 . 1
13 . 1
14.4
15 . 0
1 4 .e
13 . 6
15 . 0
14.3
13.9
11.3
15.0
14 . 6
12.4
10.4
2.6
13 . 4
13 . 1
11.5
12 . 0
12.1
12.5
9.2
14 . 1
14.3
13.1
15.0
14.0
11.1
1.1
11.8
9.7
13 . 1
14 . 4
12 . 1
14 . 4
9.5
8.3
14. 4
12.1
11 . 5
14 . 5
14 . 6
14.3
14. 4
10.6
13 . 4
13.4
12.7
14.1
12.4
12 . 8
10.5
14.1
14 .1
9.0
13.0
14.0
14.2
14.3
12.8
7.1
13.4
14.5
15.0
13 . 6
13.8
15.0
14 . 6
12.73
2 . 08
1 2 . 09
2 . 71
1 3 . 19
1 . 76
STICK
9.3
11.1
1.9
14.4
13 . 8
14 . 5
12.8
12 . 3
3.1
12 . 6
8.9
.8 . 9
13.2
10 . 1
14 . 6
14 . 4
2.1
14 . 1
11.1
13 . 5
8.8
13 . 8
13 . 6
13.0
14.6
12 . 5
14 . 4
14.0
9.0
11.3
11 . 61
3 . 16
IV
Schoo l
e t udenta
SAMPLE
PANEL
IB
19
20
21
8.8
11.4
3. I
2.0
11 . 5
3 . 5
8.8
2.8
11.3
4.9
.4
1 4.
10.1
4.5
7.5
3.5
7.2
14.0
10.6
13.9
11 . 3
13.8
14 . 4
1.9
9.9
13. 1
7.4
12.4
12.9
12 . 3
7.8
9.9
11.4
8.8
13.2
13 .0
3.1
15.0
7.3
10.2
11 .0
9.1
23
24
25
26
27
28
29
30
1).9
2.'
.7
3.8
1 . 4
.5
1 .8
2.1
1).1
2.3
0.0
2.'
1.0
2.3
.3
2.3
4.8
0.0
0.0
.4
1 .7
1.1
0.0
.2
.5
.9
.1
.7
5.7
6.6
.1
2.0
12.2
10.7
10 . 4
13.0
7.9
8.8
14.6
13.7
3.1
11.3
2.3
5 .7
6.1
4.6
6.9
5.3
12.5
1.5
3.3
10.2
7.2
4.7
6.6
6.0
9.0
9.5
1 .0
7.2
8.
7.5
7.B
7.7
1 .0
11 .5
4.1
14.3
3.1
11.9
. 1
11.8
4 . 89
3 . 63
I
2
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
SO
STICK
6.9
10.8
2.9
11.0
2.6
. 4
5.5
3.8
.4
8.8
4.2
7 . 3
4.6
4.2
5.6
B.3
I. B
1 .9
3.5
3.1
7 . 1
PANEL
HElIN
SMOOTH
6.7
7.2
3.2
10 . 5
5.5
. 1
13.4
11.8
7.3
10.0
1.6
12.9
8 . 1
4.7
6.2
14. I
13 .1
1 . 6
5.8
B.6
9.6
SO
243
ADJUST
11.2
14 . 1
11.9
14.4
9.7
13.2
9.7
10.7
12.3
12.1
1.7
3.1
1 . 7
7 .4
7 . 6
13.2
13.2
.8
11.7
10.0
8.4
HEAtI
243
243
243
243
243
243
243
243
243
243
243
243
243
243
243
243
243
243
243
243
243
243
243
243
243
243
243
COVERAGE
3.9
3.0
1.5
.4
3.6
. 5
3.6
6.5
.1
5.6
1.0
2.6
3.0
2.7
4 . 6
.1
1 .2
.B
3.8
4.6
9.7
243
243
O I S I llTEGR
.1
1.2
1.2
1.9
.9
1 . 7
6.0
.1
.3
1.2
0.0
0.0
2 . 4
10.4
6 . 2
0.0
12.5
6.2
4 ."8
3 .9
3.B
74i
SAMPLE
SLIP
1.6
. 8
3 .0
1 .4
3 0
1.4
3.8
5.3
1.1
0.0
.1
2.3
7 . 1
10.2
3 . 3
0.0
5.6
9.3
3.6
8.7
5.9
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
74.
PURCHASE PRICE
4.2
1 . 4
3.9
2.6
2.8
2.2
5.3
8.1
4.0
1.7
2.2
7.1
7 . 3
10.1
7 . 1
.9
9.9
11.8
3.7
8.5
.
5.7
742
742
742
742
742
742
742
742
742
742
742
742
742
742
7 42
742
742
742
742
742
742
742
742
742
742
742
742
742
ACCEPT
ACCEPT
7.0
.5
2.4
0.0
10.6
2.5
10 . 6
0.0
1.5
12 . 3
3.4
3.2
6.7
3.2
5.1
12.6
4.0
2.9
1 .4
2.7
4.0
12.3
4.2
5.3
5.2
11.1
7.6
0.0
3.2
11.1
5 . 22
4 . 03
6,)
3 . 60
3 . 43
PURCHASE
3.9
. 1
.7
0.0
12.8
2.3
5.9
. 1
0.0
12. 5
1..
.6
7.1
0.0
2.6
14.6
2.8
.8
1.2
3.7
3.2
10. 5
3.9
6.8
2.4
11.7
7.1
0.0
1.1
11.7
4 . 38
4 . 59 '
3.6
2 . 55
3 . 17
PRICE
3.8
. 1
0.0
.1
2.2
3.4
6.0
.1
0.0
8.8
0.0
0.0
2.4
0.0
9.9
8.3
.1
.1
. 1
1 . 1
1 . 6
7 . 4
0.0
3 .9
0.0
6.0
2.4
0.0
0.0
5.4
2 . 44
3 . 12
5,4
2 . 83
2.45
SLI P
11.3
4.2
5.1
13.I
11.2
2.8
11.9
7.8
.9
11.6
12.9
12.2
8.0
8.9
5.7
12.8
11.6
6.4
10.8
6.6
11 . 5
1l.5
2.9
4.0
3.1
12.1
7.0
8.4
13.9
9.5
8 . 66
3 . 73
5.5
9 . 83
3 . 92
D I S I NTEGR
13 .4
12 .6
13 3
14 . 9
10 9
15.0
3.8
9.3
14 .9
2.4
13.6
12.5
3.9
15.0
6.5
7.6
6.1
13.7
4 . 1
6.8
14.8
12.8
10.6
10.5
13.1
4.7
4.9
14 .5
.4
12.5
9 . 97
4 . 50
10.6
1 .6
12 . 4
5 . 2B
3 . 25
7 . 87
3 . 94
COVERAGE
ADJUST
SMOOTH
5.9
3.2
1 .7
9.2
5.0
6.5
5.9
10 4
.5
10 9
13.0
11.3
6.5
.1
6.8
5.8
1.8
5.9
7.I
5.0
8.0
5.3
12.0
10.8
4.3
3.7
8.6
2.5
13.2
3.3
9.5
8.0
1 . 4
12.1
1.6
3.0
9.9
1.3
2.4
10.5
3.4
9.1
9.8
15.0
6.5
13.8
13.5
3.4
S.8
7.8
8.6
3.7
10.6
12 . 9
7.5
14 . 3
7.2
2.5
10.1
11.6
3.0
9.3
1.6
.5
8.0
11.9
11.1
8.7
10.I
13.9
14.5
14.4
4.7
11.6
9.2
14.4
4.7
1.6
11.9
I!. I
10.8
13.3
9.8
5.0
5.0
11.1
11.6
.1
13.
7.7
7 . 26
3 . 83
6 . 47
3 . 65
7 . 99
4 . 21
B . BO
4 . 42
II.
SAMPLE
5 57
557
557
557
557
557
557
557
557
557
557
557
557
557
557
5 57
557
557
557
557
557
557
557
557
557
557
557
557
5 57
557
5.8
7.9
5.0
12.6
1 1 .7
.1
11 . 7
2.9
1 . 4
5.4
2.2
3.1
10.1
11.9
7 .9
11 . 2
15.0
13.8
11.9
5.7
3.5
13.4
11.9
9.3
2.4
.2
5.8
10.8
1 . 5
3.6
7 . 32
4 . 59
4
5
6
7
8
9
10
11
12
13
14
15
16
17
IB
19
20
21
22
23
24
25
26
27
2B
29
30
SO
PAUEL
SAMPLE
1
2
3
4
5
6
7
8
9
10
181
1 81
181
181
181
181
181
181
181
1 81
181
181
181
1 81
181
1 81
181
1 81
181
181
181
lSI
101
1 81
lSI
181
181
1 81
1 81
181
II
12
13
14
IS
16
17
18
19
20
21
22
21
24
25
26
27
28
29
30
HEll"
SD
ACCEPT
12.I
14 . 1
12.5
13.2
10.2
4 . 5
13 . 1
8.7
6.6
4 . 3
11.7
1.9
4.6
5.2
. 11.6
14 . 0
1 . 5
12.3
7.9
6 . 1
3.0
5.9
13 . 1
1 . 7
12.5
13.1
8.8
9.1
11.6
9.5
8 . 81
4 . 08
HEAtI
9 . 67
4 . 10
STICK
PAtEL
ACCEPT
1l.5
14.9
11.6
12.7
2 . 1
6.2
12.8
7.6
12.7
9.5
10.5
13.3
II.0
4.8
12.4
8.2
6.5
14.2
13.0
6.8
7 . 3
13.2
4.2
10 . 7
9.6
7 . 4
12 . 0
12.5
6.9
13.7
9 . 99
3 . 34
PURCHASE
13 . 8
12 . 0
13 .2
12.9
10.3
4 . 9
1 3. 1
12.2
2.5
1 . 9
10.5
0.0
3.5
7 .8
9.3
15.0
0.0
12.2
10.5
6 . 1
1.9
2.
12 . 8
1 . 5
13.1
13 .0
8.3
B.2
15.0
10.7
8 . 62
4 . 87
PURCIIASE
12.9
14.3
12.3
1l.8
1.6
7.4
12 . 5
4.4
12.9
7.9
9.2
10.0
10.2
7.2
7.8
3.4
6.6
14.0
13 .2
6.7
7.3
12.0
2.4
10.6
9.7
7 . 4
11.9
10.6
13.9
12.7
9 . 49
3 . 54
PR I CE
12.5
10. I
7.3
7.3
1 .8
4 . 8
7.4
.1
2.4
2.3
2.9
0.0
0.0
8.0
12.4
9.9
.1
7 . 5
B.8
1 . 4
.6
2.4
3.5
0.0
7.4
7.'
'.9
6.1
9.9
4.7
5 . 13
3 . 88
PRICE
11.2
11 . 5
6.1
5.5
.5
6.2
7.4
.1
12 . 5
4.8
2.3
4.9
3 . 7
6.6
11 . 3
6.7
8.8
10.1
12.5
1.4
6.0
8.8
0.0
8.0
4.8
1.6
8.8
8.8
7.4
7.4
6 . 52
3 . 65
SLI P
O I S I NTEGR
n.7
12.1
13.1
15.0
11.9
13.6
1(.0
H .
13 . 8
13 . 1
12.2
14.4
9.4
11.3
9.8
9.7
12.8
1 .9
9.3
I (.8
12.1
1 2. 7
2.0
9.8
13.6
13.9
10.7
11.0
.1
12.8
1 1 . 30
3 . 77
SLIP
13.I
14.1
12.2
1l.2
9.8
12.7
11.I
9.1
11.0
12.3
11.2
7 .2
4 . 4
10.2
12.0
6.4
10.0
14 . 5
13.3
12.1
10.6
8.8
1 .2
7 .5
12.5
2.8
10.7
13 .5
12.4
11.0
1 0 . 30
1 . 23
1.3
3.2
1 . 3
.2
1.6
.6
2.0
.5
13.9
1.1
12.3
1.9
9.4
1.5
2.9
.9
10.5
2.0
14.2
.6
'.9
11.9
6.4
1.9
1.6
.8
1 . 2
2.7
15.0
2.2
4 . 35
4 . 81
D I S I NTEGR
2.2
.7
3.1
Il.O
.7
1 . 3
1.7
4.4
1.0
2.1
11.3
7 . 3
5 . 9
1l.5
1.9
.4
3 . o
10 . 1
.
5.6
3.0 .
5.4
1 . 6
5.3
7.0
2 . '6
8.2 .
. 8
. 3
.1
4.0
4 . 18
3 . 67
COVERAGE
ADJUST
SMOOTII
9.1
11.4
13.4
15.0
1 3. 1
8.6
11.8
8.9
1.9
2.5
11 .5
6.0
1.8
3.6
10.9
4.1
8.3
3.5
4.9
5.0
5.
9.2
13.5
4 . 5'
13.4
3.7
3.I
1.8
8.2
5.7
13 . 8
13 . 6
2.6
9.9
4.6
9.3
10 . 1
3.1
7.7
6.3
15.0
14.2
9.5
9.0
1.4
2.4
8.5
13.7
13 . 1
9.9
7 .8
14 . 5
9.0
14 . 3
10 . 5
13.7
14.9
11.8
7.6
12.1
13.1
9.1
6.3
12.7
14 . 4
10. 1
7.9
12 . 5
11.8
10.6
12 . 6
7.0
12 . 5
6.1
10. 2
13.8
12.0
14.4
13.2
12.2
11 .8
2.8
11.5
1 4. 0
B . 06
4 .35
COVERAGE
5
B
4
1
.6
. I
.8
.3
.6
8.8
7 . 81
4 . 24
1 1 . 12
2 . 91
ADJUST
SHOOTH
II .0
13.6
11.7
. 13.2
12 . 4
13.9
1.1
11.4
9.6
13.2
12 . 0
10.5
13 .9
4.4
2.7
12.I
7.6
6.7
13.1
11.1
8.6
6.2
6.5
14.6
9.6
2.9
12.5
10.0
14 . 2
14 . 3
7.7
2.4
4 . 1
9.0
6.9
11.3
10 . 1
4 .6
6.4
9.8
7.2
11 . 2
.9
6.8
11.1
9.9
8.8
.9
5.2
13.8
3.2
12. I
4.3
9 .2
13.0
4.1
6.0
11.9
9 . 95
3 . 77
II.6
14.B
8.14
3 . 96
14 . 1
12.2
10 . 7
12.2
STICK
U.7
2.4
11.5
.2
.2
1.6
9.6
12.1
.3
1.2
11.8
4.2
7.0
3.3
9.0
.9
8.4
5.5
9.3
5.7
2.8
14.1
1.3
6.4
9.6
1.2
9.2
7 .3
2.7
2.2
5 . 96
4 . 22
STICK
13 . 6
4.7
13.0
)4,f
12 . 2
1 . 0
12.9
12.9
13.5
l3.3
9.2
14. 4
4.7
3.5
13 . 4
13.2
8.2
9.7
9.6
13.9
12.2
13.3
.9
10.8
10.9
5.7
12.l
14.0
12.0
4.7
5.1
3.0
13.1
13.6
13.6
4.4
11.0
13. 6
12.2
10 . 1
7.0
5.5
7.2
12.7
8.3
14.3
6.6
15 . 0
12 8
11.0
10.8
12.6
11.2
13.3
5.6
6.7
1 0 . 38
3 . 94
1 0 . 23
3 . 64
<.n
School
students
SAMPLE
PANEL
847
847
847
847
847
847
847
847
847
847
2
4
5
6
7
847
847
847
}\CCEPT
8.5
11 . 0
9.1
9.3
4.1
7.6
8
9
10
11
12 . 0
.3
8. 4
7.4
7.0
12
4.6
13
14
PURCH,>.SE PRICE
5.7
8.8
10 . 1
6.1
4 . 8
8.3
4.6
5.5
7.1
9.6
3 . 8
1 . 3
7 . 3
12.0
10.5
6 . 8
8.1
9.6
10.2
13 . 4
13 . 6
2.9
7.1
9.5
8.3
5.2
8.6
10.0
.6
3.9
7 . 9
4.1
1.2
6.0
.1
4 . 8
4 . 8
0. 0
SAMPLE
428
42B
428
428
2
3
5
6
7
428
428
428
428
428
428
428
428
8
9
10
11
12
13
14
428
428
428
428
428
15
16
17
18
428
428
428
428
428
19
20
21
22
23
428
428
428
428
24
25
428
428
428
8.1
4 . 2
0.0
6.2
1.8
7.0
2 . 3
4.7
8.6
8.9
4.7
5.8
13 . 6
13 . 3
6.4
2.1
7.0
9.5
2.9
4 . 9
5.7
.3
6.6
3.6
3.6
. 8
6.0
ACCEPT
6.2
2.4
5.4
1 . 6
1.4
8.9
11 . 1
7. 6
11.2
7.4
4.4
5.9
6.2
6.0
3.1
6.9
8.8
10 . 6
11. 0
13 . 5
8.5
10.6
12 . 5
3.0
11.2
28
8.6
7.6
.5
29
30
9.0
4.4
26
27
428
8 . 5
5.
.
8.
8.
1 . 1
0.0
6 . 98
3 . 49
PANEL
10.8
8.5
2 . 4
'30
MEAN
SO
4.6
3.2
0.0
29
MEAN
7 . 18
SO
3 . 51
5 . 98
3 . 67
PURCHASE
2.4
2.1
5.2
.7
.3
9.5
9.0
4 . 4
9 . 6
7.9
IDEI\L
17
14.5
IOE}\L
18
14.5
1.2
6.4
IOEI\L
19
14.2
.8
.7
13.9
7 . 0
IOEl\L
20
14.8
.1
10.0
11.4
11.8
12 . 8
11.4
14 . 1
IOEl\L
I DEAL
PRICE
1.3
2.1
2.3
.9
0.0
9.9
6.0
.1
7 . 4
SLI P
6.0
OIS INTEGR
12 . 5
11 . 4
3.2
1.8
2.6
10.9
9.6
9.1
12 . 5
10.7
.5
13 . 3
7.0
10 . 8
6.7
8.5
3 . 5
1.7
8.7
8.6
.1
4.5
7 . 5
11.1
9.1
4. 3
1.5
5.7
4.4
3.2
10.7
8.9
6.74
8.6
6.7
6 . 26
4 . 10
12.9
4.4
6.7
3 . 66
. 6
6 . 1
6.2
1.2
14.6
4.2
4.4
12.0
9.8
1.7
2 . 76
3.0
11.4
12.9
8.9
7.2
4 . 1
12 . 0
5.3
8.0
7.9
7 . 48
13 . 4
7 . 0
12.3
15
16
6.3
9.2
3 . 52
11 .5
3.5
9.1
1 . 5
2.2
7.7
IDEI\L
IOE,\L
7.6
4 . 12
4 . 8
0.0
0.0
0.0
10.0
5 . 1
4 . 1
11 . 2
2.5
9.9
10 . 0
7 . 4
4 . 9
2 . 3
2.1
1.7
IDEI\L
.7
2.6
7.3
8.0
3.9
9.5
2.4
IDE,\L
4 . 0
11.9
13.0
4. 9
1.6
8.8
23
24
loS
11.0
7.2
7.7
22
2.8
10 . 1
13.1
11.4
1.1
12 . 8
7.2
847
847
9.5
4.6
3.6
7.3
5.1
1.8
3.9
9.9
13 . 8
6.8
2
6
4
1
6
.
.
.
.
.
0
3
4
9
7
2.5
13 . 6
3.2
2.9
4 . 5
3.5
7. 6
8. 43
2 . 74
COVERAGE
13 . 5
9.2
4.5
12.7
6.7
3.1
10 . 1
9.6
11.0
9.4
2.7
14 . 5
2.7
5.
7 .
2.
7 .
8
6
0
4
3.5
8.0
11.9
14.4
8.7
8.8
6.0
6.2
3.6
6 .
11.
1 .
1 .
7.
AruUST
SMOOTH
5.6
3.6
4.4
9.5
9.5
13 . 5
10 . 0
12 . 9
12 . 3
9.0
1.0
14 . 4
4. 7
5.4
6
12
8
1
.3
.4
.4
.7
8.2
9.8
5.8
8.4
12 . 2
7.3
9.7
6.8
3 . 9
8.4
9.9
5.7
6.0
13 . 8
8.0
12 . 8
1.5
3.1
4 . 4
8.6
7 . 6
9.2
6.2
2 . 6
:3 . 2
.9
15.0
6.8
6.6
4 . 15
3 .70
7 . 14
3 . 77
6 . 37
8 . 54
3 . 91
3 . 80
7.0
7.1
5.1
13.0
4.5
3 . 86
2.0
12.8
7.4
9.6
B.8
10 . 5
11 . 5
9 . 45
5 . 6
8.5
13 . 8
4 .7
1 . 5
12.2
9 . 5
11.6
5.9
6 . 61
3 . 38
8
8
6
5
10.2
14.9
.2
.1
12
9.9
7.0
20
15.0
14.B
7.2
5
6
7
IDE,\L
13 . 5
847
847
1.3
IDE,\L
14 . 2
19
1.2
12.2
12.5
9.0
847
13 . 5
3
4
IOE,\L
IOE}\L
12 . 6
11.3
. 8
7.4
8.4
11.6
9.4
8.8
11.8
1 . 6
1.1
10 . 8
6.1
I OE,\L
IOE,\L
IDEAL
DISINTEGR
14 .0
8.4
13.9
12 . 9
14.2
8.4
2.9
7.4
SLIP
P'>'NEL
IDE,\L
5.0
10.1
4.4
12 . 1
3 . 6
9.5
8.8
2.3
SAMPLE
11.0
.G
6.0
7 . 4
12 . 5
18
28
12 . 0
13.1
7.9
7 . 1
13.7
STICK
9
10
11
6.0
9.6
17
26
27
10.0
2.9
10 . 3
4. 8
SMOOTH
10 . 8
4. 8
13 . 2
10.4
13. 9
7.9
2.4
8.5
8.9
10.7
847
B47
847
847
3.6
9.5
6.0
12 . 0
7.7
2 . 6
847
847
.8
.2
.6
.0
9. 0
10 . 5
5.8
14.4
3.6
6
1
3
1
3
5
6
3
8.0
8.2
8.3
2.4
12 . 2
16
25
6 . 1
1.6
4.9
1 . 5
2.8
9.1
8. 7
3 . 5
8.6
8.4
0.0
0. 0
IS
847
... ruUST
7 . 2
847
847
847
847
COVER,\GE
6.4
11.2
5.9
.9
21
DISINTEGR
8.5
5.7
9.9
10 . 7
7.9
11 . 1
9.1
847
S LI
6.1
5.2
7 . 20
3 . 49
10 . 6
9.1
7.2
8.3
J OE1.
IOEL
IOEL
13 . 0
14 .
"
14.7
12.7
14 . 4
8.7
14 . 1
10 . 8
13 . 4
15 . 0
14.9
12.2
8.8
1 3 . 1
14.9
14.2
14 . 7
11 . 6
13 . 6
13 . 7
13.0
14 . 8
14 . 1
14 . 9
14 . 8
13 . 2
9.5
. 3
14.8
)4 . )
14.8
13.8
14. 0
7.3
7.2
14.6
14 . 6
7.8
1 4 . 9
14 . 9
12.8
14.8
H.9
15 . 0
14 . 5
4.2
14.9
1(.4
13.6
12.2
13
9.0
2.6
13 . 9
7.8
11.3
11.3
14
10.5
.9
13 . 4
9.9
14 . 4
11.1
12 . 2
2.9
2.0
13 . 9
13. 1
15.0
