Measurement and Scaling

SCALES OF MEASUREMENT: NOMINAL

SCALE
• Nominal scale: When data are labels or names used to identify the
attribute of an element, the nominal scale is used.
• Example: Marketing research company conducting a survey in three
Indian towns: Bhopal, Nagpur, and Baroda. Numeric codes assigned for
convenience: 1 for Bhopal, 2 for Nagpur, and 3 for Baroda. Numeric
values used for identification, not indicating ranking or rating.
• Other examples of nominal data: Employee identification numbers,
contributory provident fund numbers, personal identification number,
and so on.
SCALES OF MEASUREMENT: ORDINAL
SCALE
• Ordinal Scale: In addition to nominal level data capacities, ordinal scale
can be used to rank or order objects.
• Example: Manufacturing company surveys 150 consumers for product
perception. Three options given: excellent, good, or poor. Numeric
values assigned for convenience: 1 for excellent, 2 for good, and 3 for
poor. Numeric values imply a ranking: 1 is the best (excellent), 2 is
intermediate (good), and 3 is the worst (poor). This numeric scale
represents an ordinal level of measurement.
• Ordinal data allows for ranking but doesn't specify the magnitude of
differences between values.
SCALES OF MEASUREMENT: INTERVAL
SCALE
• Interval Scale : In interval level measurement, the difference between
the two consecutive numbers is meaningful
• Example: Three M.Sc. Statistics students scored 65, 75, and 85 in
reliability theory. Performance can be ranked: 85 is the highest, 65 is the
lowest, and 75 is in the middle. Numeric values indicate both ranking
and meaningful differences between scores. This scenario involves an
interval level of measurement.
• Interval data allows for ranking and meaningful differences, but there is
no true zero point.
SCALES OF MEASUREMENT: RATIO
SCALE
• Ratio Scale : Ratio level measurements possess all the properties of
interval data with meaningful ratio of two values.
• Example: Company offers two toothbrushes priced at Rs 30 and Rs 15.
In the ratio scale, the difference between the prices (Rs 30 - Rs 15 = Rs
15) is meaningful. Ratio scale allows meaningful calculations and
comparisons, such as determining that the first product's price is two
times that of the second product. This situation involves a ratio level of
measurement.
• Ratio data includes a true zero point, allowing for meaningful ratios and
proportions between values.
FOUR LEVELS OF DATA
MEASUREMENT
FOUR LEVELS OF DATA
MEASUREMENT
• Nominal data has limited utility in analytical and statistical tools.
Ordinal data provides more statistical analysis features compared to
nominal data. Interval level data has additional properties beyond
nominal and ordinal levels. Ratio level data allows for ratio
comparisons and supports statistical analyses applicable to nominal,
ordinal, and interval data.
• Statistical tools can be categorized into parametric and non-parametric
based on data level:
• Parametric statistics: Suitable for interval and ratio data.
• Non-parametric statistics: Suitable for nominal and ordinal data.
MEASUREMENT SCALES
I. SINGLE ITEM SCALES
• As clear from the name, the single-item scales measure only one item as
a construct.
• Some of the commonly used single-item scales in the field of business
research are:

A. Multiple- choice scales,

B. Forced-ranking scales,
C. Paired-comparison scales,
D. Constant-sum scales,
E. Direct quantification scales, and
F. Q-sort scales
A. MULTIPLE-CHOICE SCALE
B. FORCED RANKING SCALES
• In forced-choice ranking scaling technique, the respondents rank different
objects simultaneously from a list of objects presented to them.
C. PAIRED COMPARISON TECHNIQUE
 D. CONSTANT SUM SCALES
• In the constant-sum scaling technique, respondents allocate points to
more than one stimulus objects or object attributes or object properties,
such that the total remains a constant sum of usually 10 or 100.
E. DIRECT QUANTIFICATION SCALE
• The simplest form of obtaining information is to directly ask a question
related to some characteristics of interest resulting in ratio- scaled data.
II. MULTI-ITEM SCALES
•Multi-item scaling techniques generally
generate some interval type of
information. In interval scaling
technique, a scale is constructed with the
number or description associated with
each scale position. Therefore, the
respondent’s rating on certain
characteristics of interest is obtained.
A. SUMMATED SCALING TECHNIQUE: THE LIKERT SCALES

• The Likert scale, developed by Rensis Likert, is a widely used technique

in business research. It has five rating categories for each item response,
with "strongly disagree" to "strongly agree" as extremes and "disagree"
to "agree" in the middle.
• The scale typically ranges from 1- to 5-points, but some researchers use
other numbers. Analysis can be done using profile analysis or summated
analysis, which sums scores across scale items, resulting in an average
for all respondents.
• The summated approach is widely used, making the Likert scale also
known as the summated scale.
A. SUMMATED SCALING TECHNIQUE:
THE LIKERT SCALES
Parasuraman et al. (2005) developed a
multi-item scale (E-S-QUAL) to
measure the service quality delivered by
Websites in which online shopping is
available for the customers (Figure
3.10). Each item of the scale is rated on
a 1- to 5-point rating scale in which 1
indicates strongly disagree and 5
strongly agree.
 B. SEMANTIC DIFFERENTIAL SCALES
• Semantic differential scale was developed by Charles Osgood, George Suchi,
and Percy Tannenbaum in 1957.
• The semantic differential scale consists of a series of bipolar adjectival words
or phrases placed on the two extreme points of the scale. Good semantic
differential scales keep some negative adjectives and some positive adjectives
on the left side of the scale to tackle the problem of the halo effect.