Unit 4

Census and Sample Survey

Census/ Complete Survey

Enumeration
Study of each and every Study of each and every
item of the population item of the sample selected
from a population
 Sample Design

• A technique or procedure adopted for selecting items

(sample size) for the sample from the population.
• Determined before data collection
 Steps in Sampling Design
• Type of universe
• Sampling unit
• Sampling frame/ Source list– contains the names of all items of
a universe
• Sample size – optimum size which fulfills requirements of
efficiency, representativeness,, reliability and flexibility.
• Parameters of interest
• Budgetary constraint
• Sampling procedure
Essentials of Sampling
Representativeness

Adequacy

Independence

Homogeneity
 Characteristics of a Good Sample Design
• Truly representative sample
• Small sampling error
• Viable in context of fund availability
• Generalization of sample results
 Errors in Research
Sampling Error Non-Sampling Error
Due to drawing inferences about • Errors that occur in acquiring, recording
population on the basis of sampling. or tabulating statistical data.
• Biased Errors: arise from any bias in • Sample Size ↑ --- Non-Sampling Error↑
selection, estimation etc. Bias arises
due to faulty process of selection, data
collection and faulty methods of
analysis. Can be avoided by drawing
the sample either entirely at random
or at random subject to restrictions.
• Unbiased Errors: due to chance
differences between the members of
population included in the sample and
those not included.
Methods to reduce Sampling Errors
• Increase the sample size. A larger sample size leads to a more precise
result because the study gets closer to the actual population size.
• Divide the population into groups
• Know your population
• Randomize selection to eliminate bias
• Train your team
• Perform an external record check
Systematic BIAS
• INAPPROPRAITE SAMPLING FRAME: Is sampling frame is
inappropriate. i.e. a biased representation of the universe, it will
result in a systematic bias
• Defective measuring device
• Non respondents
• Indeterminacy principle (Individual act differently under observation/
non-observed)
• Natural bias (Salary for tax purposes and social affluency)
 Types of Sampling Designs

2. Non-Probability
1. Probability Sampling/ Non-
Sampling/ Random Random Sampling
Sampling Method Method
Probability/ Random Sampling

Every item has an equal chance of

being selected in the sample.
No personal bias in selection of
items.
Selection of items is by chance or
random selection.
 Random Sampling Method
Advantages
• No dependence upon the existence of detailed information about the
universe.
• Unbiased estimates having measurable precision.
• Possible to evaluate the relative efficiency of various sample designs.

Limitations
• Requires very high level of skill and experience.
• Requires a lot of time to plan and execute.
• Generally large cost involved.
Probability Sampling Design
1. Probability
Sampling/ Random
Sampling Method

Unrestricted
Sampling/ Simple
Random
Sampling

Restricted
Sampling
Unrestricted Sampling/ Simple Random Sampling

Every item has an equal chance of

being selected in the sample.

No personal bias in selection of items.

Selection of items is by chance or

random selection.
 Unrestricted Random Sampling

Merits Limitations

Necessitates a
No possibility of personal
completely
bias
catalogued
universe

With increase in size sample

becomes more representative
Sample size
of the population
usually larger

Easy assessment of accuracy of

estimates Costly and time
taking
 Restricted Random Sampling

Every item has an equal chance

of being selected in the sample.

Works under certain

conditions.
 Restricted Sampling Method

Systematic
Sampling/ Quasi
Random
Sampling

Multistage
Sampling/ Cluster Stratified
Sampling Sampling
 1. Stratified Sampling

Universe to be sampled is subdivided into groups (mutually

exclusive) and include all items in the universe.

A simple random sample is then chosen independently from

each group.
 Stratified Sampling

Greater
More Greater
Merits representative accuracy
geographical
concentration

Random
Utmost care Cost per
selection of
Limitations exercised for
stratification
items from each
observation
quite high
stratum
2. Systematic Sampling/ Quasi Random Sampling
A systematic sampling is formed by selecting one unit at random
and then selecting additional units at evenly spaced intervals
(sampling interval/ sampling ratio) until the sample has been
formed.

Method is used when a complete list of population is available.

List may be prepared in alphabetical, geographical, numerical or

some other order.

Sampling Interval (k) = Universe Size (N)/ Sample Size (n)

Systematic Sampling/ Quasi Random Sampling

Simple and Less time Results

Merits convenient and work generally
to adopt involved satisfactory

Becomes less representative

Limitation in case of populations
having ‘hidden periodicities’
3. Multistage Sampling/ Cluster Sampling
There are several stages in which the sampling process is carried
out.

First stage units are sampled by some suitable method such as

simple random sampling.

Then a sample of second stage units is selected from each of the

selected first stage units again by some suitable method that
may be the same or different from the method employed for
first stage units. Similarly further stages can be added as
required.
 Multistage Sampling/ Cluster Sampling

More flexibility Permits large Permits field

Merits in sampling
method
area to be
covered
work to be
concentrated

In general, less accurate than a sample containing the

Limitation same number of final stage units selected by some
suitable single stage process
Non-Probability/ Non-Random
Sampling
Every item has an unequal chance of
being selected in the sample.