9.4
13 . 3
7 . 3
14 . 6
13.9
12.6
10.7
7.8
14 . 4
12.0
15 . 0
13 . 9
11.7
13.5
13 . 1
8.3
10.5
13 . 9
14.3
9.2
14 . 7
8.2
11.2
11.8
10.4
14.9
7.8
.3
14 . 9
14 . 7
15.0
14.9
10 . 2
10 . 1
15 . 5
13 . 6
14.3
14 .9
15 . 0
14 . 2
8.3
12.5
8.7
10.0
14 . 2
1.6
4.2
21
9.3
22
23
14 . 5
14 . 3
.6
IO
L
IDEI\L
24
25
7.0
14.5
8.1
10.8
IDEI\L
6.5
14.5
10.7
8.4
9.7
IDEAL
IOEI\L
IOEl\L
26
27
13 . 5
14 . 9
10.2
IOEI\L
28
29
30
.3
. 8
.3
9 . 51
3 . 46
SMOOTH
13 . 8
14.6
7.2
IDEI\L
El\
,>.ruUST
13 . 8
.1
.3
.5
7 . 3
2.6
7.4
5 . 1
MEI\N
SO
.2
.1
.1
10.0
2.2
1 2 . 30
1 . 44
2 . 85
1 . 96
15 . 0
15.0
9.2
12 . 2
11.2
13 . 0
15 . 0
7.7
10 .1
14 . 0
14 . 3
1 3. 3
. 1
9 . 0
11 . 5
13 . 1
12 . 9 6
2 . 42
1 1 . 13
13 . 87
1 1 . 40
3 . 16
1 . 31
3 . 62
STICK
12 . 3
12.6
3.9
11. 3
6.3
11.9
9.1
13 . 6
4.8
9.
6.
12 .
7 .
3
0
5
7
2.5
10.8
14.5
15 . 0
9
13
7
12
.5
.3
. 0
.0
8.8
12.5
4.7
6.5
10.8
12 . 4
14.0
6.6
11.5
10.3
7.4
1.9
11.0
9.1
10 . 4
9.8
14.3
9.5
15.0
11 . 5
13 . 1
8 . 88
3 . 44
IDE}\L
1.3
1.2
. 1
STICK
COVEAAGE
10 . 2 4
3 . 35
tv
JoI::>.
C1'I
APPENDIX 6.2 (continued)
Analysis of Variance of Sensory Scores Obtained from the Three Groups of Consumers
F-ratio
Attributes
Acceptability
Sample
Group
Purchase intention
Slipperiness
Coverage
Smoothness
0.000
Sample
Group
47.62
0.000
0.05
0.949
Sample
22.84
0 .000
0 . 80
0.476
49.44
0.000
0.56
0.588
1 6. 67
0 .000
3.75
0.061
Sample
2.50
0 . 1 02
Group
3 .38
0.076
31 . 68
0. 000
6.43
0.01 6
1 0.98
0.001
9.48
0.005
Sample
Group
Sample
Sample
Group
Stickability
,0.073
0.050
Group
Adjusta bility
0 . 000
3.43
4.09
Group
Disintegra tion
58.89
54.33
Sample
Group
Price to buy
P value
Sample
Group
Ideal Ratio Scores of the Prototypes' Sensory Attributes (Original Data)
APPENDIX 6.3
Kasetaart Univer s i ty
PANEL
SAMPLE
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
742
742
742
742
742
742
742
742
742
742
742
742
742
742
742
742
742
742
742
742
742
21
22
742
23
24
25
26
27
2&
29
30
742
742
7 42
742
742
742
742
742
HEAN
SO
PANEL
SAMPLE
243
243
243
243
2 43
243
243
243
243
243
243
243
243
243
243
243
243
243
243
243
243
243
243
243
243
243
243
243
243
243
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
MEAN
SO
Students
SLIP
DEFORM
COATING
ADJUST
SKOOTH
4 6 . 67
1 3 . 00
12 . 30
8 . 00
1 . 04
6 . 00
7 . 65
41 . 33
11 . 3 3
27 . 00
1 .95
5 . 38
9 . 71
1 . 90
1 . 69
2 . 98
4 . 55
1 4 0 . 00
3 . 38
85 . 00
. 06
. 87
. 61
. 55
1 . 04
. 67
. 22
. 19
. 90
. 27
.79
. 76
. 52
. 75
. 77
. 84
. 59
.70
. 85
.71
35
18
28
06
06
13
14
58
44 . 00
2 1 . 17
1 .39
67 . 5 0
7 . 65
2 6 . 20
1 3 3 . 00
4 . 05
.40
.10
. 35
. 22
. 27
. 24
.22
. 92
.73
. J4
.10
.73
,
62
1 8 . 00
2.45
2 . 71
. 49
. 10
. 84
.73
.15
1 . 15
. 82
. 17
. 03
. 34
.75
. 90
. 85
. 65
. 83
. 96
. 29
. 82
1 . 00
. 94
. 84
.65
1 . 18
. 62
.11
. 58
. 02
. 27
.24
. 61
. 61
. 43
. 3J
2 4 . 98
36 . 69
. 57
. 29
5 . 78
23 . 49
. 60
.35
. 12
. 09
. 54
. 43
1 . 02
.78
.21
. 03
. 53
. 61
. 82
. 32
. 30
. 86
. 81
.21
. 80
. 90
.20
.26
. 29
. 90
.
.
.
.
.
.
.
.
SLIP
1 2 . 00
1 . 49
DEFORM
. 70
. 9J
COATING
129
1
1
1
. 00
. 62
. 51
. 07
. 98
. 70
.43
. 09
.29
. 84
1 . 13
.75
. 34
1 . 00
1 . 04
1 . 19
.71
. 17
. 68
.75
. 21
ADJUST
SHOOT!!
STICK
.
.
1.
.
.
.
.
.
.
87
83
28
52
65
97
85
32
98
1 . 02
1 . 17
. 65
1 .17
1 . 08
. 74
1 . 08
1 . 00
1 . 04
. 80
1 . 43
1 . 20
J . 44
. 88
. 67
1 . 5J
. 96
.79
. 75
.70
1 . 11
SAMPLE
STICK
2 6 . 00
1 3 . 80
8 . 20
1 . 55
1 . 08
9 . 75
6 . 18
46.00
8 . 00
3 5 . 67
2 . 24
6 . 19
13 . 43
6 . 75
1 . 54
3 . 40
5.75
1 49 . 00
5 . 23
119 . 00
2 . 40
2 . 23
1 2 9 . 00
1 6 . 00
3 . 47
6 0 . 50
8 . 65
24 . 40
1 49 . 00
4 . 77
. 28
. 67
. 69
. 49
1 . 01
. 65
. 29
. 65
. 92
. 49
. 66
. 82
. 44
. 64
. 82
. 29
.45
. 50
.23
. 20
1 . 51
. 65
. 53
. 65
. 45
. 61
. 42
. 67
. 15
.4]
5 . 00
1 . 54
.75
.92
.98
.74
.19
. 92
. 31
. 53
1 . 13
. 96
. 26
1 . 00
. 96
1 . 8]
.71
. 84
.43
. 75
.91
. 84
. 69
.70
. 81
2 . 65
1 1 6 . 00
1 . 47
4 . 00
. 60
. 55
.79
. 44
.75
1. 12
.90
. 24
. 39
. 39
.70
. 86
1 . 00
. 89
. 35
. 96
. 20
. 82
.93
1. 00
. 84
1. 26
. 34
. 95
. 80
.23
. 57
. 09
1 . 09
.19
.76
98
64
40
34
69
.91
1 . 04
. 54
. 96
. 96
. 98
. 88
1 . 26
. 89
. 72
1 . 02
1 . 00
1.47
. 96
1 . 14
. 72
1 . 08
.13
. 39
1 . 17
. 47
. 61
. 85
. 52
. 6]
. 69
.26
28.97
4 5 . 30
. 57
. 28
4 . 98
2 0 . 99
. 68
. 32
. 84
. 31
.
.
1.
.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
557
557
557
557
557
557
557
557
557
557
557
557
557
557
557
557
557
557
557
557
557 .
557
557
557
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
MEAN
. 95
. 28
. 35
. 84
.73
.78
. 98
. 82
. 29
. 81
.78
1. 15
. 61
. 94
. 45
. 45
. 94
. 24
. 37
1 . 00
.47
. 11
1 . 22
.76
. 80
.73
. 61
.71
. 63
. 83
. 67
. 64
PANEL
557
557
557
557
557
557
SO
SAMPLE
PANEL
181
J81
2
3
4
5
6
7
8
9
10
11
12
13
14
J5
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
181
181
181
181
18]
181
181
181
18]
18]
181
181
]81
181
181
181
181
18J
18]
181
] 81
181
181
181
181
181
18]
181
MEAN
SO
SLIP
DEFORM
ADJUST
SMOO'nI
STICK
. 00
. 00
. 00
. 00
. 98
2 . 38
5 . 06
2 . 33
1 . 00
20 . 67
1. 00
1 . 69
5.29
1. 0 0
2.31
.45
1 . 60
1 1 . 00
4 . 38
5 . 00
.20
.44
1'. 0 0
4 . 00
3 . 00
1 . 00
. 06
4.20
17 . 0 0
1. 68
00
88
91
01
1 . 06
. 92
. 80
. 95
1. 00
. 92
. 66
. 97
. 89
1 . 00
. 71
. 98
. 86
. 83
. 27
1 . 00
1 . 60
1 . 10
.10
. 77
. 94
1 . 00
1 . 33
. 32
1 . 00
.29
. 00
. 77
. 70
. 48
. 98
. 87
. 78
1 . 05
. 31
. 90
1 . 22
1 . 00
. 45
1 . 00
. 94
1 . 24
. 90
. 17
. 88
. 75
1 . 25
1 . 08
. 31
. 83
. 60
. 21
4 8 . 00
2 . 24
. 62
1 . 00
. 92
1 . 00
. 92
. 92
1 . 07
. 91
. 83
. 83
1 . 00
. 86
1 . 00
1 . 00
. 94
1 . 00
. 96
. 98
1 . 00
1 . 00
. 90
. 84
1 . 33
1 .36
. 84
.90
.72
1 . 00
. 87
. 65
. 95
.93
. 95
. 76
1 . 05
. 13
. 74
. 67
. 94
1 . 16
. 98
. 98
. 98
. 58
1 . 03
. 89
. 82
1 . 10
1 . 00
1 . 32
. 72
1 . 43
. 53
.93
1 . 27
. 48
. 80
.31
. 46
. 92
1 . 00
. 77
1 . 00
. 19
3 . 52
4 .74
. 84
. 34
2 . 52
8 . 60
.95
. 14
. 86
. 29
DEFORM
COATING
ADJUST
SHOOT!!
1 . 00
. 33
. 58
. 15
. 97
. 77
.74
. 80
. 34
. 73
. 95
1 . 00
. 94
. 94
1 . 06
.63
. 90
1. 00
1 . 04
. 75
.48
. 93
.21
. 88
.46
.21
1. 00
1 . 00
. 03
. 89
1 . 00
1 . 00
. 80
. 54
. 94
. 88
.72
. 87
.96
. 83
1 . 00
. 93
. 97
1 . 00
. 96
1 . 02
1 . 00
. 23
. 98
' . 84
1. 20
1 . 36
1 . 06
1 . 00
. 87
.95
1 . 00
1 . 00
1 . 00
. 89
SLIP
3
1
2
1
COATING
1 . 00
1 . 07
. 91
1 . 09
1 . 08
. 98
. 93
1 . 06
1 . 00
1 . 04
1 . 00
1 . 00
1 . 05
1 . 00
1 . 07
. 98
. 91
1 . 00
. 98
1 . 00
1.45
1. 25
1 . 00
. 92
. 16
1 . 00
1 . 00
1 . 15
1 . 00
1 . 05
1 .
.
.
.
. 95
. 94
. 81
. 94
. 90
. 86
. 65
. 94
. 93
. 94
1 . 00
1 . 00
1 . 00
1 . 00
. 73
1 . 02
. 91
1 . 00
. 86
.73
1 . 10
1. 25
. 9J
1 . 00
. 89
. 91
. 83
1 . 00
. 86
1 . 00
1 . 00
2 . 00
9.40
2.73
1 . 18
6 .00
3 . 35
5 . 00
1 . 50
27 . 0 0
1 . 00
1 . 50
1 . 00
1 . 90
1 . 92
. 35
1 . 60
105 . 00
1 . 19
2 9 . 00
2 . 40
.21
18 . 00
2 . 67
4 2 . 50
1 . 00
1 . 00
3 . 00
3. 23
. 95
. 92
. 85
. 62
. 96
. 80
. 53
. 89
. 96
. 81
. 79
. 97
. 94
. 95
. 73
1 . 03
. 92
1 . 00
. 93
. 80
1 . 41
1 . 18
.33
.85
1 . 00
1 . 00
1. 00
1 . 00
. 92
.83
. 93
. 11
9 . 29
2 0 . 65
.90
. 19
1 . 18
1
1
1
1
. 72
. 31
. 93
. 19
STICK
1 . 15
1. 00
1 . 45
. 72
1 . 46
. 82
1 . 15
1 . 42
. 96
. 92
. 98
1 . 00
1 . 00
. 89
. 97
1 . 08
1 . 00
1 . 23
1 . 00
1 . 69
. 88
1 . 71
. 53
1 . 00
1 . 42
1 . 09
. 88
1 . 00
. 93
. 88
1 . 07
. 27
<Xl
Kasetsart
SAMPLE
Univers i ty St udents
PANEL
SLIP
. 77
DEFORM
5 . 00
. 37
1 . 08
.43
. 37
.91
. 84
. 88
. 07
1 . 00
1 . 49
1 . 14
.25
. 91
. 98
. 62
. 45
.75
. 97
. 95
. 24
. 98
. 87
. 92
1 . 51
. 95
1 . 18
. 54
. 47
847
847
847
847
.20
.95
6
7
.78
. 57
1 1 . 38
6 . 76
847
. 70
1 5 . 33
847
9
10
11
12
13
. 81
.36
3 . 13
9 . 71
. 86
. 83
. 94
5 . 10
14
. 82
5 . 33
. 93
. 34
. 36
17 . 00
1 . 38
4 . 05
. 38
. 36
.86
. 52
. 68
. 94
3 . 15
.73
2 . 55
. 91
. 15
1 . 00
. 95
.75
. 92
1 . 02
.71
.72
.71
847
15
16
17
847
18
. 96 .
1 4 0 . 00
. 46
19
. 65
4 . 19
. >2
847
. 92
847
20
. 26
847
21
22
. 69
. 68
85 . 00
1 5 . 80
1 . 74
. 39
.71
23
. 06
. 80
1 . 00
847
247
. 80
847
2.
. 68
140 . 00
26
27
. 64
. 44
3 5 . 00
847
847
847
29
847
30
25
847
847
28
MEAN
SO
SAMPLE
PANEL
428
4 28
42&
428
428
428
4
5
6
7
8
9
10
428
428
428
428
428
428
428
. 38
5 . 36
1 . 47
.75
. 81
. 50
. 53
. 77
. 69
. 8J
1 . 00
. 63
23 . 51
. 71
. 29
39 . 9 6
.23
3 . 21
1 2 . 46
.86
.85
.70
. 53
29
30
. 96
. 08
00
32
59
57
97
93
.70
. 48
. 34
. 95
1 . 00
SMOOTH
. 99
. 35
1 . 05
. 39
. 98
. 86
1 . 17
. 55
. 64
. 81
. 53
1 . 00
. 96
. 81
. 51
.70
1 . 31
.71
. 97
. 80
1 . 04
. 89
. 82
1 . 00
1 . 00
.75
. 84
1 . 14
1 . 11
1 . 08
. 93
.41
1 . 25
1 . 36
1 . 33
1 . 26
54 . 00
2 2 . 67
1 . 10
. 18
. 12
2 . 13
5 2 . 00
. 40
. 48
. 25
3 . 65
. 45
. 52
.36
1 8 . 60
6 4 . 00
5 . 36
. 56
.30
1 5 . 19
1 8 . 31
. 84
1 . 00
. 84
1 3 . 50
.18
314
. 12
1 3 . 64
. 19
314
314
314
5
6
7
. 15
. 44
.22
. 53
. 06
. 03
1 5 . 25
8 . 06
. 65
. 17
4 8 . 67
.11
. 06
2 3 . 50
. 45
.29
4 8 . 33
3 . 05
.18
314
314
9
10
11
12
314
314
314
13
314
314
314
14
15
16
17
314
314
18
314
314
19
20
21
22
314
23
2f
.45
.11
. 01
. 10
.21
. 12
. 19
2 . 50
2 1 . 43
6 . 30
1 1 . 54
3 . 13
6 . 75
. 09
. 06
1 5 0 . 00
5 . 46
. 88
.72
. 01
. 11
. .52
. 24
.15
. 07
. 34
. 01
150 . 00
. 03
. 06
27 . 20
.24
. 05
2 . 91
1 5 0 . 00
2 4 . 67
. 05
. 01
. 02
. 01
. 01
. 26
. 02
.25
. 09
314
314
25
314
314
26
27
.70
314
28
29
. 01
2 4 . 20
. 01
. 01
314
. 01
1 4 7 . 00
. 01
314
30
. 03
6 . 32
. 06
. 12
. 16
3 5 . 43
4 8 . 57
. 27
.26
MEAN
SO
3 . 24
7 5 . 00
7 . 65
1 . 62
. 56
. 42
. 34
.47
. 48
. 09
SMOO'I'II
STICK
. 01
.74
1 . 08
.20
. 25
.70
.79
1 . 17
.75
. 06
60
35
11
.16
34
. 10
.
.
.
.
. 29
. 34
. 10
. 81
. 36
. 10
. 95
. 36
. 86
. 82
. 88
. 93
. 62
1 . 16
. 97
. 60
1 . 11
. 10
. 57
. 10
. 09
. 01
. 87
1 . 01
. 62
.70
. 94
. 89
. 84
1 . 05
. 64
. 53
. 81
. 02
. 90
4 . 41
1 50 . 00
. 50
.28
. 93
. 04
. 75
. 48
. 01
.10
1 . 39
. 26
. 01
3 . 95
4 . 41
. 13
. 01
. 02
. 10
. 01
. 73
1 . 17
1 . 00
1 . 47
. 83
.57
.01
. 11
1 . 00
. 05
5 . 96
27 . 23
. 45
. 38
. 58
. 41
.50
. 80
. 95
1 . 06
1. 4 2
1 . 00
. 97
1 . 13
4 . 55
6 . 00
2 . 77
ADJUST
. 57
.7]
. 85
. 93
COATI NG
4 9 . 33
1 4 . 10
1 . 35
.4S
. 75
. 79
. 92
DEFORM
STICK
03
89
63
56
04
88
65
25
.
.
.
.
1.
.
.
.
1 . 69
1 . 16
. 82
.21
. 32
. 17
138 .
1.
1.
.
.
.
.71
. 27
2 . 14
5 . 10
8 . 23
1 . 09
28
. 01
.75
. 75
.41
. 94
.76
. 48
.35
. 71
. 93
. 91
2 . 24
22
23
428
428
MEAN
SO
. 89
. 41
3 . 82
1 . 23
4 . 25
5.76
3 1 . 33
1 . 50
3 0 . 67
20 . 40
24
428
. 63
. 65
.62
. 88
.90
. 59
.11
a . 67
1 0 . 80
1 0 . 90
2 9 . 00
25
26
27
. 60
. 39
.29
42.8
428
428
. 63
. 01
. 09
. 04
314
314
1 . 10
. 18
SLIP
3
4
3 14
1 14 . 00
ADJUST
1 . 36
1 . 61
. 25
. 28
. 22
COATING
314
314
1 . 69
1 . 00
DEFORN
. 84
1 . 14
.76
PANEL
314
1. 23
. 91
SLIP
SAMPLE
.89
. 87
.20
21
428
428
428
1 . 00
6 9 . 00
2 . 03
19
20
18
. 94
.76
.9]
. 83
. 98
. 29
.73
428
1 . 17
6.76
2 1 . 80
. 86
. 81
15
16
17
. 56
. 88
1 . 04
1 . 06
1 . 09
12
428
428
428
. 55
. 55
. 83
. 96
. 97
. 82
. 04
. 52
. 79
.70
13
14
.73
. 92
.75
. 36
. 64
.65
. 85
. 89
. 44
. 21
. 57
2 . 38
11
428
428
1 1 . 50
1 . 34
. 29
. 89
. 68
. 04
428
428
STICK
. 88
. 47
1
2
3
847
847
847
847
SHOOTIi
9 . 00
4 . 60
9 . 82
847
847
847
847
ADJUST
. 67
847
847
.43
. 43
COATING
1 4 . 33
