19-Sep-20

Topic-8

I welcome all of you to today’s session on

sampling.

Out line of the session

 Sampling
Definition, Necessity, Advantages
Understanding Relevant Concepts
Types with Advantages and Disadvantages
Choosing Sampling Technique
Sampling in Qualitative Research
Determining Sample Size

1
 19-Sep-20

Sampling-Definition
Sampling is the technic/technics of selecting
sample/samples from a population. and

•Sample is a relatively small subset from a

population.

•I will explain population as a concept a bit later.

Necessity of sampling
•The population of interest is usually too large to
attempt to survey all of its members.

•A carefully chosen sample can be used to

represent the population.

•The sample reflects the characteristics of the

population from which it is drawn.

2
 19-Sep-20

Advantages and Utilities

•Helps to identify various representative units

•Saves a lot of time, energy and money without sacrificing quality

•Unlimited data cannot be manage without this technique

•Offers high degree of accuracy

•Saves destruction of units

•Achieves greater response rate

•Easy to supervise

Concepts related to understanding of sampling

• Population
• Target Population
• Elements
• Sampling Units
• Sampling Frame
• Sample
• Parameter
• Statistic

3
 19-Sep-20

Continued
•Population is a group that includes all the cases
(individuals, objects, or groups) in which the
researcher is interested.

•A population is a collection of elements about

which we wish to make an inference.

•It is critical to the success of the research project

to clearly define the target population.

Continued
• Rely on logic and judgment.

• The population should be defined in connection with the

objectives of the study.

• An element is an object on which a measurement is taken.

• Sampling units are non-overlapping collections of

elements from the population that cover the entire
population.

4
 19-Sep-20

Continued
• A sampling frame is a list of sampling units.

• A sample is a collection of sampling units drawn

from a sampling frame.

• Parameter: numerical characteristic of a population

• Statistic: numerical characteristic of a sample

Types of sampling
• A) Probability Sampling
1. Random, 2. Systematic, 3. Stratified
4. Cluster, 5. Multi-stage
6. Probability Proportionate to Size (PPS)

• B) Non-probability Sampling
1. Convenience, 2. Judgment/Purposive, 3. Quota
4. Snowball, 5. Extreme, 6. Self-selected

5
 19-Sep-20

A) Probability Sampling
• Probability Samples: each member of the population
has a known non-zero probability of being selected
• Methods include random , systematic, stratified,
cluster, multi-stage, and PPS sampling.

• Used when researchers want precise, statistical

descriptions of large population.

• A sample of individuals from a population must contain

the same variations that exist in the population.

A(1) Random Sampling

Random sampling is the purest form of probability sampling.

•Each member of the population has an equal and known

chance of being selected.

•When there are very large population, it is often ‘difficult’ to

identify every member of the population, so the pool of
available subjects becomes biased.

•You can use software, such as minitab to generate random

numbers or to draw directly from the columns

6
 19-Sep-20

Continued
Advantages/strengths:
Simple theory and easiest to understand.

Can be used in conjunction with other methods in

probability sampling

Requires minimum level of knowledge on

population

Degree of sampling error is low

Continued
Disadvantages/limitations/challenges:
Where population variability is very sporadic and irregular,
cannot be used

Assigning numbers to all sampling units in many cases is

impractical

Data collection by geographical unit will be very difficult

No guarantee of typical sampling selection

7
 19-Sep-20

A(2) Systematic Sampling

• Systematic sampling is often used instead of random sampling.
It is also called an Nth name selection technique.

• After the required sample size has been calculated, every Nth
record is selected from a list of population members.

• As long as the list does not contain any hidden order, this
sampling method is as good as the random sampling method.

• Its only advantage over the random sampling technique is

simplicity (and possibly cost effectiveness).

Continued
Advantages/strengths:
Quicker than simple random sampling

Possible to select a sample without sampling frame

Especially appropriate if (wants to spread sample over

population, draw from a chronological file)

Mistakes in drawing elements are relatively unimportant

8
 19-Sep-20

Continued
Disadvantages/limitations/challenges:
If the population is not random, it is not suitable to use

If there exists periodic regularities in list, may get

redundant similar samples

Over representation and underrepresentation of groups

can occur

Systematically ignores some elements

A(3) Stratified Sampling

Stratified sampling is commonly used probability method that is superior
to random sampling because it reduces sampling error.

A stratum is a subset of the population that share at least one common

characteristic; such as male and female.

Identify relevant strata and their actual representation in the population.

Random sampling is then used to select a sufficient number of subjects

from each stratum.

Stratified sampling is often used when one or more of the stratum/strata in

the population have a low incidence relative to the other strata.

9
 19-Sep-20

Continued
Advantages/strengths:
More efficient than simple random sampling.

