SAMPLING TECHNIQUES

CHAPTER 4
SAMPLING TECHNIQUES

4.1 Sampling
• Researchers usually draw conclusions about large
groups by taking a sample.
• A Sample is a segment of the population selected
to represent the population as a whole.
• Ideally, the sample should be representative and
allow the researcher to make accurate estimates of
the thoughts and behavior of the larger population.
• Sampling: is the process of selecting a sample from
the population.
Rationale for Sampling
 In many cases complete coverage of population
is impossible.
 Studies based on sample require less time and
labor
 Sampling is relatively economical
 Sampling provides more detailed information
and high degree of accuracy
Designing the sample calls for three decisions:
1. Who will be surveyed? (The Sample)
• The researcher must determine what type of information is
needed and who is most likely to have it.
2. How many people will be surveyed? (Sample Size)
• Large samples give more reliable results than small samples.
However, it is not necessary to sample the entire target
population.
3. How should the sample be chosen? (Sampling)
• Sample members may be chosen at random from the entire
population (probability sample)
• The researcher might select people who are easier to obtain
information from (non-probability sample)
• The needs of the research project will determine which method
is most effective.
1. Assessment of urban informal settlement: the case of
Shashemenne town------abduljebar
2. Assessing the challenges of informal sectors on municipal
revenue generation: the case Shashemenne Town------
Gemechu
3. Assessing Challenges of peri-urban land compensation
payment: the case of shashemene town------daniel
4. Exploring challenges of urban infrastructure project
management: the case of Shashemenne, town---Josef
5. Assessing practice of liquid waste management: the case of
Shashemenne town------husien.
6. An assessment of socio-economic conditions of street-
children in Shashemenne town-----Adunga
7. Appraising challenges of urban structural plan
implementation: the case Shashemenne town-----Tekalign
In general, the following points must be considered while
determining sample size.
1. Homogeneity of the population:- the more homogeneous
the characteristics of the population, the smaller will be the
sample size. To check the homogeneity of the population we
can use:
– Simple pilot study
– Casual observation
2. Precision of statistical analysis:-this determines the
maximum error you commit, while conducting your work.
Your error should not be greater than 0.05. Increasing our
sample size will improve our precision.
3. The population Size: most of the time the sample size
depends on the total population size. Sample size under 30 is
not possible, unless it is case study.
4. The characteristics you plan to study:- important question
to be raised here is “can I get different data from different
samples?”
 In qualitative study usually small sample size is
recommendable, because you can cease collecting data, if
you get your objectives answered. This avoids redundancy.
 In general, Borg and Gall (1979:195) suggest that, as a
general rule, sample sizes should be large where:
– there are many variables
– only small differences or small relationships are expected or
predicted
– the sample will be broken down into subgroups
– the sample is heterogeneous in terms of the variables under
study
– reliable measures of the dependent variable are unavailable
3.2 Types of Sampling

• There are basically two types of sampling. These

are: probability and non probability sampling.
A. Probability Sampling
• It is sometimes called chance sampling, because
the members are not selected with personal
judgment of the researcher.
Importance of Probability Sampling
 It is highly reliable
 It has high degree of accuracy
 Its findings are highly generalizable
 Sampling Methods

Probability Sampling Non-probability Sampling

(Usually for Quantitative Methods) (Usually for Qualitative Methods)