1 . 04
1 . 26
. 89
93
.
.
1.
1.
90
00
04
. 92
. 19
. 64
2 . 88
. 75
. 20
.47
. 52
1 . 10
. 51
18 . 00
. 53
2 . 87
. 45
.70
1 . 18
. 21
. 88
. 55
.26
1 . 19
. 68
. 30
6 . 12
.70
. 97
25.11
. 30
. 32
1 . 09
. 17
1. 63
. 91
. 35
. 62
. 58
\0
OFFICE WORKERS
SAMPLE
PANEL
742
742
742
742
742
742
742
742
742
742
742
742
742
742
742
72
742
742
742
742
742
742
742
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
742
24
742
742
742
742
742
742
25
26
27
28
29
30
MEAN
SD
SAMPLE
PANEL
1
2
3
4
5
6
7
8
9
10
II
12
13
14
15
.
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
243
243
243
243
243
243
243
243
243
243
243
243
243
243
243
243
243
243
243
243
243
243
243
243
243
243
243
243
243
243
MEAN
SD
SL P
. 06
.71
. 29
. 15
. 42
1 . 03
. 17
.10
.16
.46
.29
. 02
. 25
.40
. 17
. 25
. 07
. 53
. 06
. 53
. 50
. 63
. 78
DEFORM
COATING
3 . 62
2 . 40
1 3 . 80
57 . 00
15.71
4 5 . 00
.28
4 . 50
6 . 62
5 . 27
1 . 30
1 6 . 33
3 . 62
9 1 . 00
2 . 92
14 . 25
3 00
1 . 24
24 80
5 . 67
3 . 45
. 40
1 . 14
. 60
. 62
. 47
.78
.78
. 66
1 . 19
. 48
.70
1 . 02
.41
. 17
. 47
. 69
. 05
1 . 12
1 . 10
. 38
. 24
. 64
.79
. 42
. 77
1 8 . 75
ADJUST
SHOOTH
.29
.70
. 18
. 97
. 93
.42
.25
. 46
. 07
.48
. 52
. 06
. 50
. 78
. 97
.51
. 19
1 . 00
. 06
. 51
. 53
.28
1 . 12
. 72
. 83
. 60
. 06
.10
. 26
. 54
.16
. 88
. 98
1 . 17
.75
3 . 23
1 . 03
. 56
. 73
. 55
. 69
. 68
. 03
2 . 19
. 41
1 . 15
. 83
. 33
. 98
. 75
. 77
1 . 00
1 . 07
.23
. 87
1 . 48
. 62
. 50
. 32
. 93
. 61
16
30
01
78
12
22
4 8 . 33
6 . 20
1 . 00
1 . 54
1 0 2 . 00
44 . 00
. 52
. 46
.80
. 02
. 35
. 25
. 57
.40
. 74
. 25
. 97
. 58
. 50
. 27
.47
.10
. 86
. 52
. 07
.46
. 54
. 96
.48
. 34
. 53
.75
. 55
.21
.28
1 .26
. 73
. 04
. 92
. 04
. 65
.17
. 62
. 35
. 32
1 8 . 17
2 6 . 54
. 58
. 30
. 51
. 30
. 01
.
1.
.
.
.
.
SLIP
. 18
.74
. 47
. 24
. 42
. 47
1 . 00
.84
. 59
. 40
. 15
. 55
. 35
. 15
. 65
. 06
.70
. 03
. 12
. 06
. 89
.76
.15
. 17
. 01
1 .11
. 01
.78
. 12
. 32
. 42
. 32
DEFORM
COATING
ADJUST
SHOOTH
7 . 10
2 . 40
1 1 . 90
57 . 00
19.14
1 2 6 . 00
1 . 00
5 . 44
5 . 29
1 0 . 64
3 . 85
3 . 22
3 . 28
1 1 5 . 00
1 . 58
17 . 88
3 . 29
1 . 56
27 . 8 0
1 1 . 50
1 . 55
. 95
1 . 27
1 4 . 88
48 . 33
9 . 90
14 . 0 0
1 . 83
147 . 00
82 . 00
. 13
. 62
.69
. 78
. 78
. 97
. 85
. 58
.76
. 98
. 35
. 67
. 15
.22
. 61
. 76
. 46
. 66
. 45
. 10
. 85
.81
. 33
.44
.25
. 93
. 18
. 63
. 25
. 29
. 09
. 69
. 60
. 97
. 58
. 58
2 . 73
.61 .
. 31
. 90
. 41
.72
. 30
. 25
. 48
. 35
. 90
. 74
1 . 00
1 . 17
. 53
. 95
1 . 44
. 54
.18
. 36
. 13
1 . 21
1 . 60
. 47
.21
. 70
. 39
. 97
. 93
. 23
1 . 15
.96
. 13
. 42
. 87
.70
. 42
. 82
. 61
. 26
1 . 00
. 32
. 29
. 85
. 53
. 57
1 . 39
. 78
. 83
. 94
. 18
. 67
. 15
. 31
2 5 . 22
4 0 . 03
. 55
.27
. 73
. 62
. 33
. 53
STICK
1 . 52
. 96
1 . 33
STICK
. 18
. 49
1 . 00
. 95
. 88
. 94
. 75
. 63
. 71
1 . 11
1 . 10
1 . 36
.30
. 64
. 95
. 24
3 . 78
1
1
1
1
. 25
. 59
. 66
. 33
. 89
. 83
. 86
. 00
. 19
. 97
. 87
. 28
. 12
. 90
. 63
SAMPLE
PANEL
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
557
557
557
557
557
557
557
557
557
557
557
557
557
557
557
557
557
557
557
557
557
557
557
557
557
557
557
557
557
557
MEAN
SD
SAMPLE
PANEL
1
2
3
4
5
6
7
8
9
1 81
181
181
181
181
181
181
181
181
181
181
181
181
181
181
181
181
181
181
181
181
181
181
181
181
181
181
181
181
1 81
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
MEAN
SD
SLI P
1 .08
. 83
1 .00
DEFORM
ADJUST
SMOOTH
1 . 09
. 94
1 . 03
. 97
. 93
. 97
. 43
. 82
1 . 00
. 93
. 77
. 80
. 63
. 77
. 85
. 12
1 . 13
. 43
1 . 00
. 70
1 . 00
. 91
1 . 56
.9
1 . 00
1 . 00
. 65
. 51
. 91
1 . 00
. 62
. 96
1 . 53
. 05
. 84
. 93
. 54
. 85
1. 00
. 82
. 82
. 89
. 36
1 . 09
.75
. 51
. 97
.66
. 87
. 78
. 16
. 91
. 88
1 . 00
. 97
.73
. 36
5 . 65
. 83
1 . 00
.97
. 91
. 90
. 78
. 67
. 68
. 20
1 . 00
. 63
. 34
. 74
1 . 00
1 . 08
1 . 14
. 92
. 38
. 92
. 60
1 . 09
1 . 60
1 . 00
:i
1 . 00
.70
1 . 27
. 05
. 78
. 72
. 62
. 95
1 . 00
1 . 07
.15
1 . 05
. 78
.35
. 72
. 32
1 . 00
. 92
. 86
. 26
.76
. 36
COATING
ADJUST
SHOO'ni
. 87
1 . 00
. 62
. 97
. 93
. 92
1 . 34
, 93
.90
. 98
. 92
. 80
. 73
. 84
. 61
1 . 00
. 41
. 72
.76
. 10
. 85
. 97
1 . 69
1 . 04
. 48
1 . 00
. 92
1 . 00
. 91
. 92
. 39
. 96
1 . 00
. 9
. 88
1 . 00
. 90
1 . 06
1 .23
1 . 10
1 . 22
1 . 36
. 66
1 . 27
. 58
. 93
2 . 70
1 . 00
1 . 00
1 . 00
. 78
. 98
.40
1 . 04
1 . 00
1 . 16
.75
1 . 00
1 . 48
1 . 26
. 87
. 27
1 . 03
.40
COATING
1. 00
. 89
1 .00
.35
1. 0 0
1 . 21
1 . 00
. 63
1 . 00
1 .78
. 92
. 92
. 92
1 . 00
. 48
1 . 07
3 . 20
4 . 50
1 . 86
3 . 00
1 . 99
2 . 44
1 . 00
5 . 55
1 . 91
2 . 78
2 . 41
3 4 . 00
. 13
2 . 25
. 44
. 29
4 . 00
7 . 83
1 . 00
1 . 24
1 . 00
. 38
1 . 0 0.
. 10
132 . 00
1 . 25
1 4 . 00
19 . 00
1 . 06
. 86
1 . 00
. 19
. 72
. 97
.29
. 84
1 . 00
. 42
1 . 07
.72
.25
. 78
. 32
. 65
. 59
. 54
1 . 00
. 30
. 95
1 . 19
1 . 39
1 . 01
. 30
8 . 40
2 4 . 36
.96
2
1
1
1
.91
. 99
. 03
. 2
. 00
. 04
1 . 16
.
.
.
1.
85
62
87
21
. 88
SLIP
DEFORM
. 66
. 66
1 . 00
. 83
. 72
. 42
. 61
1 . 36
1 . 42
. 63
. 98
1 . 04
. 95
. 69
. 35
1 . 35
1 . 00
. 30
1 . 00
. 78
1 . 00
.89
1 . 17
. 93
. 83
. 92
1 . 68
. 71
1. 00
. 71
. 86
3 . 24
. 78
8 . 60
2 1 . 00
1 5 . 71
3 0 . 00
1. 34
. 56
1 . 43
2 . 09
2 . 15
2 . 33
2 . 07
7 1 . 00
. 13
1 . 00
1 . 00
1 . 00
1 . 00
1 . 00
1 . 55
. 23
1 . 70
1 6 . 75
1 . 00
1 . 80
1 4 2 . 00
1 . 00
7 8 . 00
7 . 00
. 56
. 94
. 80
.78
. 78
. 52
. 52
. 94
. 93
.90
1 . 22
.79
. 62
. 73
. 41
1 . 24
. 88
1 . 00
. 89
1 . 00
. 85
1 . 27
1 . 00
. 36
. 51
. 97
. 92
1 . 00
. 66
.76
. 53
. 95
. . 72
. 97
. 58
. 58
4 . 92
. 95
. 85
. 93
1 . 04
. 84
. 65
. 38
. 48
1 . 00
. 64
1 . 00
. 93
. 10
1 . 36
1 . 14
. 92
. 98
1 . 00
1 . 07
. 95
1 . 00
. 48
. 89
. 89
. 31
1 3 . 95
3 0 . 79
. 83
.23
. 96
. 79
STICK
.ll
STICK
OFFICE WORKERS
PANEL
SAMPLE
847
847
. 67
. 47
.90
.71
1. 40
10 . 10
2 1 . 00
. 86
. 94
. 78
. 82
.91
9 . 14
57 . 0 0
. 78
.46
. 97
. 73
.49
847
6
7
847
847
847
847
8
9
10
11
. 61
1 . 00
. 33
.70
. 50
.91
847
847
12
13
847
847
14
15
847
16
847
847
847
17
18
19
20
21
22
:n
847
841
24
25
847
26
27
847
847
847
28
29
30
847
847
PANEL
428
428
428
428
428
428
428
428
4
5
6
7
8
9
10
11
428
428
428
12
13
14
428
428
428
428
15
16
428
17
18
19
428
428
428
20
21
22
428
428
428
428
428
23
24
25
428
428
428
26
428
428
28
29
428
30
27
MEAN
SO
.23
. 16
.42
. 84
. 33
. 53
. 50
. 67
. 50
. 38
, 85
.
1 .
.
.
.
.
77
92
36
15
39
23
76
SLIP
. 02
.76
.47
. 54
. 66
. 72
. 77
1.11
. 51
. 63
. 83
.61
. 81
. 37
.41
. 14
. 53
. 39
. 22
. 86
. 50
. 92
. 68
. 72
. 10
1. 59
. 08
. 25
. 33
.13
. 56
. 34
SMOOTH
STICK
School
s t udents
SAMPLE
PANEL
.42
. 46
742
. 85
. 96
. 00
. 95
. 88
. 97
. 86
.97
.93
. 74
742
742
1 0 . 18
11 .75
742
4
5
6
7
. 03
.24
7 2 . 00
742
742
1 . 00
1 . 00
1 . 00
1 . 00
.31
. 42
. 83
. 63
. 44
. 61
. 54
. 83
742
. 39
1 . 07
. 86
. 51
2 . 19
1 . 06
. 42
1. 10
. 75
742
742
83 . 0 0
. 63
1 . 58
1 3 . 00
2 . 20
. 53
. 68
1 . 00
. 73
. 60
. 42
. 36
. 13
. 77
.76
. 82
.71
.93
. 78
.78
. 83
. 13
1 . 00
1 . 36
. 57
20
9
4
1
1 , 53
2 . 88
1 . 00
8.
105.
.
102 .
32 .
60
00
74
00
00
17 . 46
2 9 . 44
DEFORM
. 81
. 36
1 . 00
. 59
1 . 00
. 80
. 82
.76
. 80
. 60
15
. 38
. 01
17
. 23
. 06
. 85
1 , 33
1 , 23
. 90
742
742
742
742
742
.86
1. 15
742
1 . 00
742
742
742
742
742
742
. 82
. 51
.48
.50
. 89
. 76
. 19
. 76
.30
.71
. 27
. 04
. 55
COATING
. 96
.49
.45
. 35
. 88
. 19
.87
1.76
.75
. 86
16 .60
3 . 33
4 . 77
. 59
. 50
. 79
1 . 05
. 09
1 . 85
1 7 . 88
1 . 00
. 66
.81
. 00
1. 80
120 . 00
. 83
ADJUST
. 14
. 69
. 53
. 97
. 73
. 40
4 . 92
. 74
. 35
. 88
.77
.
.
.
.
51
21
49
96
. 09
. 52
SMOOTH
STICK
. 14
.70
. 52
. 97
.93
. 48
.25
.75
. 22
. 27
. 89
.43
. 32
. 82
. 81
. 97
. 04
. 90
. 02
. 83
. 57
. 83
. 93
.23
. 30
. 21
. 17
. 35
. 85
1 . 05
1 . 00
1. 00
. 96
.75
. 95
. 88
. 97
. 68
. 97
. 77
.
.
.
.
89
43
16
95
. 44
. 55
. 21
. 24
23
24
25
26
27
28
29
30
MEAN
SO
SAMPLE
PANEL
1
2
3
4
5
6
7
243
243
243
243
243
243
243
243
243
8
9
10
11
243
243
243
243
243
243
243
12
13
14
15
16
243
243
. 90
243
. 03
. 67
243
243
28
29
. 00
243
30
. 35
. 26
. 18
1 9 . 40
33. 74
. 73
.76
. 62
.30
. 84
. 34
. 09
. 73
1 . 00
. 15
41.67
. 15
. 23
. 68
20 . 33
1 . 32
1 . 83
7 . 53
COATING
ADJUST
DEFORM
12 .
10 .
10.
74 .
1 09 .
11.
3 .
93 .
149 .
18
50
23
50
00
54
17
00
00
8 . 00
27 . 20
1 .71
1 . 50
1 6 . 67
2 . 24
3 . 80
3.81
17 . 1 3
.43
. 23
. 13
.61
. 34
. 53
. 67
. 79
. 03
. 77
. 88
. 81
. 83
.01
. 49
. 44
1 . 38
. 28
. 72
.01
. 00
. 30
. 87
.72
71
00
00
50
1 . 30
. 72
56. 00
.74
1 . 04
16 .
100 .
2.
27 .
1 3 1 . 00
243
243
243
.76
1. 23
. 29
1 . 92
. 57
. 21
1 . 00
1 . 16
.48
126 . 00
.51
. 48
. 93
. 81
. 60
. 38
1 . 07
25
26
27
. 84
. 25
34 . 2 2
4 3 . 48
. 88
1 .69
. 89
. 85
. 47
. 66
. 62
. 24
.22
. 06
.91
. 72
1 . 19
. 60
. 05
.53
. 49
. 01
.20
.76
. 39
. 92
. 54
. 89
.31
. 25
. 16
. 26
1 04 . 00
43 . 3 3
9 . 88
29 . 3 3
146 . 00
6.23
. 81
.31
. 42
. 87
. 83
1 . 86
.45
.21
1 . 45
. 15
. 42
.61
. 32
. 46
. 69
SLIP
. 87
76
33
03
00
. 08
. 87
.11
. 42
. 51
.43
. 87
. 64
.42
.
.
.
.
. 15
2 . 62
5 . 86
68 . 0 0
3 . 52
6 4 . 00
1 7 . 67
23
24
. 62
. 28
1 . 00
.41
1 . 59
2 . 00
6 . 60
8 . 25
1 . 00
. 44
. 76
. 45
1 . 24
. 79
. 20
. 00
. 98
.76
.49
.50
. 03
. 13
.01
. 35
18
. 92
. 60
. 65
. 27
.31
1 . 04
.37
8 . 22
2 . 19
. 34
. 82
1 . 00
1 . 10
. 86
. 63
. 19
. 90
17
19
20
21
22
243
243
243
243
243
. 58
. 46
. 03
1 . 04
742
18
19
20
21
22
. 25
. 65
. 84
1 . 00
742
. 03
40 . 33
3 . 40
. 42
14
. 85
. 18
1 . 00
.22
.77
. 01
. 90
. 95
.97
. 44
742
16
. 68
. 78
. 49
.70
.11
. 93
. 27
742
742
. 13
. 52
. 33
.26
STICK
. 18
. 70
. 37
.
1 .
.
.
SMOOTH
. 61
. 84
. 23
8 . 08
107 . 00
. 07
. 36
ADJUST
50
00
25
74
1 2 3 . 00
13
3 . 2"2
. 85
. 82
. 53
1 . 00
.31
97 . 00
1 0 . 15
. 49
. 51
. 24
.01
742
.75
. 82
.22
. 75
.78
4.73
1 . 64
7 . 11
1 . 83
. 52
. 81
.78
9 . 15
COATING
.28
.45
742
742
742
. 85
. 89
. 43
.77
. 17
. 62
. 59
.78
.78
. 67
.70
.75
1 . 15
15. 63
. 33
.41
. 34
. 07
8
9
10
11
12
1. 00
1 . 12
. 84
. 93
. 67
. 50
6.76
2 . 40
13 . 80
5 1 . 00
9 . 14
30 . 00
. 28
1 . 56
2 . 24
78 . 0 0
2 . 92
. 55
. 79
DEFORM
.28
.22
.12
1 . 00
. 72
SLIP
1
2
3
3 . 56
1. 81
5 . 00
3 . 48
13 . 44
4 . 10
. 19
. 58
. 29
MEAN
SO
SAMPLE
ADJUST
2 . 95
. 54
. 66
847
847
847
COATING
.28
4
5
847
847
DEFORM
847
847
SLIP
. 12
. 41
. 36
. 81
. 83
1 5 . 67
. 27
. 65
. 62
6 . 13
. 86
48 . 3 3
. 18
. 93
. 95
4 . 00
5 . 68
. 88
MEAN
. 76
30 . 88
SO
. 43
41. 73
.78
. 36
. 65
. 67
.
.
.
.
.
.
1 .
69
74
12
82
11
23
04
. 16
. 16
.
.
1 .
.
1.
74
23
25
87
52
. 69
1 . 04
1 . 85
.
.
.
1 .
.
28
65
74
05
36
35 . 33
1. 26
. 52
1 . 01
. 83
. 61
. 94
. 07
.51
. 69 .
. 54
. 55
. 67
. 58
. 30
SMOOTH
.21
. 68
. 12
. 03
. 54
. 86
. 79 '
. 58
. 68
. 93
1 . 00
1. 00
. 42
.81
. 63
1 . 04
. 31
. 11
. 91
. 80
. 96
. 89
. 67
. 50
.33
1 .03
.49
1 .23
.77
. 77
. 76
27
1 . 39
. 33
1 .64
.28
. 89
1 . 00
. 90
. 94
.49
STICK
. 67
1 . 10
.38
. 86
. 80
. 01
.91
. 20
. 09
1 .29
. 15
.23
. 83
. 07
.63
1 . 05
1 . 92
. 92
1 . 43
. 40
. 30
. 89
. 80
1 . 12
. 19
. 96
.78
.78
. 95
. 02
. 52
.01
. 94
. 81
1 5 . 00
.43
. 19
1 . 00
1. 29
. 67
. 27
. 53
1 . 89
1 . 16
. 32
6 . 33
. 64
. 31
2 . 65
tv
U1
.....
School
students
PANEL
SAMPLE
847
SLI P
.41
847
1
2
847
847
3
4
. 80
5
6
.71
. 94
847
847
847
847
847
847
847
847
847
847
847
847
847
847
847
. 62
. 67
15
16
17