Personal bias in selection of items.

Selection of items is not by chance.

Non-Probability Sampling Design
Non-Probability
Sampling
Method

Convenience
Sampling

Quota
Sampling

Judgement
Sampling
 1. Convenience Sampling/ Chunk

Chunk refers to that fraction of population being

investigated which is selected neither by probability
nor by judgement but by convenience.
Convenience samples are prone to bias by their
very nature – selecting population elements which
are convenient to choose.

Often used for making pilot studies.

 Convenience Sampling

• Results obtained can hardly be

representative of the population.
• Results generally biased and
Dem unsatisfactory.

erits
 2. Quota Sampling

A type of judgement sampling.

Quotas are set up according to some specified characteristics

such as so many in case of several income groups, so many in
each age, so many with certain political or religious affiliations
and so on.

Selection of sample items depends on personal judgement.

 Quota Sampling

• Often used in public opinion studies.

• Occasionally provides satisfactory results.
M
eri • Cost per person interviewed relatively
ts small.
• Numerous opportunities for bias.
De
me • Not widely used in practical work.
rits
 3. Judgement Sampling

Includes the items

Choice of sample in the sample which
items depends are most typical of
exclusively on the universe with
investigator’s regard to the
judgement. characteristics
under investigation.
 Judgement Sampling

• Used in solving many types of economic & business problems.

• Most suitable when only a small number of sampling units is in
Mer the universe.
• Useful in making public policy decisions.
its

• Not a scientific method hence can not b recommended for

general use.
Limi • No objective way of evaluating the reliability of sample results.
tatio • Success depends upon the excellence in judgement.
ns
Factors to be Considered while deciding the Sample Size
Universe
Size
Homogenei
ty/ Availability
Heterogen of
eity of Resources
Universe

Factors
Degree of Nature of
Accuracy Study

Nature of Sampling
Respon- Method
dents Adopted
 Less time
consuming

Used to
judge
accuracy of Less
information financial
obtained burden
through
census
Merits of
Sampling

Possible to
More
collect more
reliable
detailed
results
information
 Qualified &
experienced
personnel

Census if
Carefully planned Limitations information
& executed of Sampling required for all
units

Complicated if
sample requires
large proportion
of population
Determine the sample size
• Having the right sample size is crucial in finding a statistically
significant result.
• The larger the sample size, the more reliable the results; however,
larger sample size means more time and money.
• So, how do you determine the right sample size for your market
research?
Methods of Determining Sample size
• Arbitrary Approach: Sample can be between 5% and 10% of the
population.
• Conventional Approach: Based on the average size of similar studies
conducted.
• Statistical Analysis Approach: Based on the proposed statistics.
• Cost-Benefit Approach: Base on availability of resources and benefits.
Mainly used in Non-Probability sampling.
• Confidence Interval Approach: Based on standard deviation, margin of
error and confidence level.
Sample Size Variables Based on Target Population

• Population Size — How many total people fit your demographic? For instance, if you want to know
about mothers living in the US, your population size would be the total number of mothers living in
the US. Not all populations sizes need to be this large. Even if your population size is small, just know
who fits into your demographics. Don’t worry if you are unsure about this exact number. It is common
for the population to be unknown or approximated between two educated guesses.
• Margin of Error (Confidence Interval) — No sample will be perfect, so you must decide how much
error to allow. The confidence interval determines how much higher or lower than the population
mean you are willing to let your sample mean fall. If you’ve ever seen a political poll on the news,
you’ve seen a confidence interval. For example, it will look something like this: “68% of voters said yes
to Proposition Z, with a margin of error of +/- 5%.”
• Confidence Level — How confident do you want to be that the actual mean falls within your
confidence interval? The most common confidence intervals are 90% confident, 95% confident, and
99% confident.
• Standard of Deviation — How much variance do you expect in your responses? Since we haven’t
actually administered our survey yet, the safe decision is to use .5 – this is the most forgiving number
and ensures that your sample will be large enough.
• Your confidence level corresponds to a Z-score. This is a constant
value needed for this equation. Here are the z-scores for the most
common confidence levels:
• 90% – Z Score = 1.645
• 95% – Z Score = 1.96
• 99% – Z Score = 2.576
Sample Size Formula

• Sample Size = (Z-score)2 * StdDev*(1-StdDev) / (margin of error)2

Example
• Here is an example of how the math works assuming you chose a 95%
confidence level, .5 standard deviation, and a margin of error
(confidence interval) of +/- 5%.
• ((1.96)2 x .5(.5)) / (.05)2
(3.8416 x .25) / .0025
.9604 / .0025
384.16
• 385 respondents are needed