Simple Random Sampling Convenience sampling

Systematic sampling Judgment/Purposive sampling

Stratified random sampling Quota sampling

Snowball sampling
Cluster sampling
Volunteer sampling
Multistage sampling
Spatial/Grid Sampling
• Probability sampling incorporate several techniques like:
Simple random sampling, Systematic sampling, Stratified
random sampling, Cluster (area) sampling, Stage sampling
1. Simple random sampling:
• In simple random sampling, each member of the population
under study has an equal chance of being selected and the
probability of a member of the population being selected is
unaffected by the selection of other members of the
population, i.e. each selection is entirely independent of the
next.
• The method involves selecting at random from a list of the
population (a sampling frame) the required number of
subjects for the sample.
• Every member of the population has a known and equal
chance of being selected.
• Randomization is a method and is done by
using a number of techniques as:
(a) Tossing a coin
(b) Throwing a dice
(c) Lottery method
(d) Blind folded method
(e) By using random table
 2. Systematic Sampling
• This method is a modified form of simple random
sampling. It involves selecting subjects from a
population list in a systematic rather than a random
fashion.
• For example, if from a population of, say, 2,000, a
sample of 100 is required, then every twentieth person
can be selected.
• The starting point for the selection is chosen at
random.
3. Stratified random sampling:
• Stratified sampling involves dividing the population
into homogenous groups, each group containing
subjects with similar characteristics.
• For example, group A might contain males and group
B, females.
• In order to obtain a sample representative of the
whole population in terms of sex, a random selection
of subjects from group A and group B must be taken.
• Population is divided into mutually exclusive groups
such as age groups and random samples are drawn
from each group.
Stratified sampling may be of three types:
1. Disproportionate stratified sampling
2. Proportionate stratified sampling
3. Optimum allocation stratified sampling
1. Disproportionate sampling: means that the size of
the sample in each unit is not proportionate to the
size of the unit but depends upon considerations
involving personal judgement and convenience.
 It is more effective for comparing strata which have
different error possibilities.
 It is less efficient for determining population
characteristics.
2. Proportionate sampling refers to the selection from
each sampling unit of a sample that is proportionate to
the size of the unit.
• Advantages of this procedure include
representativeness with respect to variables used as the
basis of classifying categories and increased chances of
being able to make comparisons between strata.
• Lack of information on proportion of the population in
each category and faulty classification may be listed as
disadvantages of this method.
3. Optimum allocation stratified sampling is
representative as well as comprehensive than other
stratified samples.
 It refers to selecting units from each stratum should be
in proportion to the corresponding stratum of the
population.
 See table below
4. Cluster (area)sampling:
• When the population is large and widely dispersed,
gathering a simple random sample poses administrative
problems.
• Suppose we want to survey students’ fitness levels in a
particularly large community.
• It would be completely impractical to select students and
spend an inordinate amount of time travelling about in order
to test them.
• By cluster sampling, the researcher can select a specific
number of schools and test all the students in those selected
schools, i.e. a geographically close cluster is sampled.
• The population is divided into mutually exclusive groups such
as blocks, and the researcher draws a sample of the group to
interview.
5. Stage sampling
• Stage sampling is an extension of cluster
sampling.
• It involves selecting the sample in stages that
is, taking samples from samples.
• Using the large community example in cluster
sampling, one type of stage sampling might be
to select a number of schools at random, and
from within each of these schools, select a
number of classes at random, and from within
those classes select a number of students.
 B. Non Probability Sampling
• This is sometimes called purposive sampling.
• All the members of the population have no equal probability
of selection.
• This is useful for ethical and historical research, when we need
to conduct qualitative research.
• It does not involve random selection.

Importance of non-probability sampling

• Useful when the researcher has no access to an entire
population
• Useful when the generalizability of the study is not important
• Useful when there is one or limited size of sites are available
• Non-probability sampling includes different
techniques of sample drawings like:
convenience sampling, judgment sampling,
quota sampling, snowball sampling and
dimensional sampling.
1. Convenience sampling
• Convenience sampling—or as it is sometimes
called, accidental or opportunity sampling—
involves choosing the nearest individuals to
serve as respondents and continuing that
process until the required sample size has been
obtained.
• The researcher selects the easiest population
members from which to obtain information.
2. Judgment sampling
• In purposive sampling, researchers handpick the cases to be included
in the sample on the basis of their judgment of their typicality.
• In this way, they build up a sample that is satisfactory to their specific
needs.
• As its name suggests, the sample has been chosen for a specific
purpose, for example:

(a) a group of experts and senior managers of town is chosen as the

research is studying the incidence of corruption amongst senior
managers;

(b) a group of disaffected residents has been chosen because they

might indicate most distinctly the factors which contribute to resident’s
disaffection (they are ‘critical cases’ akin to ‘critical events’ discussed.
• The researcher uses his/her judgment to select
population members who are good prospects for
accurate information.
3. Quota sampling
• Quota sampling has been described as the non
probability equivalent of stratified sampling (Bailey,
1978).
• Like a stratified sample, a quota sample strives to
represent significant characteristics (strata) of the
wider population; unlike stratified sampling it sets out
to represent these in the proportions in which they
can be found in the wider population.
4. Snowball sampling
• In snowball sampling researchers identify a small
number of individuals who have the
characteristics in which they are interested.
• These people are then used as informants to
identify, or put the researchers in touch with,
others who qualify for inclusion and these, in
turn, identify yet others—hence the term
snowball sampling.
• This method is useful for sampling a population
where access is difficult, maybe because it is a
sensitive topic
5. Dimensional sampling
• One way of reducing the problem of sample
size in quota sampling is to opt for
dimensional sampling.
• Dimensional sampling is a further refinement
of quota sampling.
• It involves identifying various factors of
interest in a population and obtaining at least
one respondent of every combination of those
factors
6. Volunteer sampling
• This sampling technique consists of participants
becoming part of a study because they volunteer
when asked in response to an advert.
• It is a biased sampling.
 Characteristics of Good sampling

 It should be truly representative of the

population
 It should be resulted in small sampling error
 It should be economical in time, money and
energy
 Systematic bias should be controlled
effectively