18
19
20
21
22
23
6 47
847
847
847
PANEL
1
1
11
2
7
.
.
.
.
.
17
18
19
20
21
22
23
24
25
26
27
28
29
30
30
00
67
00
20
5 . 50
5 . 50
1 9 . 29
85 . 00
1 . 38
6 8 . 00
2 2 . 17
. 79
21 . 00
ADJUST
SMOOTH
. 57
. 89
. 91
. 58
.69
. 52
. 89
. 92
.40
. 98
. 92
. 21
. 54
. 73
. 88
.76
. 63
. 55
. 68
. 39
. 53
. 19
. 73
. 33
. 86
I . 23
. 63
. 64
. 82
. 75
. 63
.71
. 60
. 78
. 52
. 86
. 92
.21
. 67
. 34
1 .51
. -4 1
. 36
. 78
. 38
. 56
1 . 00
. 13
. 37
. 60
. 15
2 . 33
1 . 09
. 64
. 35
. 94
.70
. 93
. 53
. 17
.81
. 89
. 91
. 78
.
.
.
.
08
29
29
09
1 . 00
. 89
. 79
. 87
1 . 13
. 81
STICK
. 97
1 . 50
. 47
. 58
. 95
1 . 65
1 . 11
. 57
. 19
1 . 76
.
.
.
.
.
85
92
90
36
94
1 .21
1 . 65
. 97
. 77
. 49
. 97
. 94
. 17
. 89
. 41
I . 24
.79
.76
. 44
. 61
. 56
. 54
. 86
. 75
. 17
. 67
. 95
. 42
. 73
. 60
9 . 67
6 . 13
1 9 . 00
67 . 00
4 . 05
I . 00
.
.
97 .
.
. 65
.30
1 8 . 31
2 3 . 14
. 67
. 23
. 67
.43
. 69
. 29
4 . 08
1 7 . 55
. 65
. 70
.44
SLIP
DEFORM
. 84
15
16
SO
3 0 . 00
7 . 20
.63
. 85
12
ME:AN
4 . 33
6 6 . 00
3 0 . 00
8
9
13
14
428
428
428
428
. 61
7 . 50
2 8 . 00
2 . 92
5
6
7
10
11
428
428
428
428
428
. 12
. 49
.73
9.
8.
69 .
68 .
1.
5.
428
428
428
428
428
428
428
428
428
. 31
. 87
COATING
8 . 00
3 . 83
3 . 77
.11
.42
. 26
. 12
. 18
.61
. 74
1
2
3
4
428
428
428
.51
1 . 00
. 32
428
428
428
428
428
428
428
428
428
.78
. 68
. 20
1 . 13
26
27
ME:AN
SO
SAMPLE
1 . 26
. 97
1. 23
28
29
30
847
847
. 69
. 48
.11
24
847
. 69
8
9
10
11
12
13
14
847
847
847
847
.61
DEFORM
. 03
1. 85
. 78
. 03
. 55
. 38
1. 42
. 24
. 56
.41
. 67
. 25
. 94
1. 23
. 46
. 60
1 1 . 36
COATING
ADJUST
. 98
.41
4 4 . 00
1 9 . 00
. 65
. 34
. 85
. 45
.25
1 . 15
.73
.74
.33
. 38
. 88
. 14
. 97
.78
. 76
22 . 3 3
. 66
. 60
. 11
1 . 12
5
1
1
3
.
.
.
.
50
38
00
00
54
25
00
86
23
22
. 18
1 . 04
.35
1 . 55
1 . 75
. 55
. 15
4 . 75
1 4 . 88
1 2 . 43
1 . 00
6 8 . 00
2 . 81
8 . 00
2 . 50
. 91
57 . 00
2 0 . 00
. 43
. 49
.
.
.
.
.
1.
.
63
48
92
94
32
08
89
. 59
. 86
. 54
I . 00
. 84
. 87
. 59
. 89
. 92
1 . 00
. 81
. 50
. 37
. 74
. 67
SMOOTH
. 49
. 52
. 37
. 64
. 64
. 98
.71
. 87
.83
. 60
. 07
1 . 00
.42
. 38
.73
1 . 41
1 . 75
. 30
.77
. 43
. 53
.71
. 92
. 32
2 . 21
.40
. 92
. 63
. 23
. 86
I . 36
. 90
. 42
. 49
1 . 02
. 83
. 94
. 67
. 44
. 73
. 65
. 77
. 13
. 90
.70
I . 00
1 . 00
. 80
. 76
1 . 00
.45
3 . 88
1 4 . 67
. 86
. 48
. 26
.36
. 64
.41
1 6 . 50
. 67
1 . 69
. 64
2 1 . 54
. 29
5 . 73
. 24
9 . 00
3 . 09
STICK
. 65
.48
. 55
. 50
3 2 . 00
. 97
. 90
. 10
I . 00
58
63
00
78
1 . 92
. 63
1 . 60
1 . 24
. 59
1 . 05
. 93
.71
150 . 00
1 . 00
5 . 92
27 . 22
PANEL
SAMPLE
557
557
557
1
2
3
4
557
557
5
6
557
557
7
8
9
10
11
12
557
557
557
557
557
557
557
557
557
557
557
13
14
15
16
17
18
19
20
21
22
23
557
557
557
557
557
557
24
25
26
27
28
29
557
557
557
557
557
557
30
MEAN
SO
PANEL
SAMPLE
181
181
181
181
181
181
181
181
4
5
6
7
8
9
10
11
12
181
181
181
181
13
14
15
181
181
181
181
16
17
181
181
181
181
181
181
181
181
181
181
181
181
SO
. 90
1 . 07
1 . 00
. 80
1 . 89
1 . 08
1 . 00
. 94
1. 03
. 84
2 . 00
1. 04
1 . 08
. 80
. 81
2 . 42
. 13
.65
1 . 00
1 . 30
DEFORM
COATING
1 . 18
2 . 67
.66
ADJUST
. 98
SMOOTH
. 99
1 . 00
1 . 00
.81
.42
. 77
1 . 00
1 . 16
1 . 00
1 . 00
. 88
. 25
. 22
1 . 00
. 80
. 46
.70
1 . 34
. 14
. 86
. 55
. 86
. 68
.13
. 69
. 93
. 96
. 18
1. 36
.41
. 94
1. 07
1 6 . 00
1 . 67
5 . 00
1 3 9 . 00
3 . 67
2 4 . 60
. 26
3 . 62
1 . 67
1 . 00
. 45
6 . 56
2 . 50
2 0 . 29
6 . 00
1 . 17
. 18
.78
.43
.23
. 27
.78
.31
. 55
. 24
. 35
. 23
I . 05
. 05
. 38
. 84
.52
. 15
29.33
. 83
. 96
. 88
1 . 08
. 60
1 . 44
.
.
.
.
88
61
42
88
I . 00
. 89
. 55
. 86
. 85
.71
. 84
. 53
1 . 40
. 94
. 96
1 . 00
.81
. 01
1 . 28
1 . 50
9 . 00
1 5 0 . 00
1 . 00
1 . 04
. 95
. 63
. 09
. 31
1 . 00
. 90
. 54
. 68
. 94
1 . 19
. 96
. 93
1. 00
. 84
. 20
1 . 00
1 . 01
.49
1 6 . 34
36 . 78
.62
. 32
1. 66
5.24
. 81
.21
COATING
ADJUST
. 88
. 14
SIrIP
. 94
1 . 04
1 . 00
. 75
. 66
1 . 76
. 85
. 63
.75
. 97
. 77
1 . 00
.49
. 97
. 98
.53
1 .
1.
.
.
1 .
.
23
24
25
1 . 07
. 86
29
30
t<E:AN
. 98
18
19
20
21
22
26
27
28
181
181
SLIP
89
00
94
82
14
61
. 08
. 19
1 . 00
1 . 00
. 83
1 . 10
. 89
. 36
59 . 50
1 0 . 67
. 23
1 6 . 00
2 . 67
DEFORM
. 43
. 59
.63
. 94
.81
. 97
2 . 00
. 58
2 . 38
6 5 . 00
7 . 00
1 . 00
1 . 42
4 4 . 00
10 . 00
. 80
. 96
. 87
.79
. 93
. 09
1 . 30
.73
. 89
7 . 00
2 2 . 60
1 . 00
. 85
.71
2 . 27
1 2 . 78
. 66
. 20
1 . 88
12.
8.
30 .
1 .
8.
63
00
00
29
00
8 . 83
. 86
2 6 . 00
27 . 3 3
1
1
1
1
1 . 00
.56
. 20
. 87
. 58
.45
. 91
. 80
. 74
. 67
. 44
. 98
1 .23
.19
. 92
. 00
. 00
. 00
. 82
1 . 00
1 . 00
. 95
10 . 32
1 5 . 04
.78
1 . 00
. 28
.
1 .
1.
1.
.
.
96
15
00
00
17
31
. 95
. 83
.76
.72
. 31
. 88
. 87
.73
1 . 19
. 07
.93
. 93
.73
SMOOTH
. 98
. 90
.78
. 82
. 82
. 07
. 92
. 87
. 91
. 89
STICK
1 .46
. 33
. 88
.01
. 36
.21
.75
.82
. 02
.29
.79
.31
. 57
. 30
.71
. 08
1 . 08
. 37
1 . 12
.40
. 24
. 94
. 09
.77
.77
. 14
. 82
. 55
2 7 . 00
.17
I.41
4 . 85
STICK
1 . 56
.65
1 . 00
1 . 00
.39
. 38
1 . 02
.92
.91
. 63
1 . 05
. 74
1 . 00
. 42
1 . 00
I . 00
. 24
. 92
. 95
.91
.56
. 5 1
. 55
. 92
. 65
.73
. 85
1 . 00
1 . 00
. 56
. 84
.11
. 50
46 . 00
. )1
. 85
. 29
1. 0 0
1 . 00
1 . 09
1 . 00
.41
. 67
1 . 00
. 84
.41
1 . 00
1 . 00
56 . 00
.51
2. 21
8. 28
.75
. 28
1 0 . 06
. 89
. 06
1 . 07
.73
.40
1 . 00
1 .
.
1 .
.
1.
00
86
33
86
45
2 . 73
IV
<JI
IV
APPENDIX 6.4
Kasetsart
SAMPLE
University
PANEL
742
742
1
2
742
742
742
742
5
6
7
742
742
742
8
9
742
742
742
742
742
10
11
12
13
14
15
16
17
742
742
742
742
742
18
19
742
20
742
21
742
22
23
24
742
742
742
742
742
25
26
27
742
28
29
30
742
742
MEAN
SD
SAMPLE
1
2
3
4
243
243
243
243
243
243
5
6
7
8
9
10
243
11
12
13
14
243
243
243
243
243
243
243
243
15
16
17
18
19
20
243
243
21
22
243
243
23
243
243
24
25
26
243
243
27
243
28
243
243
29
30
MEAN
SD
students
SLIP
- . 92
- 1 . 03
- . 27
- . 37
.
-.
-.
-1.
- .
-
01
11
68
60
27
. 22
. 09
. 49
. 53
.
.
.
.
07
09
67
10
DEFORM
1 . 67
1.11
1 . 09
l. 62
1 . 05
1 . 43
. 29
. 73
.99
. 28
.23
. 47
- . 04
- . 69
- . 53
1 . 08
- . 59
- . 04
-1
-1
-
.
.
.
.
.
.
.
45
75
56
21
20
88
84
- . 24
- . 52
SLIP
- . 45
- . 08
- . 13
- .11
- . 01
- . 09
- . 53
- . 09
- .11
. 06
- . 22
- . 02
- . 35
- . 35
- . 03
- . 63
- . 43
. 00
- . 33
- . 97
. 09
- . 12
- . 10
- . 14
- .21
- . 15
- . 20
- . 08
- . 17
- . 20
- . 21
.22
COATING
-1. 21
- . 06
1 . 93
. 17
1. 64
1 . 33
. 14
1 . 83
. 88
1 .42
2 . 12
.61
1. 00
. 61
DEFORM
1 . 41
1 . 14
. 91
. 19
. 03
. 99
.79
1 . 66
. 90
1 . 55
. 35
. 79
1 . 13
. 83
. 19
. 53
. 76
2 . 17
. 72
2 . 08
. 38
.35
2 . 11
1 . 20
. 54
1 . 78
. 94
1. 39
2 . 17
. 68
. 02
. 62
ADJUST
- 1 . 00
. 18
. 03
- .14
.21
- . 26
. 02
- . 17
- . 66
- . 71
- . 05
- . 56
.10
- . 12
- . 28
- . 12
- . 12
- . 07
- . 23
- . 15
- . 07
- . 15
- . 16
- . 04
- . 40
- l . 02
- . 46
- . 67
- . 57
- . 62
- . 66
- . 03
- . 33
. 31
COATING
- . 01
- . 15
- . 36
- 1 . 04
- . 54
- . 07
. 05
- . 12
- . 47
. 00
. 02
. 08
- . 15
- . 76
- . 17
- . 12
- . 67
. 14
. 85
. 99
. 14
. 56
1 . 26
.39
.43
. 00
- . 19
- . 54
- . 18
- . 03
- . 31
- . 18
- . 09
- . 36
- . 19
- . 09
- . 54
.
.
.
.
35
30
64
70
. 18
- . 19
- . 28
- . 19
- . 34
- . 21
- . 38
- . 18
- . 83
- . 39
- . 29
. 22
- . 07
- . 83
. 06
- . 09
- .78
- 1 . 48
- . 47
- . 12
- . 05
- . 07
- . 19
.
.
.
.
08
02
53
08
. 00
- . 02
- . 08
- . 18
. 07
- . 21
- . 97
- . 24
- 1 . 70
- . 57
- . 62
- . 21
- . 21
- . 31
- . 06
.63
- . 36
. 45
ADJUST
SMOOTH
- . 55
- . 17
- . 16
- . 31
SMOOTH
2 . 11
- . 21
. 90
. 02
.78
. 88
. 66
2 . 15
. 53
. 41
PANEL
243
243
243
243
Log of Ideal Ratio Scores of the Prototypes' Sensory A ttributes (Transformed data)
.70
. 19
. 13
. 03
. 01
. 13
. 72
. 04
. 51
- . 28
. 05
- . 02
- . 59
. 00
- . 02
. 26
- . 15
- . 08
- . 37
- . 12
- . 04
- . 08
- . 16
- . 15
- . 09
.42
2 . 06
. 17
. 60
- . 22
. 02
. 49
.
.
.
.
26
10
35
12
. 05
. 05
. 61
.41
.41
. 15
-.
.
- .
-.
07
00
05
46
- . 02
- . 70
- . 08
- . 03
. 00
- . 08
. 10
- . 47
- . 02
- . 10
- . 63
- . 24
- 1 . 05
. 04
- . 73
STICK
. 06
- . 08
. 11
. 29
- . 19
- . 01
. 07
- . 50
- . 01
. 01
. 07
- . 19
. 07
. 03
- . 13
. 03
. 00
. 02
- . 10
. 16
. 08
. 16
- . 05
- . 18
. 18
- . 02
- . 10
- . 12
- . 15
. 05
- . 04
. 14
STICK
- . 01
- . 19
. 15
- . 47
-.16
- . 04
.02
- . 27
- . 02
- . 02
- . 01
- . 06
. 10
- . 05
- . 14
. 01
. 00
. 17
- . 02
. 06
- . 14
. 03
- . 88
- . 40
. 07
- . 33
- . 21
- . 07
- . 28
- . 12
- . 22
- . 24
- . 11
. 21
. 28
PANEL
SAMPLE
557
557
557
1
2
557
557
557
557
557
557
557
557
SLIP
. 00
18
19
20
21
557
.
.
.
.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
181
181
181
181
23
181
24
181
181
181
181
25
26
27
28
181
181
181
29
30
11EAN
SD
- . 01
. 00
. 16
. 20
1 . 04
. 64
.70
- . 70
- . 35
. 00
. 27
. 53
PANEL
181
181
. 00
. 36
- . 35
- . 01
- . 04
. 00
. 72
- .01
. 15
SD
181
181
. 00
. 03
. 02
. 00
. 23
- 1 . 23
. 62
1 . 23
MEAN
181
181
. 00
1 . 32
- . 04
- . 80
. 00
. 00
. 06
. 00
30
181
181
181
181
181
181
181
181
181
181
181
181
00
02
00
00
24
25
28
29
SAMPLE
- . 04
- . 09
- . 02
. 00
- . 01
. 10
26
27
557
. 38
.70
. 37
- . 03
. 02
. 30
22
23
557
557
557
557
- 1 . 83
- . 01
15
16
17
557
557
557
557
. 00
. 03
14
557
557
557
. 04
6
7
557
557
557
557
. 03
- . 04
557
557
. 00
. 02
SLIP
-
COATING
. 48
. 00
3
4
8
9
10
11
12
13
DEFORM
.
.
.
.
.
.
- .
-.
-.
-.
.
.
02
03
09
03
04
07
19
03
03
03
00
00
. 00
. 00
- . 14
. 01
- . 04
. 00
- . 06
- . 13
. 04
. 60
. 48
. 00
. 23
DEFORM
. 00
. 30
. 97
. 44
. 07
. 78
. 53
.70
. 18
1 . 43
. 00
. 18
. 00
. 28
. 28
- . 46
. 20
2 . 02
. 08
1 . 46
. 38
ADJUST
. 00
SMOOTH
. 00
- . 04
- . 06
- . 04
. 25
. 23
- . 04
. 02
- . 01
. 03
- . 03
- . 18
- . 02
- . 05
. 00
- . 15
- . 01
- . 07
- . 08
- . 57
. 00
. 20
. 04
- 1 . 01
. 17
- . 06
- .11
. 02
- . 50
- . 05
. 09
. 00
- . 35
. 00
- . 02
. 09
- . 04
- . 76
- . 06
- . 12
. 10
. 03
- . 51
- . 02
- . 12
. 02
- . 04
- . 90
- . 04
- . 08
- . 08
. 00
- . 06
- . 18
- . 03
. 07
. 00
- . 02
- . 05
. 00
. 00
- . 03
-.
.
.
-.
- .
.
.
01
00
00
04
08
12
13
- . 07
- . 04
- . 14
. 00
- . 06
- . 19
- . 02
- . 13
- . 01
- . 01
- . 01
- .23
. 01
- . 09
. 04
. 00
. 12
- . 14
. 16
- . 27
- . 03
. 10
- . 11
- . 02
. 00
. 12
- . 03
.
.
.
.
.
.
- .
- . 17
. 40
- . 03
. 41
- . 03
. 06
- . 10
.21
COATI NG
ADJUST
SMOOTH
- . 50
. 00
- . 54
.
.
.
.
.
.
.
.
.
.
.
.
.
.
02
04
07
21
02
09
27
05
02
09
10
02
03
02
- . 13
. 01
- . 04
. 00
. 08
. 22
. 69
. 68
. 00
STICK
. 35
- . 21
. 00
. 00
- . 48
- . 24
- . 82
- . 01
. 11
. 13
. 10
. 47
. 13
. 02
. 00
- . 03
- . 03
. 02
- . 06
- . 02
- . 08
. 00
- . 03
- . 01
. 00
- . 02
- . 20
- . 04
. 01
. 00
- . 64
- . 01
- . 08
. 08
- . 67
. 02
- . 10
. 15
. 07
.
.
-.
-.
-.
- . 07
. 00
. 00
- . 05
- . 34
- . 69
. 00
. 00
- . 03
. 00
. 00
. 10
. 27
. 03
. 06
. 14
00
02
12
32
03
.13
32
10
50
33
04
00
11
STICK
.
.
.
- .
.
- .
.
.
- .
06
00
16
14
16
09
06
15
02
- . 04
- . 01
. 00
. 00
- . 05
- . 01
.
.
.
.
.
- .
03
00
09
00
23
06
. 23
- . 27
. 10
- . 04
- . 68
1 . 26
- . 04
l . 63
- . 07
. 00
. 00
. 48
. 51
- . 04
- . 08
- 1 . 53
- . 05
. 00
- . 05
- . 03
- . 06
- . 04
. 45
- . 06
- . 22
- . 05
. 02
. 00
- . 05
- . 08
. 00
. 05
. 43
. 07
. 00
.61
- . 48
. 00
. 11
. 00
. 34
. 00
- . 06
- . 02
. 00
. 00
. 13
. 00
. 15
. 04
- . 05
. 00
. 11
N
(JI
VJ
Kasetsart
SAMPLE
University
PANEL
1
2
3
4
