SURVEY METHODOLOGY
Prof. Amar KJR Nayak
SURVEY METHODOLOGY
What is a Survey?
What are the key features of this
Methodology?
What is the philosophical foundation of
Survey Methodology?
What are the basic statistical assumptions of
this methodology?
What Is a Survey?
Survey is a means to get a prompt and accurate flow of
information on preferences, needs, and behavior. It is
in response to this critical need for information on the
part of the government, business, and social
institutions that so much reliance is placed on surveys.
"Survey" is used most often to describe a method
of gathering information from a sample of
individuals. This "sample" is usually just a fraction
of the population being studied.
Census v/s Survey
The sample is not selected haphazardly or only
from persons who volunteer to participate. It is
scientifically chosen so that each person in the
population will have a measurable chance of
selection.
Who undertakes Survey?
An Experience
Impact assessment of RLTAP
Drought Proofing, Poverty alleviation &
Quality of Life
8 schemes, 130 variables, 4500 respondents in
58 of the 80 blocks
Workshop – ‘Do not get into kaduo/mud’
Highly efficient method of research
Low Cost
Fast
Possibility of Generalization
Proper Sampling
Representative and Adequate (H & V)
Highly Clinical in Approach
Garbage in – Garbage out
Statistical Assumptions
Interpretation with caution
Philosophical Foundation of Survey Methodology
Ideational Perspective
V/s
Materialistic Perspective
Respondents say what they think
Application
Answers to questions like what?
Conditions for good Survey
Sample size is significant
Sample is representative of the population
Principles of good estimation is followed
(unbiased, efficient, consistent, sufficient)
All respondents are sincere in their responses
Biases in the responses get normalized due to
proper sampling
Criteria of a good Estimator
Unbiased
If the statistic tends to assume values that are above the
population parameter as frequently as it assumes values
that are below the population parameter.
Efficiency
It refers to the size of the standard error of the statistic
If we compare two statistics from a sample of the same size,
and try to decide which one is the more efficient estimator,
we would pick the statistic with the smaller standard error
Consistency
If as the sample size increases, the statistic becomes
closer to the values of the population parameter, then
that statistic is consistent.
Sufficiency
An estimator is sufficient if it makes so much use of the
information in the sample that no other estimate could
extract from the sample, additional information about the
population parameter.
Some Basics of Survey
Sampling & Estimation
Sampling
1. Simple random Sampling: equal probability of being
picked
2. Systematic Sampling: selected at an uniform
interval
3. Stratified Sampling: selected from homogeneous
groups / strata
4. Cluster Sampling: make clusters and choose
any one of them
Statistical inference is based on
simple random sampling
Sampling Distribution:
¾ Sampling distribution of the mean:
A probability distribution of all the possible means of the
samples is a distribution of sample means.
¾Sampling distribution of the median:
¾Sampling distribution of the proportion:
How can we infer population characteristics
from a Sample?
The Central Limit Theorem:
The mean of sampling distribution of the mean will equal
the population mean regardless of the sampling size, even
if the population is not normal.
As the sample size increases, the sampling distribution of
the mean will approach normality, regardless of the shape
of the population distribution.
The significance of the central limit theorem is that it
permits us to use sample statistics to make inferences
about population parameters, without knowing anything
about the shape of the frequency distribution of that
population other than what we can get from the sample.
The sample frequency distribution:
x1 x2 x3 x4
Research Principles & Ethics
Descriptive statistics
(parametric)
¾ Sample
¾ Population
¾Statistics
¾Parameter
¾Raw data, source, & authenticity
Standard error:
The standard deviation of the distribution of a sample
statistic is known as the standard error of the
statistics.
Example: standard deviation of the distribution of sample
means is termed as standard error of the mean.
μ
σ = standard deviation of this distribution
The population distribution:
μ = the mean of the distribution
The sampling distribution of the mean
μx
μx = mean of the sampling distribution of the means
σx = standard error of the mean
= standard deviation of the sampling distribution of mean
Sampling from Normal Populations
Properties of the sampling distribution of the mean when
the population is normally distributed
μx = μ
σx = σ/√ n
Standard error of the mean for infinite population:
σx = σ/√ n
Standard error of the mean for finite populations:
σ (N - n)
σ x = ---- * √ ----------
√n n-1
With a finite population multiplier
Standardizing the sample mean:
Standard score; standard deviation from the mean of a
standard normal probability distribution
x - μ
Z= -------
σx
Sample mean, population mean, standard error of the
mean
Estimation
Reason for estimates: To make statistical inferences about the
population from a sample.
Types of estimate:
Point Estimate: It is a single number that is used to estimate
an unknown population parameter.
Limitations
Often insufficient, right or wrong
Example: Total weight of students, CGPA of students in a
high school
Point estimate is more useful, if it is accompanied by an
estimate of the error that might be involved.
Interval Estimate: It is a range of values used to estimate
a population parameter.
How Large Must The Sample Size Be?
The sample size required for a survey partly depends on
the statistical quality needed for survey findings
Much depends on the professional and financial resources
available.
For example, the well-known national polls frequently
use samples of about 1,000 persons to get reasonable
information about national attitudes and opinions
⎡ Zα ⎤
n = ⎢ 2
⎥ pq * DE
⎢⎣ E ⎥⎦
n = sample size
E = Margin of Error
DE = Design Factor
p = 0.5
q = 0.5
Surveys provide a speedy and economical means
of determining facts about our economy and about
people's knowledge, attitudes, beliefs,
expectations, and behaviors.
The recommended policy for survey organizations
to safeguard such confidentiality includes
Using only number codes to link the respondent to a
questionnaire and storing the name-to-code linkage
information separately from the questionnaires
Refusing to give the names and addresses of survey
respondents to anyone outside the survey organization,
including clients
Destroying questionnaires and identifying information
about respondents after the responses have been
entered into the computer
Omitting the names and addresses of survey
respondents from computer files used for analysis
Presenting statistical tabulations by broad enough
categories so that individual respondents cannot be
singled out