Note: systematic bias results from wrong way of

selecting samples.
 Errors in Sampling

• Basically, two types of errors occur in sampling. These

are sampling and non sampling errors.
• Sampling Error: this arises due to drawing faulty
inference about the population based up on the result
of samples.
• In other words, it is the difference b/n that is obtained
by the sample study and entire population study.
• Example: if results of sample study indicate that about
25% of youths in Shashemenne are unemployed, and
the entire population study indicates 30%, then the
difference is considered as sampling error.
• Sampling error can be minimized if the sample size is
large in relative to the population size.
• Non-sampling Error: this error is introduced due to
technical faults during the processing of data.
• This error could also arise due to defective
methods of data collection, incomplete coverage
of population, inaccurate information provided by
the participants in the sample.
• Moreover, errors may also occur while editing,
tabulating and mathematical manipulation of data.
• In general, both the sampling and non-sampling
errors must be reduced to a minimum, in order to
get as representative as a sample of the population
as possible.
 Sample Size Determination
 Sampling is the process of obtaining information from a
subset (sample) of a larger group (population)
 The results for the sample are then used to make
estimates of the larger group
 Faster and cheaper than asking the entire population
Two keys
1. Selecting the right people
Have to be selected scientifically so that they are
representative of the population
2. Selecting the right number of the right people
To minimize sampling errors I.e. choosing the wrong
people by chance
Why use a sample?
 Cost
 Speed
 Accuracy
 Destruction of test units
Steps sampling
 Definition of target population
 Selection of a sampling frame (list)
 Probability or Nonprobability sampling
 sampling Unit
 Error
– Random sampling error (chance fluctuations)
 Non-sampling error (design errors)
 Target Population
• Who has the information/data you need?
• How do you define your target population?
- Geography
- Demographics
- Use
- Awareness
Operational Definition
• A definition that gives meaning to a concept
by specifying the activities necessary to
measure it.
Eg. Student, employee, user, area, major news
paper.
• What variables need further definition? (Items
per construct)
Sampling Frame
• List of elements
• Sampling Frame error
• Error that occurs when certain sample
elements are not listed or available and are
not represented in the sampling frame
Two alternative approaches for determining the size of the
sample.
• The first approach is “to specify the precision of estimation
desired and then to determine the sample size necessary to
insure it” and the second approach “uses Bayesian statistics to
weigh the cost of additional information against the expected
value of the additional information.”
• The first approach is capable of giving a mathematical solution,
and as such is a frequently used technique of determining ‘n’.
• The limitation of this technique is that it does not analyse the
cost of gathering information vis-à-vis the expected value of
information.
• The second approach is theoretically optimal, but it is seldom
used because of the difficulty involved in measuring the value of
information.
• Hence, we shall mainly concentrate here on the first approach.
• Selecting the right number of the right people
Three Issues
1. Financial
2. Managerial
3. Statistical
• Generally, the larger the sample size the
smaller the statistical error, but the greater
the cost, both financial and in terms of
managerial resources
• Balance between financial and statistical issues
1. What can I afford
2.Rule of thumb
• past experience
• historical precedence
• gut feeling
• some consideration of sample error
3.Make up of sub-groups (cells)
• What statistical inferences do you hope to make
between sub groups (rare to fall below 20 for a sub
group)
4. Statistical Methods
 Some of the commonly used sample size
determination techniques are
1. Using census: This infers use of all populations
as a sample
2. Using published table formula developed by
Krejcie and Morgan, 1970.
Using published table formula developed by Krejcie and Morgan, 1970
3. Using published table as developed by Glenn (1990)
Sample Size for 95% degree of precision (±5% tolerable error and 90% of precision
or ±10% tolerable error
By Standard of accuracy and acceptable confidence
level:
4. Using Krejcie and Morgan formula (1970)

5. Cochra nformula (1975):

• For populations that are large and the number is un known,
Cochran developed the Equation to yield are representative
sample for proportions.

• The value for Z is found in statistical tables which contain the area
under the normal curve (i.e.1.96 for 95% and 2.58 for the 99% of
degree of confidence at the normal curve)
6. Cochran correction formula

7. Solvin’s formula (1960) and Yamane simplified formula

(1967)
8. Applying comments of scholars:
• 10-50 % of total population (commented by Mugenda and
Mugenda, 2003)
• 10-30% of total population as commented by Best and Kahn,
200
• Gay (2005), suggests 20% -30% of the total population as a
sample size and are good to take and used as the sample size
for descriptive study.
9. Cost
• E.g.,if cost of surveying of one individual or unit is 30birr and if
the total available fund for survey is say 1800 birr, the sample
size then will be determined as,
• Sample size(n)=Total budget of survey/Cost of unit survey,
accordingly, the sample size will be 60units(1800/30=60units).
Sample Confidence
• “Probability” we can take results as “accurate
representation” of universe (i.e. that “sample
statistics” are generalisable to the real
“population parameters”)
• Typically a 95% probability (i.e. 19 times out of
20 we would expect results in this range)
Example:
• We can be 95% sure that, say, 65% of a target
market will name X in an unprompted recall test
plus or minus 4%
• We can be 95% sure (level of confidence) that, say, 65%
(predicted result) of a target market (of a given total
population) will name X in an unprompted recall test
plus or minus 4% (to a known margin of error)
95% confidence
• If we do the same test 20 times then it is statistically
probable that the results will fall between 61-69 %, (i.e.
65 +/ 4%) at least 19 times
• If we lower the probability then we lower the sample
error
• e.g.. at a 90% confidence level, result might be between
64% - 66% (a tighter range but we are less sure the
sample is representative of the real population)