5
6
84 7
84 7
84 7
84 7
84 7
847
84 7
84 7
847
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
847
84 7
847
84 7
84 7
847
84 7
84 7
84 7
847
847
84 7
847
847
23
24
847
25
26
27
28
29
30
847
847
847
847
847
847
MEAN
SO
SAMPLE
PANEL
1
2
3
4
5
6
428
428
428
428
428
428
428
428
428
428
428
428
428
428
428
428
428
426
428
428
428
428
428
428
428
428
428
428
426
428
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
MEAN
SO
s t udents
SLIP
DEFORM
- . 11
- . 37
- . 36
- . 69
- . 02
-.11
- . 25
- . 15
- . 08
- . 45
. 14
- . 09
- . 44
- . 07
- . 08
- . 03
- .10
- . 02
- . 19
- . 59
- . 16
- . 17
1 . 16
. 95
. 66
. 99
. 06
1 . 06
. 83
1 . 19
. 73
1 . 23
.61
. 49
. 99
.71
. 50
- . 14
.41
2 . 15
. 62
1 . 93
1 . 20
. 24
- . 17
- . 10
- . 20
- . 36
- . 56
- . 66
- . 30
- . 27
. 26
-1 . 2 6
SLIP
COATING
ADJUST
SMOOTH
STICK
1 . 06
. 00
1 . 54
. 83
1. 34
2 . 06
. 73
- . 18
- . 06
- . 32
- . 06
- . 01
- . 20
- . 34
- . 12
- . 03
- . 42
- . 44
- . 06
- .28
. 02
. 12
- . 04
- . 14
- . 34
- . 41
. J5
.13
- . 14
- . 04
- . 26
- . 05
-.17
- . 08
- . 41
- . 27
- . 12
.70
- . 37
- . 44
- 1 . 19
- . 01
- . 02
- . 62
- . 04
- . 47
- . 17
- . 03
- . 04
- . 81
.00
. 01
- . 15
- . 15
- . 04
. 04
- . 12
- . 44
- . 19
- . 26
- . 25
- . 25
. 38
1 . 64
. 31
. 17
- . 13
. 43
- . 04
- . 08
- . 60
- . 01
- . 06
- . 54
- . 27
- . 32
- . 19
- . 07
- . 05
- . 35
- . 08
- . 02
- . 01
- . 08
- . 05
- . 06
- . 08
. 06
. 04
- . 75
- . 20
- . 10
- . 04
- . 69
- . 42
- . 09
- . J5
. 03
. 00
. 17
- . 04
. 18
. 02
. 07
- . 68
. 02
- . 28
- . 06
. 02
. 07
.
- 03
- . 12
- . 04
. 00
. 09
. J2
. 23
. 13
.21
- . 61
- . 22
. 03
. 04
. 00
- . 16
. 09
. 00
. 94
. 60
- . 17
. 15
- . 09
. 51
- . 19
. 22
. 04
. 20
ADJUST
SMOOTH
2 . 15
DEFORM
COATING
- . 93
- . 61
- . 35
- . 28
- . 45
- 2 . J7
-.13
- . 12
- . 39
- . 03
- . 12
- . 31
- . 46
- . 02
- 1 . 08
- . 03
- . 04
- . 07
- . J2
- . 10
- . 09
- . 29
- . 15
- . 01
- . 10
. 07
- . 09
. 04
- . 75
- . 40
- . 32
. 04
- . 29
- . 35
- . 15
2 . 14
. 12
. 20
- . 25
- . OJ
- . 03
- . 16
- . 32
- . 47
- . 02
. 00
- . 03
- . 09
. 00
- . 01
. 12
- . 15
. 08
.02
. 12
. 05
. 03
- . 38
- . 72
- . 19
. 46
1 . 26
. 46
. 07
- . 68
- 1 . 54
- . 05
- . 20
- . 25
. 02
- . 06
- . 19
- .61
- . 41
- . 26
-.19
- . 03
- . 27
. 00
- . 02
- . 05
- . 08
. 00
. 00
- . 08
. 03
. 10
- . 12
- . 69
- . 32
- . 28
- . 28
- . 05
- . 26
- . 59
. 02
- . 15
. 13
- . 34
- . JO
- . 02
. 03
. 15
- . 01
. 07
. 07
. 02
. 10
- . 05
. 03
. 05
. 00
. 02
. 04
. 06
.13
.12
. 04
. 78
. 21
- . 04
- . 46
- . 21
- . 24
. 08
- . 35
. 35
. 86
. 55
- . 27
. 43
. 04
. 53
- . 22
. 32
- . 05
.20
- . 77
PANEL
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
314
314
314
314
314
314
314
314
314
314
314
314
314
314
314
314
314
314
314
314
314
314
314
314
314
314
314
314
314
314
MEAN
SO
SLIP
COATING
ADJUST
SMOOTH
. 00
. 21
- . 25
- . 37
- . 46
- . 33
- . 32
- 1 . 04
- . 54
- . 09
- . 02
- . 44
- . 21
. 06
. 05
. 14
- . 59
-2 . 12
- 1 . 00
- . 05
- . 20
- . 28
- . 09
-1 . 66
- . 05
. 64
2.18
. 60
. 64
- . 98
-2.15
1 . 13
- . 66
1 . 18
.91
1 . 69
1. 37
1 . 68
. 48
. 40
1 . 33
. 80
1 . 06
. 49
. 83
2 . 18
.74
2 . 18
1. 43
. 46
2 . 18
1. 3 9
. 51
1 . 88
. 88
1. 38
2 . 17
. 80
- . 24
- . 08
- . 76
-.71
- . 27
- . 19
- . 76
- . 95
- . 34
- . 74
- . 06
- . 14
-2 . 11
- . 96
- . 28
- . 62
- . 82
- 1 . 16
- . 47
- 1 . 48
- . 62
- 1 . 27
- . 55
- 1 . 66
- . 60
- 1 . 02
- 2 . 03
-2 . 14
-2 . 18
- 1 . 26
1 . 16
. 65
- . 88
. 64
- .22
. 75
DEFORM
- 1 . 85
-J . 0 3
-1 . 44
- . 92
- . 82
- . 36
- 1 . 19
- 1 . 60
-1 . 21
- . 53
- . 34
- . 96
-2 . 1 1
- 1 . 00
- . 67
- . 93
- . 72
- 1 . 06
- 1 . 22
-2 . 1 8
-1. 2 2
- 1 . 27
-2 . 1 6
- 1 . 70
- 2 . 08
-2.11
- . 15
-1. 99
-2 . 1 4
-1 . 55
1 . 69
1 . 15
-1 . 26
. 60
1 . 13
- .13
STICK
. 03
- . 70
. 07
- . 81
- . 61
- . 16
- .10
- . 13
- 1 . 26
- 1 . 00
- . 47
- . 45
- . 07
- . 09
- . 01
- . 22
- . 25
- 1 . 00
- 1 . 03
-2 . 18
- . 06
- . 08
- . 30
- . 89
- 1 . 85
- 1 . 79
- . 99
- 2 . 18
- . 13
. 07
. 00
- 1 . 29
.22
. 45
. 96
. 99
. 98
. 06
. 03
. 00
-.21
- . 16
- . 03
. 00
. 17
- . 08
. 02
- . 56
- . 03
- 1 . 35
- . 12
- . 32
- 1 . 84
- . 99
-. 24
- 1 . 87
- . 30
- . 69
. 73
- . 45
. 55
- . 47
STICK
1. 69
1 . 03
1 . 04
. 58
. 09
. 63
.76
. 50
. J8
1 . 49
. 35
. 23
. 33
.71
.92
. 05
. 66
. 78
. 44
1 . 46
1 . 31
. 10
1 . 73
1 . 36
.33
1 . 72
. 56
1 . 27
1 . 8J
.73
- 1 . 45
- . 20
- . 19
- . 21
- . 06
- . 05
- . 23
- . 97
- . 05
- . 39
- . 15
- . 06
- . 09
- . 07
- . 07
- . 15
- . 27
- . 01
- . 54
- . 13
- . 53
. 04
- . 68
- . 50
SAMPLE
N
CJI
Of f i ce workers
PANEL
SAMPLE
742
742
1
2
3
742
742
4
5
742
6
7
742
742
742
742
742
742
742
742
SLIP
- . 53
- . 83
1 . 14
1 .76
- . 32
- . 11
- . 60
- . 01
- . 15
- . 75
- . 01
- . 38
1 . 20
- . 11
.01
- . 76
1 . 65
- . 55
. 65
. 82
- . 18
. 08
- . 32
- . 16
- 24
- . 30
- . 38
- . 57
- . 33
- . 59
- . 34
- 1 . 00
- 1 . 17
- . 34
53
81
60
. 72
. 11
.21
- . 07
39
. 96
.01
- . 39
- . 77
- . 33
- . 16
- . 32
- 29
-1. 25
. 30
- .11
- . 78
-.
-1.
- .
- .
-. 76
742
15
16
- . 60
742
17
742
742
742
18
19
- 1 . 13
- . 28
22
742
742
23
24
25
742
742
742
742
742
26
27
28
29
30
742
742
MEAN
SO
PANEL
SAMPLE
243
243
243
243
243
243
243
243
243
2
3
4
5
6
7
8
243
243
243
243
243
- 1 . 26
- . 27
- . 74
- . 13
- . 32
. 62
- . 38
- . 33
. 00
12
13
14
15
. 85
. 38
1 . 08
1 . 76
1 . 28
2 . 10
.00
.74
. 72
1 . 03
82
26
46
82
18
. 59
. 51
. 52
2 . 06
.20
- 1 . 21
- . 16
1. 25
. 52
-1. 55
- . 93
. 19
1 . 44
.
.
.
.
.
- 1 . 24
21
22
23
24
- . 05
- . 12
- . 84
- . 77
1 . 06
. 19
- . 02
. 10
243
25
- 1 .86
1 . 17
1 . 68
243
243
26
27
243
243
28
. 05
- 1 . 86
- . 11
1 . 00
1 . 15
. 26
243
30
- . 91
- . 50
- . 59
. 53
. 01
- . 25
- . 14
.26
. 00
- 1 . 19
- . 29
- . 39
. 06
- . 08
. 81
. 68
DEFORM
- . 13
. 51
- . 28
- . 12
- . 26
- . 70
. 58
SLIP
. 07
- 1 . 56
. 34
- 2 . 16
- . 11
- . 91
- . 66
20
SO
- . 19
- . 06
- . 01
- . 02
- . 29
- .. 3 4
. 54
. 12
. 80
- . 29
- . 72
- . 32
1 . 27
1 . 68
. 79
. 00
.19
2 . 01
1 . 64
18
19
MEAN
- . 63
. 18
- . 02
- . 47
- . 55
. 10
20
11
15
80
11
29
. 04
- . 42
- . 38
.
.
.
.
.
243
243
243
243
. 09
. 39
. 75
- 1 . 28
. 05
- . 39
. 06
-2
-
243
243
243
. 46
1 . 15
. 48
- . 03
STICK
- . 16
- . 17
- . 26
- . 02
- . 68
9
10
16
17
. 56
- . 29
- 1 . 13
- . 33
, 10
> , }O
- . 07
- . 23
- . 40
11
243
243
243
- . 54
- . 39
. 13
- 1 . 00
20
SMOOTH
- . 22
- .21
9
10
11
12
13
21
ADJUST
. 56
. 38
742
7 42
COATING
- 1 . 22
- . 15
14
742
PANEL
SAMPLE
DEFORM
- .
27
- . 55
. 05
- . 48
- . 08
10
69
45
60
25
- . 14
- 1 . 38
- . 04
- 1 . 38
- . 19
. 77
- . 21
.
.
- .
.
- . 22
- 1 . 23
- . 99
- . 59
- . 27
.
- .
- .
.
-.
- . 33
. 38
- . 41
. 39
- . 43
. 39
ADJUST
SMOOTH
- . 11
-1
-
.
.
.
.
.
COATING
- . 90
-.21
- . 16
- .11
- . 11
- . 01
- . 07
- . 24
- . 12
- . 01
- . 46
. 17
. 84
. 66
- . 22
- . 12
- . 34
- . 18
. 35
- . 98
- 1 . 06
- . 16
- . 22
.01
. 24
- . 24
. 44
- . 21
- . 50
- . 05
. 39
- . 14
- . 52
- . 60
- . 32
- . 46
- . 05
- . 13
. 00
- . 14
.
.
.
.
.
.
.
- .
67
15
41
01
03
64
06
02
- . 87
- . 38
-
. 06
. 15
. 37
. 09
- .
.
.
- .
22
58
00
50
. 07
- . 53
- . 07
-.07
- . 09
- . 27
- . 27
- . 02
- . 24
- . 48
- . 36
. 16
- . 27
. 14
- . 11
- . 08
01
12
11
00
03
64
06
17
21
- . 14
. 37
STICK
.74
. 31
. 00
. 02
. 06
. 02
. 12
. 20
. 15
. 05
. 04
. 13
- . 53
- . 19
. 02
. 63
. 58
- .61
- . 23
557
557
557
557
557
557
. 03
- . 08
. 00
. 65
. 27
. 31
. 30
17
. 05
. 00
557
557
557
557
557
18
. 05
19
20
. 00
- . 45
. 00
. 08
21
22
23
24
25
26
27
28
29
30
557
557
MEAN
PANEL
1
2
3
4
5
6
7
181
181
181
181
181
181
181
181
181
181
181
8
9
10
11
181
181
181
181
181
181
181
181
181
181
00
- . 20
. 00
. 25
- . 04
- . 04
- . 03
. 00
- . 01
. 13
SO
SAMPLE
SLI P
- . 18
. 00
- . 08
- . 14
- . 38
- . 22
. 13
.15
- . 20
- . 01
12
13
14
15
16
17
18
19
20
.
.
.
.
.
.
.
02
02
16
45
13
00
53
. 00
- . 11
. 00
- . 05
. 07
. 48
21
22
23
- . 03
- . 05
181
- . 08
. 08
- . 06
181
24
25
26
- . 03
181
. 00
. 08
- . 02
181
181
28
181
18 1
29
30
- . 60
- . 74
- . 44
- . 88
. 08
2 . 17
1 . 91
- . 20
- . 60
- . 53
. 20
- . 83
- . 33
- . 51
. 05
. 93
- . 33
- . 24
- . 12
. 66
. 28
. 32
- . 29
. 28
- . 06
. 11
27
MEAN
SO
. 27
. 23
.15
.00
- . 15
- . 06
- . OB
. 17
- . 14
- . 01
- . 53
- . 04
. 04
- . 03
.01
- . 01
- . 05
. 00
- . 01
- . 14
. 45
.75
- . 03
- . 01
- . 36
- . 09
. 44
. 38
1 . 53
05
11
17
- . 90
- . 49
- .
.
.
- .
. 35
-. 36
- . 53
. 19
17
- . 69
.23
. 27
. 60
. 00
. 89
. 00
. 09
- . 52
. 02
. 07
- . 13
. 00
'. 0 0
- . 16
. 00
. 04
- . 04
- . 04
- . 43
- . 04
15
16
. 00
- .71
SMOOTH
08
00
02
04
- . 06
. 08
557
557
. 02
- . 07
ADJUST
.
.
.
- .
12
13
14
COATING
- . 07
. 00
- . 38
. 03
- . 14
- . 61
- .11
. 39
. 00
.74
11
557
557
557
. 48
. 09
557
557
557
557
557
557
557
- . 32
. 03
. 51
. 00
. 02
. 06
- . 07
- . 21
8
9
10
557
DEFORM
02
- . 04
. 01
181
181
- . 18
- . 03
- . 76
- . 03
- . 74
- . 17
1
2
3
4
5
6
7
557
557
557
557
SLIP
. 28
00
- . 43
. 00
- 1 . 00
2 . 12
. 10
1 . 15
1 . 28
. 30
. 67
DEFORM
. 51
- . 11
.93
1 . 32
1 . 20
1 . 48
.13
- . 25
. 15
. 32
. 33
. 37
. 32
1 . 85
- . 90
. 00
.
.
.
.
00
00
00
00
18
30
01
18
06
11
39
.
- .
.
- .
.
09
05
00
05
00
- . 07
. 10
1 . 22
. 00
. 26
- . 45
- . 30
1 . 89
. 85
. 46
. 74
. 06
.19
- .
.
.
- .
03
00
00
19
SMOOTH
09
10
21
14
. 00
- . 01
- . 04
. 00
- . 18
-.12
- . 10
. 14
. 04
- . 12
. 00
- . 15
ADJUST
.
.
.
.
.
- . 05
- . 45
. 05
- . 20
. 24
26
03
10
11
11
28
28
03
03
04
- . 09
- . 37
- . 10
.
.
.
.
.
.
.
.
.
.
. 00
- . 09
. 00
- . 08
. 24
.-
-.27
- . 07
- . 20
- . 47
- . 79
COATING
-1 . 3 1
- . 08
- . 03
-.95
- . 22
. 04
.20
. 00
. 19
- . 65
. 23
2 . 15
. 00
- . 20
- . 11
- . 07
-.21
- . 02
. 19
- . 92
. 14
.
.
.
.
.
.
. 00
- . 03
- . 11
- . 10
STICK
.
.
.
.
.
.
.
.
28
02
14
01
24
24
69
02
- . 07
. 03
. 02
- . 07
- . 19
- . 42
- . 32
. 00
- . 20
. 00
- . 03
- . 99
. 13
. 06
- . 04
- . 29
- . 04
. 00
. 20
. 10
-1.29
- . 11
- . 14
-.21
. 02
. 00
. 03
- . 84
.02
-. 1 1
.45
-.22
.38
STICK
06
00
21
01
03
04
13
03
04
01
-.41
- . 02
. 00
- . 02
- . 06
. 00
- . 05
. 02
. 09
. 04
- . 04
- . 10
- . 14
. 13
- . 18
-.
.
- .
- .
- .
- .
.
- .
- .
- .
- . 08
- .
.
- .
- .
- .
22
00
39
14
11
- 1 . 01
- . 07
- . 01
.09
.10
-.23
- . 03
.43
. 00
. 00
. 00
-.11
- . 01
- . 39
. 23
. 02
- . 32
. 00
- . 04
. 06
- . 12
. 00
. 00
- . 32
- . 04
- . 05
- . 04
. 17
. 10
- . 09
.26
- . 09
- . 01
.21
. 16
- . 01
. 00
. 03
- . 02
. 00
.02
.00
IV
<.n
<.n
OFFICE WORJ(ERS
SAMPLE
PANEL
SLI P
847
847
847
847
847
847
847
847
847
8'47
847
847
847
847
847
847
847
847
847
847
847
847
847
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
847
- , 1
- . 03
847
847
847
847
847
847
26
27
28
29
30
SO
PANEL
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
428
428
428
428
428
428
428
428
428
428
428
428
428
428
428
428
428
428
428
428
428
428
428
428
428
428
428
428
428
428
MEAN
SO
.
.
.
.
.
13
84
40
65
12
- . 30
. 25
MEAN
SAMPLE
. 55
. 04
. 15
. 27
. 18
. 22
. 00
- . 49
- . 16
- . 30
- . 04
- . 63
- . 78
- . 38
- . 08
- . 48
- . 27
- . 72
- . 30
- . 17
- . 30
- . 42
- . 07
SLI P
DEFORM
. 47
. 15
1 . 00
1 . 32
. 96
1 . 76
. 00
. 55
. 26
.70
. 54
1 . 13
.61
1 . 92
.20
1 . 11
. 34
-. 13
1 . 31
. 97
. 64
. 05
. 18
, 46
. 00
2
2
1
COATING
. 17
. 07
. 03
.11
.11
. 33
. 00
- .41
. 03
- . 07
. 09
- . 45
- . 15
- . 20
- . 28
. 00
. 00
11
. 12
. 09
.15
. 03
- . 24
- .
23
. 00
. 93
. 02
. 13
. 01
. 51
. 76
. 65
- . 14
. 12
DEFORM
.
.
.
.
.
10
08
12
30
05
COATING
ADJUST
. 33
. 08
. 04
. 01
. 14
. 31
. 00
- . 50
- . 38
- . 08
- . 20
- . 36
- . 26
- . 10
- . 17
-.11
- . 11
- . 08
. 89
. 00
. 13
- . 07
. 12
- . 08
- . 30
- . 32
- . 07
- . 05
. 16
-.11
SMOOTH
-
. 38
. 07
. 06
. 01
. 03
. 13
. 00
- . 21
- . 27
- . 08
- . 49
- . 39
46
. 09
. 11
- . 10
. 22
. 07
- . 75
. 00
. 66
- . 11
. 09
- . 07
. 00
. 13
. 08
. 66
. 13
11
- . 16
.21
- . 19
ADJUST
SMOOTH
. 22
STICK
. 17
- . 02
. 00
- . 02
- . 06
- . 02
. 00
- . 29
. 34
. 02
- . 38
. 04
- . 12
. 10
- . 08
. 00
.51
- . 07
- . 09
- . 28
. 00
- . 51
- . 04
. 06
. 00
. 05
- . 08
- . 03
. 22
- . 30
STICK
. 83
. 38
1 . 14
1 . 71
. 96
1 . 48
- . 55
. 19
. 35
. 67
.21
. 85
. 26
1 . 89
.46
1 . 19
. 24
- . 28
1 . 22
. 52
.68
. 02
. 27
1 . 25
. 00
.26
2 . 08
- . 32
2 . 10
1 . 75
- . 76
-.21
- . 23
- . 11
. 11
- . 17
- . 16
- . 12
. 06
. 02
- . 31
- . 35
. 45
- . 05
- . 73
. 06
- . 13
- . 07
.23
- . 30
- . 10
. 04
.22
- . 09
. 00
- . 08
- . 12
- . 31
- . 30
. 00
. 86
. 16
. 27
. 01
. 14
. 39
.69
- . 13
- . 45
- . 06
- . 12
- . 29
- . 68
- . 31
- . 02
- 1 . 04
- . 29
- . 03
- . 64
- . 52
- . 68
. 02
. 00
- . 19
- . 04
- . 23
- . 14
- . 12
. 09
- . 46
. 85
. 15
- . 29
. 01
- . 03
. 32
. 10
- . 13
- . 65
- . 05
- . 36
- . 49
- . 09
- . 09
- . 02
- 1 . 67
- . 08
- . 24
- . 77
- . 46
- . 07
- . 07
. 00
- . 19
. 00
- . 01
- . 14
-2 . 14
. 59
- .73
- . 57
- . 02
. 24
- . 02
- . 06
- . 02
- . 17
- . 01
. 25
. 02
- . 05
- . 05
- . 36
. 06
- . 02
- . 08
. 65
- . 26
. 08
- . 63
. 13
- . 09
. 00
. 04
. 00
. 06
- . 04
. 01
. 22
. 00
. 38
. 40
.73
.72
- . 18
.21
- . 25
. 33
. 35
. 50
. 02
.23
s t uden t s
PANEL
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
742
742
742
742
742
742
742
742
742
742
742
742
742
742
742
742
742
742
742
742
742
742
742
742
742
742
742
742
742
742
- . 03
.20
. 64
. 12
. 32
. 27
. 18
. 14
. 11
. 05
- . 29
- . 20
- . 08
- . 22
- . 09
- . 43
- . 39
- . 86
- . 27
- . 40
- . 66
- . 06
- . 30
- . 04
- . 17
- . 14
- 1 . 02
. 20
- 1 . 12
- . 61
- . 48
- . 89
-1
-
School
SAMPLE
MEAN
SO
SAMPLE
PANEL
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
243
243
243
243
243
243
243
243
243
243
243
243
243
243
243
243
243
243
243
243
243
243
243
243
243
243
243
243
243
243
MElIN
SO
SLI P
DEFORM
- . 56
- . 65
- . 91
- 1 . 57
- . 61
-1 . 16
- . 56
- . 35
-2 . 17
- . 36
- 1 . 16
- . 44
- .48
- . 59
- . 42
- 2 . 08
- . 65
- 1 . 26
- . 57
- . 51
. 02
- . 43
- 1 . 46
- . 89
- 2 . 16
- 1 . 32
- . 27
- . 31
-2 . 17
- .70
1. 01
1 . 07
. 96
1 . 86
1. 99
1.01
.91
. 03
. 09
.61
.53
- . 37
-.18
.91
. 42
. 82
. 92
.00
1 . 22
2 . 00
. 30
1 . 44
1. 31
.12
2 . 02
1 . 64
. 99
1. 47
2 . 16
.79
- . 89
. 62
1 . 10
.71
SLIP
DEFORM
COATING
ADJUST
SMOOTH
-2
- . 30
. 00
- . 60
- . 13
- . 73
- 1 . 52
- . 24
- . 34
- 1 . 57
- . 20
- . 55
. 00
- . 39
- . ]7
- . 22
- . 20
- . 61
- . 80
. 53
. 06
. 81
. 62
. 83
- . 64
- . 16
- . 11
- . 44
- . 18
- . 18
- . 21
- . 08
- . 65
- . 87
-.11
- . 60
- . 20
- . 73
- . 12
- . 49
- 1 . 56
. 00
- . 05
- . 51
- . 29
- . 60
- . 32
- . 03
- . 09
. 00
. 00
- . 08
-.21
- . 03
- 1 . 18
- . 29
- . 16
- . 27
- . 26
- . 17
.20
. 30
- . 84
- . 17
- . 05
- . 02
- , 01
- . 06
- . 08
. 27
- . 18
- . 08
- . 14
. 08
. 01
- . 54
. 28
- . 31
. 09
- . 11
- .11
- . 12
- . 57
- . 39
. 54
- . 34
. 37
- . 10
. 27
ADJUST
. 31
. 29
. 62
. 15
. 43
. 09
. 18
- . 05
- . 31
- . 15
- . 96
- . 0]
. 02
- . 46
- . 35
. 03
- . 06
- . 95
- . 38
- . 13
. 02
- . 14
- . 68
. 16
- . 83
- . 38
-.21
- . 49
- . 34
- . 16
- . 35
. 44
SMOOTH
STICK
. 14
. 56
. 57
. 07
.75
. 06
. 37
- . 68
- . 17
- . 93
- 1 . 47
- . 27
- . 06
- . 10
-.2
- . 17
- . 03
. 00
. 00
- . 38
- . 09
- . 20
. 02
- . 50
- . 97
- . 04
- . 10
- . 02
- . 05
- . 18
- . 31
- . 48
- . 11
- . 02
-2 . 15
- . 03
- . 17
- . 18
. 04
- . 41
- . 06
- . 10
- 1 . 89
- . 04
- .71
- 1 . 03
.11
- . 83
- . 64
- . 08
. 03
- . 20
. 02
. 28
- . 04
. 16
- . 40
- . 53
- . 05
- . 10
. 05
- . 72
- 1 . 64
- . 29
- . 09
1 . 18
- . 56
- . 42
. 46
- . 18
. 48
- . 33
. 49
- . 29
. 57
- . 37
- . 64
- . 90
- . 21
- . 47
- . 27
- . 17
- . 10
- 1 . 47
- . 11
- . 05
- . 09
- . 08
-2 . 13
- . 31
- . 35
- . 92
- . 39
- . 29
- . 37
- . 06
- . 44
- . 24
- . 67
- . 08
- . 18
- . 21
- . 03
- . 02
- . 20
. 30
. 10
. 61
- . 69
.H
- . 48
. 22
- . 56
- . 05
. 00
- . 05
. 16
. 13
. 92
. 09
. 95
. 64
. 02
- . 81
- . 79
- . 13
- . 64
. 10
- . 06
. 18
- . 16
. 02
. 27
- . 55
- . 19
- . 13
. 02
- . 45
1 . 55
. 10
- . 28
- . 02
- . 11
- . 72
. 00
.11
COATING
1 . 09
1 . 02
1 . 01
1 . 87
2 . 04
1 . 06
. 50
1 . 97
2 . 17
. 90
1 . 43
. 23
. 18
1 . 22
. 35
. 58
. 58
.23
. 77
. 83
. 55
.81
1 . 25
. 11
2 . 12
1 . 19
.79
1. 68
. 60
. 75
- . 09
- . 51
- . 38
- . 06
- . 12
- . 41
- . 04
- . 27
- 1 . 21
- . 04
- . 05
. 23
- . 05
- . 07
- . 33
. 03
. 34
- . 36
- . 12
- . 35
. 09
- . 10
- . 59
STICK
- . 09
-
N
U1
0'\
School st udents
PANEL
SAMPLE
1
2
557
557
5 57
3
4
5
5 57
557
5 57
6
7
557
557
557
557
557
557
10
11
12
557
557
557
13
14
15
16
17
557
557
18
19
557
557
20
21
22
23
5 57
5 57
557
5 57
24
25
5 57
7
557
557
557
557
557
181
8
9
181
181
10
11
12
13
14
15
181
181
181
181
181
181
181
16
17
181
18
19
181
20
21
22
181
181
181
181
23
24
25
181
181
181
26
27
181
28
181
181
29
30
181
MEAN
SO
SH<XlTH
- . 01
- . 38
- . 00
- . 11
STICK
. 16
. 48
. 03
. 00
. 00
.06
. 00
. 00
1 . 20
. 00
- . 60
- . 66
- . 87
. 00
- . 10
- . 26
. 03
. 22
- . 06
- . 16
- . 06
- . 69
- . 12
- . 06
- . 09
- . 21
- . 38
- 1 . 70
- . 54
- . 10
- . 51
- . 34
- . 06
- . 15
.13
.70
- . 17
- . 03
. 01
2 . 14
- . 08
. 30
1 . 39
- . 58
- . 89
- . 75
- . 11
. 00
.
.
- .
- .
02
03
10
09
. 38
- . 88
- . 18
. 00
.11
- . 06
- . 85
, )5
- , 0)
- , 11
. 47
2
3
4
5
6
7.
ADJUST
. 00
- . 09
. 28
- , 09
-2 . 17
, 11
181
181
181
181
181
181
181
- . 18
- . 09
28
29
PANEL
COATING
. 07
. 43
- , 02
. 00
MEAN
SO
DEFORM
- . 01
- . 05
26
27
30
SAMPLE
SLIP
SLIP
- ,
.
.
- .
- .
,
-.
03
02
00
13
18
25
07
20
13
01
12
00
,
,
.
.
.
- , 31
- . 01
- . 01
- , 27
, 28
, 00
- , 03
- , 09
, 06
- ,22
- 1 . 08
,03
- . 06
- ,71
, 00
. 00
- , 08
, 04
- , 10
. 25
. 56
.
.
.
- .
56
22
00
35
. 82
. 40
1 . 31
:78
. 07
1 . 77
1 . 03
-,6)
1 . 20
.43
,
.
2 .
.
18
95
18
00
. 56
.74
DEFORM
. 30
23
38
81
85
00
15
64
00
- .
.
1 .
.
.
.
1 .
1.
. 85
1 . 35
,
.
1 .
- ,
- ,
,
00
36
11
18
70
27
1 . 10
, 90
1 . 48
,11
. 90
. 95
- , 06
1 . 41
1. 44
, 00
, 00
, 00
.
.
.
.
.
36
64
57
11
50
- . 26
- . 61
- . 45
- . 37
- .23
- . 20
- . 03
- , 09
- . 01
. 02
- . 02
- . 20
- 1 . 03
- . 51
- . 28
. 29
COATING
-
. 10
. 02
. 06
. 10
. 03
- 1 . 04
.11
- . 14
- , 05
- . 07
15
. 00
- . 25
- . 70
- . 02
- . 76
.13
- . 39
- . 0]
. 03
- . 63
. 02
- 1 . 30
- . 42
- . 08
- . 28
- .81
1 . 47
- , OB
- , 02
- . 06
, 03
. 22
. 16
. 00
05
26
07
07
- . 15
- . 08
- . 27
- . 05
- . 26
- . 16
- . 03
, 07
- . 02
-.
.
.
.
-.
- .
- .
-.
- .
- .
02
06
00
00
77
51
02
08
12
14
- . 51
- . 06
- . 06
- . 14
- , 03
- , 13
- , 08
- , 96
- , 30
1 .66
- , 50
- , 07
- . 24
- . 53
- . 15
- 1 . 08
. 03
. 44
. 05
. 40
. 62
. 03
- 1 . 06
- , II
11
- . 03
. 00
- . 86
- . 09
- , 07
- , 71
. 00
- , 26
1 . 43
- . 77
SH<XlTH
- , 03
, 35
- ,01
, 09
- ,73
.
.
.
.
ADJUST
- .35
- . 04
, 10
, 13
, 17
-,11
. 08
- 1 . 17
- . 06
. 00
- . 20
.47
- . 06
- .24
. 15
-1
. 01
. 05
, 11
. 09
, 08
. 14
. 04
. 06
. 04
, 05
, 20
. 00
-.38
- . 61
- . 04
- , 02
- . 26
- , 19
- , 14
, 00
, 04
- , 05
- 1 . 21
- . 39
. 59
. 19
- . 19
, 00
. 00
- . 41
.41
. 01
. 04
. 04
. 02
- , 13
. 00
. 00
. 04
. 26
- . 29
- , 03
- , 07
. 00
. 00
- , 25
. 00
- , 07
- . 54
, 12
- . 06
, 16
, 00
.00
, 00
, 00
- , 02
, 00
- , 39
- , 18
, 00
- , 08
, 00
. 00
, 39
- , 29
- , 17
- . 19
, 01
- , 04
,26
, 00
- . 16
.25
, 46
,31
students
PANEL
1 . 75
, 36
847
847
847
847
847
847
847
847
847
847
- . 20
_ . 26
9
10
- . 32
1 . 48
- . 96
. 10
- .01
- . 11
- . 14
- . 49
- . 06
. 09
1 . 07
1
2
3
4
5
6
7
428
428
428
428
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
428
428
428
428
428
24
25
428
26
27
428
428
428
28
29
428
428
30
MEAN
SO
- . 70
. 05
, 09
428
428
428
428
428
428
428
428
428
428
- . 17
24
25
PANEL
428
428
428
428
428
- . 16
- . 29
. 00
- . 51
- . 06
- , 91
MEAN
SO
SAMPLE
- . 15
- . 02
18
19
20
21
22
28
29
]0
847
847
847
- . 33
-. 24
- . 06
- . 12
26
27
847
847
- . 02
- . 24
- . 29
- . 04
. 64
1 . 82
23
847
- . 28
- . 04
- . 21
- . 18
15
16
17
847
847
. 58
SMOOTH
- . 24
- . 04
14
847
847
847
847
- .21
- . 16
- . 10
- , 31
- . 21
. 47
1 . 48
.86
. 11
. 00
. 30
. 86
.74
. 74
1 . 29
1 . 93
. 14
1. 83
1 . 35
- . 10
1 . 32
. 99
. 79
- , 49
- . 18
- . 15
- . 35
- ,22
1 . 28
1 . 83
. 61
- . 24
, 26
. 94
. 56
SLIP
-
. 97
, 37
. 58
. 92
. 76
, 21
. 13
- . 20
- . 07
- , 07
- 1 . 47
, 27
- , 11
, 01
- . 26
- . 43
, 15
- , 62
- . 25
- . 39
- , 18
- , 61
- , 03
. 09
DEFORM
1 . 06
. 98
. 92
1 . 84
1 . 83
. 19
. 72
1 . 64
1 . 28
1 . 35
.70
.27
. 09
. 51
. 19
. 24
, 68
1 . 17
1 . 09
1 . 83
, 45
, 90
. 40
- , 04
- , 09
1 . 76
1 . 30
, 59
- , 34
- . 22
- . 12
1 . 17
, 00
.95
- , 35
, 49
- , 31
, 37
. 89
, 56
- . 40
- . 01
- . 71
- . 20
- . 19
- . 09
- . 12
- . 20
- . 15
- . 22
- . 11
- . 28
- . 07
- . 03
- , 69
- . 10
- , 12
- , 35
STICK
ADJUST
- . 05
11
12
13
847
847
- . 14
. 88
1. 45
6
7
847
COATI NG
. 90
- . 05
847
847
DEFORM
- . 39
- . 41
3
4
847
SLIP
. 58
847
847
STICK
, 03
- . 14
- , 40
. 58
, 66
- . 03
. 05
-1.86
- . 45
School
SAMPLE
- . 16
- . 69
- . 27
- . 14
- . 46
- . 03
- . 17
- . 16
- . 03
- . 39
- . 44
- . 04
- . 18
- . 47
. 18
- . 11
- . 42
- . 25
- . 28
- . 78
- . 09
- . 05
- . 11
- . 53
- . 53
- 1 . 04
. 00
- . 05
- .10
. 04
- . 20
, 27
- , 09
- . 17
, )7
- .02
. 22
. 05
- . 25
-.71
. 25
- . 07
- . 04
- . 04
- .44
- . 02
. 08
. 22
- 1 . 10
. 88
. 43
. 22
. 81
. 00
. 18
- .06
. 05
- .
- .
- .
- .
-.
01
11
31
01
03
- . 76
- , 05
- . 39
.
,
.
.
.
09
24
20
99
11
- . 21
- . 25
. 18
- , 38
- . 13
, 00
1
-
- , 20
. 18
- .26
, 28
- . 23
.30
- , 04
. 45
ADJUST
SMOOTH
- . 39
- . 31
- . 28
- , 4]
- . 20
- . 19
- . 01
- , 1 5'
- , 06
COATING
-
. 01
. 19
. 47
. 07
. 35
. 59
. 06
- , 14
- , 13
- . 18
- , 74
. 02
- , 46
- . 36
- , 26
- . 82
-,
,
- ,
-,
31
00
20
32
- , 03
- , 03
- . 50
, 03
- . 05
-, 23
- . 07
- . 27
, 00
, 07
, 22
,23
. 07
. 12
. 77
- , 48
-
.
.
.
.
42
06
85
01
.
.
.
.
11
12
08
22
- , 94
. 05
- . 06
- , 23
- , 05
-. 04
. 00
- , 09
- , 30
-.43
- , 13
- . 30
1 . 51
- , 01
- , 05
-1. 00
, 00
- , 02
- . 08
- . 22
-1 . 16
, 00
- . 38
- , 43
- , 14
- , 18
- , 32
- . 26
- , 17
-
, 05
, 38
, 36
. 13
, 18
- . 12
- , 87
- . 04
STICK
.15
. 24
- . 52
- . 1 1
- . )7
. 18
- , 15
- . 04
- . 49
. 35
- ,40
- . 04
- , 20
- , 65
- , 07
, 13
, 28
- .20
. 20
- , 31
, 01
- , 08
- . 03
. 09
- . 23
, 02
- , 03
- , 15
- , 32
- , 59
- , 15
- . 45
, 00
, 00
- , 21
- , 24
- , 01
.43
. 00
, 25
2 , 18
, 48
'l
APPENDIX 9.1
Questionnaire for Final Product Testing
4.
nlu'in" u n ul'fl
w aflnwnl,unnlui7un L uuw aflnWInnlnNUI U L ui1n
ul L a n U : IW 8 ni ( UI L nlnUnl,u n' )
uinu1u'w anw l n u u uuau ; wa i'uu ; L nfilnu L uu 1 fl q un
Ulfl' U ; IW 2 1 n 7
w i ln nlna uu n ; u iu' L UUUU'nl 'U n l U 7 UU WaflnW L U

,
n l w a flu ; n l YIU'UU ; L nfioiu

nWlnfl O 'WaflnWnl,un u n''UanW; L U ' n U nIU L U'in l ' u n
* ,u L !l
,un l nina u ' U I UOU 2 - 3 nWl fl Ufln ' 'U
L d D l nnl >1 : H u i a ' IHl1 L fl O >1 1Jnl'u n >1 L I; a lf'l)\;1Jnl'U n ,j i

nWln I Wl n l U O7UlnWaU0 >1 n" U n L nlU i" nluu-a
" u >1 n "
flOUnl" L ;"' L I ; a . un u n u fl" 'ui u , 'u' I U a ; L fi u fl
5.
'I nfl
l h u 1
, u n u 1uuun ' ; I flHIUUn : I L UUflU )
n1ru'fltiUUUUUO'JJ' l .1n1U'Uu
.. HJJ'JU 4 H,!
nWI O n 1JIU lnluin a o ,j 'iwaflnWnl, u n ,j u n 'lI U u l ,j i' ( L U n' ; I L '
--
ilou,tJu
,
2536
a L fi u ' nunl'nl,un
6.
u u;u
ilula ; 2 - 3 fl i
iuO; 1J nln,ui u o i u n ' WI' L flWa

,
L Oua: 2 - 3 fl i
u u n , d iu; i >1
3u '1 ( 'U ; )
_
_
_
_
_
_
_
_
nWIUOnUI O n l ' U n nIU'U ;
2.
7.
n'ln
...
iOI'i L U >1
..
n', L nuou
n>1 IUI'
. tu
'1 ( 1u : )
_
_
_
_
_
_
_
_
flIUUnflnIU'inl, u n >1 rl o o : I'
3.
oun'Uln
OlnIUflDUI n li uDUn, nWI' L flWaIWa nWn I 1 U nn' >1 l n
UHU
PRITT
.
PELIKAN
t"
'l ( 1Ufl ; 1! )
_
_
_
_
_
_
_
_
_
_
tv
U1
00
8.
",u ti f d , : itl'l1tlu u: u'1'll1'Y1'1 -l 'U"ol tl -l ",uit) Coj ii 'l.ntll';; n ' ') u n -ln'l1 tJ 1iJ
i tJ / u u : u'1
1iJitJ/ 1iJuu:u'1
9.
tJ,
"'Uti 1 ' H ii ntll';;n ',)U"-lnfi ,) : "oI'U'lU 'fi''l ( 'I1UnU L "'nUn' ')'''u'

tJ'1 : '1 ) .
31-40
tJ,
'1 'fl ' -l n 1' Coj iintu.nn , ,) u " -l ",u'ltJ t '1 ' n
5-15
1 6- 2 0
il
2 1 - 30
il
'nn1, 4 0 il
il
un L Uu
ili'l unftn'
' '1 ' 'lS n ' '1
itnnUUl1'Y1 L tln'lSU
'1 ' fl ' -l n 1' Coj ii ntll';; n ,,)u"-l,,,u'ltJ'1 : '1 L a ntJU

,
'1 ' fi' L n'nUCojntll';; n ,,)u"-ln,u'tJ'1 : '1

'1 ' fi''1n 1'Cojiintll';; n , ,) u " -l "'U'tJ'1 : '1 L anUtlU
'fi''1n 1'Cojiintu.nn , ,) u " -ln,u'tJ'1 : '1 ' n
10 .
antII : tJ1tJn L UantII : L Q' : ,) "oI tJ Coj ii ntll';; n , ') u n n a nw: tl'l : a,)u -l'
'l1n'U tl'l1 tJ'Y1 tJ -l 'Cojntu.n O ' n ' '1 ,) ' -l '1'11 ' U 'un tJ -l fl '
n ; w' L tJ n tJu1iJ L nu 3 tJ LU'l1fi: UUU -l n
1
a,)u -l ' ' n
a'U -l ' L Uiiuu 2
a,)u -l' L Uiiuu 3
Coj ii ' n u u -l ua'1tJ : 'I1a -l "oltJ -l 1'Y1U

tJtlnU tl -l u ') tl
tJtlnUtl'
L UHiinw';; ' U''I1iJ
'1 ' fl ' n n 1 , n , ') u n -l U 'un tJ -l '
11.
n '1t11' 'll1fi'1L1u: u'1n,uti1' : L uutJ'1 : LU'lSU'Un''1NU,tJiutJ'1 -l U : a -l L an''1'

,
... ..
"oI tl -l H nn',) U 'Y1 -lU
..
..
n ;W" l1 fl ,) ' L U'l1 tl tJ L aUtlL1u: L nu 'nUn' '1 tJ tln UUUatluo'fii-lU
il
APPENDIX 9.1
(continued)
4. What size of glue stick do you normally use?
Questionnaire for Filial Product Teii ting
(Translation from Thai)
The sample you have been given is a new glue stick product. This product was
Small
8 grams
__
Medium 21 grams
__
Large
40 grams
developed in Thailand for Thai consumers. It is made from tapioca flour produced in
Thailand as the basic ingredient. We should
like
Now please answer the following questions after you have tested the new
you to test it now so that it can be
glue stick.
developed further for production and marketing in Thailand.

Note:
acceptability from glue stick characteristic only, not the turning up and down of the
instruction
Please
use
Since a suitable machine for glue stick packing was not available, please evaluate
tile given sample as you would use glue stick normally. Then complete the
stick
accompanying questionnaire alter you have tried the sample at least three times.
Please place a tick in front of the answer nearest to your answer for the question. Where
we have leli a space lor you to write, please give as detailed an answer
Note:
The questionnaire will be collected by the interviewer by
as
possible.
5. Please explain how you tried the glue stick sample (e.g. stick paper together, stiCk
photo onto paper, stick fabric onto paper etc.)
J une 93.
PART I
1. How often do you use glue stick?

___
___
6. Was the new glue stick product acceptable when you used it?
times a day
Yes
.Once a day
No If not acceptable, please tell us what was wrong with it.
23 times a week
2-3 times a month
Less than once a month
___
_
_
_
_
_
_
_
_
_
_
_
_
_
__
_
_
_
2. Do you buy the glue stick you use or it is prOvided by the office?
7 . How does this product compare to the glue stick you usually use? .
Very much better
Buy the glue stick myself
Slightly better
Provided by the office
The same
_
_
_
_
_
_
_
_
_
_
_
_
Slightly worse
Very much worse
3. What brand of glue stick do you normally use?

UHU
us
how it is different from your present glue
stick. __________
_
_
_
_
_
_
_
_
__
__
_
__
__
__
__
__
__
__
__
__
_
PRrIT
PELIFIX
If better or worse, please tell
_
_
__
__
__
__
_
__
_
_
8. Would you purchase or recommend to your company to purchase this new glue
PART IV
Age
stick?
Yes
Personal details
5 to 15
16-20
__
31-40
__
Over 40
__
21-30
No
9. At what price you think this glue stick should be sold?

Very much higher than the one you normally use
Slightly higher than the one you normally use
Tne same price as the one you normal! y use
Slightly lower than the one you normally use
Very much higher than the one you normally use
10. If this glue stick was available on the market, which ones of the following would
Occupation
__
School student
__
College /University student
__
Government office worker
__
Private sector office worker
attract you to purchase or try it. Please select three characteristics and give scores
Second attractive
Third attractive
Most attractive
Made in Thailand
:Vlade from Thai tapioca
Environmentally friendly
Safe
New product
Cheaper than the ones already in the market
11. Please make any other comments that could help us to develop and market this
product
tv
'0"
.. -'
APPENDIX 9.2
Question 1:
Question 2:
Summary of Results from Final Product Testing
How often do you use glue stick?
3-4 times
Source
Once . day
23 times a
day
University
Source
Once
23 times
week
a month
a month
(1.8%)
11 (20.0%)
24 (43.6%)
19 (34.5%)
55 (32.O'r.)
1 (1.9%)
2 (3.8%)
27 (50.9%)
18 (34.O'r.)
5 (9.4'.)
53 (30.8%)
Govern.
8 (23.5%)
4 (11.8%)
8 (23.5%)
11 (32.4%)
3 (S.S%)
34 (19.S%)
Private
4 (13.3%)
4 (13.3%)
10 (33.3%)
10 (33.3'10)
(6.7%)
30 (17.4'10)
Col. total
13 (7.6'10)
11 (6.4%)
56 (32.6%)
63 (36.6%)
29 (16.9%)
172 (100.O'r.)
df
50.583
12
Significance
Oti-square
df
Significance
17.063
0.002
1.559
0.S16
(Govern.,.Private)&Univ
26.974
0.000
(Govern.,. Private)&School
14.028
0.007
Comparison
Univ&School
Govem&Private
3-4 times
Once a day
a day
Govern.,.Private
School
23 times
Once
week
a month
a month
Row total
12 (18.8%)
8 (12.5%)
IS (28.1'.)
21 (32.S'.)
5 (7.8%)
64 (37.2'.)
1 (1.9%)
2 (3.S%)
27 (50.9'10)
18 (34.O'r.)
5 (9.4%)
53
(30.S%)
(1.8%)
11 (20.O'r.)
24 (43.6%)
19 (34.5%)
55
(32.O'r.)
11 (6.4%)
56 (32.6%)
63 (36.6%)
29 (16.9'k) 172 (100.O'r.)
University
Column total
23 times a
13 (7.6'10)
Provided
(92.7%)
4 (7.3%)
School
SO
(94.3%)
3 (5.7%)
Government
(23.5%)
Private
2 (6.7%)
Column total
III
121.267
25
57
(73.5%)
3 (5.,"".)
53
1 (2.9%)
34 (19.8%)
(30.8%)
30 (17.4%)
4 (2.3'10)
(33.1%)
Significance
Row total
55 (32.0%)
28 (93.3'.)
(64.5'10)
df
Others
172 (100.O'r.)
0.000
Oti-square
dt
Significance
Univ&School
6.98
0.031
Govern&Private
4.54
CGovern.,.Private)&Univ
70.40
(Govern.,.Private)&School
80.34
0.103
0.000
Provided
University
51 (92.7%)
4 (7.3%)
School
SO (94.3%)
3 (5.,"".)
Govern+Private
10 (15.6%)
Column Total
111 (64.5%)
0.000
Buy
Source
Source
Buy
Comparison
0.000
or it is provided by your office?
51
Chi-square
use
University
Row total
School
Chi-square
Do you buy the glue stick you
Others
Row total
55
(32.0%)
3 (5.7%)
53
(30.8%)
53 (82.8%)
1 (1.6%)
64 (37.2%)
57
4 (2.3%)
172 (100.0%)
(33.1'.)
Question 4:
What branp. of glue stick do you normally use?
Question 3:
UHU
Source
PRITT
PEUFIX
Row total
Source
55 (32.0%)
(1.9"10)
Others
What size
of glue
stick do you normally use?
Medium
University
43 (78.2%)
U (21.8%)
55 (32.0%)
53 (30.8%)
School
38 (71.7%)
15
(28.3%)
53 (30.8%)
University
54 (98.2%)
1 (1.80/.)
School
45 (84.9%)
7 (13.2%)
Govern.
31 (91.2"10)
2 (5.9"10)
(2.9%)
34 (19.8%)
Govern
13 (38.2%)
19 (59.9%)
Private
25 (83.3%)
2 (6.7"t.)
3 (10.0"10)
30 (17.4"10)
Private
21 (70.0%)
9 (30.0%)
4 (2.3%)
5 (2.9"10)
172 (100.0%)
115 (66.9%)
55 (320%)
ISS (90.1%)
Col. total
Chi-square
8 (4.7%)
df
26.755
Significance
Large
Small
Column total
Chi-square
0.002
21.68
df
Significance
Comparison
Chi-square
df
Row total
2 (5.9%)
34 (19.8%)
30 (17.4"10)
2 (1.2%)
172 (100.0%)
0.001
Chi-square
df
Significance
Univ&5chool
6.28
0.043
Univ&5chool
0.605
0.-137
Govem&Private
1.40
0.497
Govem&:Private
i.23
0.027
(Govem+Private)&Univ
8.40
0.038
(Univ+School)&:Private
0.034
0.581
13.08
0.004
(Univ+School)&Govern
19.15
(Univ+School+Private)&Govem
28.885
Comparison
(Govern+Private)&Schoo1
Source
UHU
PRITT
University
54 (98.2%)
(1.8%)
School
45 (84.9%)
7 (13.2%)
Govern+Private
56 (87.5%)
PEUFIX
Others
ISS (90.1%)
8 (4.7"t.)
0.000
0.000
Row total
55 (32.0%)
1 (1.9%)
53 (30.8%)
4 (6.3%)
4 (6.3%)
64 (37.2%)
4 (2.3%)
5 (2.9%)
172 (100.0%)
Source
Univ+School+Private
Govern
Column total
Column total
Significance
Small
Medium
102 (73.9"10)
36 (26.1%)
13 (38.2%)
19 (59.9%)
2 (5.9%)
34 (19.8"10)
115 (66.9%)
55 (32.0%)
2 (1.2%)
172 (100%)
Row total
138 (80.2%)
Question 6:
Would you purchase or recommend to your office to purchase this new
Question 8:
Was the new glue stick product acceptable?
glue stick?
Acceptable
Not acceptable
University
39 (70.9%)
16 (29.1'1'0)
55
School
36 (67.9%)
12 (32.1%)
53 (30.8%)
Govern
23 (67.6%)
11 (32.4%)
34 (19.8%)
Private
19 (633%)
11 (36.7'ro)
30 (17.4%)
117 (68.0%)
55 (32.0%)
172 (100%)
Source
Column tocal
Chi-square
Question 7:
elf
0.156
Significance
Row tocal
(32.0%)
Very much
Slightly
better
better
University
1 (1.8%)
7 (12..7%)
School
1 (1.9%)
Govern
The same
Slightly
No
University
31 (56.4%)
24 (43.6%)
55
School
30 (56.6%)
23 (43.4%)
53 (30.8%)
Government
22 (64.7'ro)
12 (35.3%)
34 (19.80)
Private
15 (SO.O'l'o)
15 (SO.O%)
30 (17.4%)
Column total
98 (57.0%)
74 (43.0%)
172 (100.0%)
Chi-square
0.915
How does this product compare to the glue stick you usually used?
Source
Yes
Source
Very much
Row tocal
elf
1.436
Significance
At what price you think this glue stick should be sold?
Question 9:
Slightly
Source
worse
worse
19 (345%)
22 (40.0%)
6 (10.9%)
55 (32.0%)
9 (17.0%)
14 (26.4%)
23 (43.4%)
6 (11.3%)
53 (30.8%)
School
6 (17.6%)
10 (29.4%)
13 (38.2%)
5 (14.7'ro)
34 (19.8%)
Government
Private
Column total
2 (6.7%)
3 (10.0%)
6 (20.0%)
14 (46.7'ro)
5 (16.7'ro)
30 (17.4%)
Column
4 (2.3%)
25 (145%)
49 (285%)
72 (41.9%)
22 (12.8%)
172 (100.0%)
The same
Slightly lower
Very much
6.196
elf
12 (21.8%)
37 (673%)
6 (10.9%)
55 (32..0% )
4 (75%)
38 (72.2%)
8 (15.1%)
53 (30.8%)
6 (17.6%)
24 (70.6%)
4 (11.8%)
34
1 (33'1'0)
5 (16.7"10)
18 (60.0%)
6 (20.0%)
30 (17.4%)
4 (2.3%)
27 (15.7%)
117 (68.0%)
24 (14.0%)
172 (100.0%)
University
Chi-square
=
12
Significance
0.863
3 (5.7'ro)
10.135
Row tocal
lower
total
Chi-square
(32.0"10)
0.697
higher
Private
Row tocal
df
Significance
0.340
(19.8%)
Question 10:
If this glue was available on the market, which ones would attract you
to purchase or try it?
So=
Please choose three characteristics and give the score
Most attractive
Second attractive
Third attractive
Environmental friendly
Made in Thailand
Second
Third
attractive
attractive
Most attractive
Not attractive
Row total
University
16 (29.1%)
13 (23.6%)
7 (12.7%)
19 (34.5'1'.)
55 (32.0%)
School
24 (.3%)
15 (28.3'1'.)
(13.2'1'.)
7 (13.2%)
53 (30.8%)
Government
3 (8.8%)
9 (26.5%)
2 (8.8%)
19 (SS.9%)
34 (19.8%)
Prival1!
5 (16.7%)
8 (26.7%)
3 (10.0%)
14 (46.7%)
30 (17.4%)
48 (27.9%)
45 (26.2'1'.)
20 (11.6%)
59 (34.3%)
172 (100..)
Colwnn total
d1
0.003
Most
Second
Third
attractive
attractive
attractive
8 (14.5%)
7 (12.7%)
9 (16.4%)
31 (56.4%)
55 (32..)
Compansons
Chi.,;quare
4 (7.5%)
17 (32.1%)
4 (7.5%)
28 (52.8%)
53 (30.8%)
Univ&School
7.247
0.064
Government
4 (11.8%)
10 (29.4%)
5 (14.7%)
15 (44.1%)
34 (19.8%)
Govem&.Prival1!
1.071
0.784
Priva",
8 (26.7%)
6 (20..)
4 (13.3%)
12 (4O..)
30 (17.4%)
(GovemPrival1!)&School
24.282
0.000
24 (14.0%)
40 (23.3%)
22 (12.8%)
86 (50..)
172 (100%)
(GovemPrival1!)&:Univ
6.402
0.094
Source
University
School
Colwnn total
Chi-square
d1
13.408
Significance
Not attractive
Row total
Chi-square
25.086
Significance
(Govem+PrivaUniv)&:School
1.145
Second Third attractive
Most attractive
Source
12 (21.8%)
6 (10.9%)
25 (45.5%)
55 (32.0%)
School
U (22.6%)
11 (20.8%)
9 (17.0%)
21 (39.6%)
53 (30.8%)
Government
6 (17.6'.)
9 (26.5%)
3 (8.8%)
16 (47.1".)
34 (19.8%)
Private
6 (2O.O'r.)
6 (2O.O'r.)
2 (6.7%)
16 (53.3%)
30 (17.4%)
36 (20.9%)
38 (22.1'.)
20 (11.6%)
78 (45.3".)
172 (100.0%)
Column total
Chi-square
3.704
df
significance
0.903
0.000
Second
Third
Not
attractive
attractive
attractive
School
24 (45.3%)
15 (28.3%)
7 (13.2'1'.)
7 (13.2%)
53 (30.8%)
Univvem+Private
24 (20.20/.)
30
(25.2'1'.)
13 (10.9%)
52 (43.7%)
119 (69.2%)
Colwnn total
48 (27.9%)
45 (26.2%)
20 (11.6%)
59 (34.3%)
attractive
U (21.8%)
Most
Row total
University
Significance
attractive
Source
Not attractive
df
18.524
Made from Thai Tapioca
Row total
172
(lOO..)
Cheaper than the ones already in the market
Safe
Most
Second
Third
attractive
attractive
attractive
University
15 (27.30/.)
14 (25.5%)
3 (5.5%)
23 (41.8%)
55 (32.0%)
School
23 (43.4%)
8 (15.1%)
8 (15'!%)
14 (26.4%)
53 (30.8%)
9 (26.5%)
8 (23.5%)
1 (2.9%)
16 (47.1%)
34 (19.8%)
Private
10 (33.3%)
7 (23.3%)
3 (10.0%)
10 (33.3%)
30 (17.4%)
Column total
57 (33.1%)
37 (21.5%)
IS
63 (36.6%)
172 (100%)
Source
Not attractive
Row total
Source
Most attractive
Second attractive
(14.5%)
8 (14.5%)
9 (16.4%)
30 (54.5%)
55
11
(20.8%)
(20.80/.)
12 (22.6%)
19 (35.8%)
53 (30.8%)
5 (14.70/0)
17 (50.0%)
34 (19.80/.)
University
School
Chi-square
11.746
df
Column total
(8.7'r.)
Significance
Chi-Square
0.228
Most
Second
Third
attractive
attractive
attractive
University
6 (10.9%)
5 (9.1%)
10 (18.2%)
34 (61.8%)
55 (32.0%)
School
8 (15.1%)
8 (15.1%)
6 (11.3%)
31 (58.5%)
53 (30.8%)
Government
4 (11.8%)
8 (23.5%)
1 (2.9%)
21 (61.8%)
34 ( 19.8%)
Private
5 (16.70/.)
6 (20.0%)
7 (23.3%)
12 (40.00/0)
30 (17.4%)
23 (13.4%)
27 (15.7%)
24 (14.0%)
98 (57.0%)
172 (1000/0)
Source
Column total
11.617
df
Significance
Not attractive
0 256
Row total
(5.9%)
10 (29.4%) .
Row total
(32.0%)
9 (30.0%)
2 (6.7'r.)
2 (6.7'ro)
17 (56.7%)
30 (17.4%)
30 (17.4%)
31 (18.0%)
28 (16.3%)
83 (48.30/.)
172 (100.0%)
Private
New Product
Chi-square
Government
Government
11
Third attractive Not attractive
16.773
df
Significance
0.052
APPENDIX 9.3
APPENDIX 9. 4
Cross-tabulation o f ' Reason Consumers Thought that the Developed Glue Stick was
Cross-tabulation of Comparison of Developed Glue Stick with Glue Stick Consumers
Better or Worse than their Glue Stick' by 'Acceptability'
Normally Used by Purchase Intention
Accept
Reason
Weaker
bond
so (65.8%)
accept
Row total
44 (86.3'1.)
94 (74.0'1.)
Not
Source
Buy
Very much beller
Slightly beller
22 (88.00/.)
3 (12.00/.)
25 (14.5%)
TI,e sam.
46 (93.9%)
3 (6.00/.)
49 (2B.50/)
Slightly worse
25 (34.7%)
47 (65.3%)
72 (41.9%)
Not buy
Row Total
(HXI.O%)
4 (2.3%)
6 (7.9%)
6 (4.7%)
Not messy
(7.9%)
6 (7.4%)
Dry slowly
(530/.)
(7.8%)
8 (6.3%)
Very much worse
1 (4.5%)
21 (95.5%)
22 (12.8%)
3 (3.90/.)
(2.!)%)
4 (3.1%)
Col UUUl total
98 (57.0%)
74 (43.0%)
172 (100.00/.)
(2.0%)
3 (2.4%)
Srronger bond
Disintegrate
Low diSintegrate
2 (2.6'1'. )
No unpl.asant odour
(13%)
(0.8%)
Not wrinkly
(1.3%)
(0.8%)
Very slippery
(1 3'1'.)
(0.8%)
( 1.3%)
(O.ll%)
Colour too white

Messy
(2.00/.)
(0.8%)
51 (40.2%)
127 (100.00/.)
Low degree of coating
Column total
(0.8%)
(1 3%)
76 (59.8%)
Chi-square = 79.273
Si gnificance = 0.000
dE =4
Comparison
Chisquare
Very much&slightly better
0.535
0.464
(Vry much...slightly better)&the
0.457
0.499
df
Significance
same
Very much&Slightly worse
7.669
(Betteathe same)&slightly wor>e
54.337
0.000 .
(Beller+the same)&very much
67.057
0.000
0,006
worse
Source
Buy
Not buy
Row toW
Bellu+the same
72 (923'1'.)
6 (7.7%)
78 (453%)
Slightly worse
25 (34.7'1'.)
47 (65.3%)
72 (41.9%)
Very much worse

Coiumn total
(4.5%)
21 (95.5%)
22 (12.8%)
98
(57.00/.)
74 (43.0%)
172 (100.00/.)

Consumer Buying Behaviour On Glue Stick

Uploaded by

Consumer Buying Behaviour On Glue Stick

Uploaded by

Copyright is owned by the Author of the thesis.

Permission is given for

CONSUMER RESEARCH AND PRODUCT OPTIMIZATION

and in various industries.

I would like to express my deepest gratitude to my supervisor, Professor

I wish also to thank my co-supervisors, Professor Vichai Haruthaithanasan, Product

"'The staff of the Consumer Technology Department, Massey

I NIT IA L DE VELO P ME NT ST U DY O N GL UE STIC K FOR M U LAT IO N

Criteria for Selection of Preliminary Formulation

A TT RlBUTE S OF CO M ME RCIAL GL UE STIC K S

CO M ME RCIAL P ROD UCT S

RELATIO N S HI P S BETWEE N GL UE STIC K ATTR IB UTE S

Correlations bet ween Glue Stick Physical Attributes

RELATIO N S HI P S BETWEE N I N GREDIE NT S A N D GL UE ST IC K

Effect of Ingredients on Attributes of Glue Sticks

Selection of Physical Attribute Measures

PRO DUCT FO R M ULATIO N DE VELO P ME NT WIT H CO N S U MER

AIM AND OBJECTIVES

Consumer Panel in Thailand

IDEAL RATIO SCO RE S A ND LO G OF I DEAL RATIO SCORE S

Sensory Attributes of the

FINAL CONSUMER TESTING OF PRODUCT

USE OF EXPERIMENTAL DESIGNS IN PRODUCT

Six components extreme vertices

Users of each glue product

Frequency of glue product usage

Purposes of glue product usage

Attributes of glue stick product needing improvement

Attributes of self-adhering note pad needing improvement

The attributes of a glue stick product

Importance of glue stick attributes from open-ended elicitation method

Importance of self-adhering note pad attributes

I mportance of glue stick product attributes from panel in New

I mportance of self-adhering note pads a ttributes from panel in

Importance of glue product attributes from Thailand panel

Importance of permanent bonding glue stick attributes

Importance of temporary bonding glue stick attributes

Acceptability of glue products by panel in Thailand

Buying intention of consumers towards the new product

Usage time for one stick of glue

Price of.glue stick consumer prepared to pay

Formulations used in preliminary experimentation

Properties of glue stick samples from preliminary experimentation

Experimentation plan for dextrin experiment

Physical attributes of dextrin glue sticks

Experimentation plan for casein experiment

Empirical equations showing relationships between ingredients and

Empi rical equations showing relationships between ingredients and

Physical attribu tes of the commercial glue sticks

Sensory attribu tes of the commercial glue sticks

Experimental plan for prototype development

Physical attributes of the glue stick samples from constrained mixture

Sensory a ttributes of the glue stick samples from mixture design

Comparison between physical attributes of prototype products and

Comparison of sensory attributes of prototype products and commercial

Correlations between physical attributes of glue stick samples from

Correlations between sensory attributes of glue stick samples from

Correlations between physical and sensory attributes of glue stick

Correlations between ingredients and output variables

Empirical equations showing relationship between ingredients and

Experimental plan for mixture design with three components

Physical attributes of the prototypes

Comparison between physical attribu tes of the prototypes and

Acceptability, purchase intention and price to buy of the prototypes

Sensory attribu tes of the prototypes

Ideal scores of the sensory attributes of the prototypes

Correlations between sensory scores obtained from three groups of

Ideal ratio scores of the sensory attributes of the prototypes (Original

Ideal ratio scores of the sensory a ttributes of the prototypes

Correlations between physical attributes of glue stick samples

Correlations between sensory attributes of glue stick samples

Correlation between acceptability and sensory attributes

Correlations between sensory attributes and physical attributes

Correlations between the components and the attribu tes

Empirical equations showing relationships between glue stick a ttributes

Costs of the ingredients used in glue stick formulation