Allows for more comprehensive data analysis, as

considers each stratum

Convenient to administer

Can be used for comparing sub-groups

Continued
Disadvantages/limitations/challenges:
Additional prior info. of population is required

A separate frame is needed for each stratum

Needs a larger sample size

Classification errors can increase

10
 19-Sep-20

A(4) Cluster Sampling

Cluster Sample: a probability sample in which each
sampling unit is a collection of elements.

Effective under the following conditions:

A good sampling frame is not available or costly, while a

frame listing clusters is easily obtained

The cost of obtaining observations increases as the

distance separating the elements increases

Continued
Examples of clusters:
City blocks – political or geographical

Housing units – college students

Hospitals – illnesses

Automobile – set of four tires

11
 19-Sep-20

Continued
Advantages/strengths:
No needs to construct a list of elements

If a list of elements is available, sampling is cheaper

Easier to administer in large population and/or large

geographical area of study

Respondents can be easily substituted

Continued
Disadvantages/limitations/challenges:
Reliability is lower than simple or stratified sampling

Dissimilarity of cluster size can lead to sampling error

Lacks representation

Overleaping of respondents in more than one clusters is

difficult to check

12
 19-Sep-20

A(5) Multi-stage Sampling

• Used when it's not possible or practical to create a list
of all the elements that compose the target population.

• Involves repetition of two basic steps: listing and

sampling.

• Highly efficient but less accurate.

• Combination of random and stratified samplings.

Continued
Advantages/strengths:
More representative

More efficient and flexible than single stage

sampling

Sampling frame is required for selected units only,

not for all

13
 19-Sep-20

Continued
Disadvantages/limitations/challenges:
Complexity in theory

Estimation procedures are difficult

Takes much time at the planning stage

A(6) Probability Proportionate to Size (PPS) Sampling

•When samples from different sized subgroups are
used and sampling is taken with the same
probability, the chances of selecting a member
from a large group are less than selecting a
member from a smaller group.

•This is known as probability proportional to size

(PPS).

14
 19-Sep-20

Continued
• For example, if one sample had 20,000 members, the
probability of a member being selected would be
1/20000 or .005 percent. If another sample had 10,000
members, the chance of a member being selected
would be 1/10000 or .01 percent.

• Sophisticated form of cluster sampling.

• Used in many large scale survey sampling projects.

Continued
Advantages/strengths:
Helps to avoid underrepresenting and
overrepresenting one subgroup and yields more
accurate results.

Disadvantages/limitations/challenges:
Time consuming and sophisticated to use

15
 19-Sep-20

B) Non-probability Sampling
Technique in which samples are selected in a way
that is not suggested by probability theory.

Members are selected from the population in some

nonrandom manner

Methods include convenience, judgment/purposive,

quota, snowball, self-selected and extreme sampling

Continued
Reliance on available subjects:

Only justified if less risky sampling methods are

not possible.

Researchers must exercise caution in generalizing

from their data when any of these methods is
used.

16
 19-Sep-20

B(1) Convenience Sampling

• Convenience sampling is used in exploratory research
where the researcher is interested in getting an
inexpensive approximation.

• The sample is selected because they are convenient.

• Often used during preliminary research efforts to get an

estimate without incurring the cost or time required to
select a random sample.

Continued
•Advantages/strengths:

It is quick and economical

•Disadvantages/limitations/challenges:

Can be treat as biased sample

17
 19-Sep-20

B(2) Judgment/Purposive Sampling

Judgment sampling is a common non-probability method.
Another name is purposive sampling.

The sample is selected based upon judgment/particular

purpose.

It is an extension of convenience sampling

When using this method, researcher must be confident that the

chosen sample is truly representative of the entire population.

Continued
Advantages/strengths:

It is purposeful, researcher can easily reach

Disadvantages/limitations/challenges:

It is deliberate choice of researcher and based on

prior judgment

18
 19-Sep-20

B(3) Quota Sampling

Quota sampling is the non-probability equivalent of
stratified sampling.

First identify the strata and their proportions as they

are represented in the population

Then convenience or judgment sampling is used to

select the required number of subjects from each
stratum.

Continued
Advantages/strengths:
Less costly than other non-probability techniques
Does not require sampling frame
Can be completed in a short period of time

Disadvantages/limitations/challenges:
Not representative
Has researcher’s bias
Sometimes fulfilling quota is not possible

19
 19-Sep-20

B(4) Snowball Sampling

Snowball sampling is a special non-probability method used
when the desired sample characteristic is rare.

It may be extremely difficult or cost prohibitive to locate

respondents in these situations.

This technique relies on referrals from initial subject to generate

additional subjects.

It lowers search costs; but introduces bias because the technique

itself reduces likelihood that sample will represent a good cross
section from the population.

Continued
Advantages/strengths:
Can use when target population is unknown
When difficult to approach respondents (lets say drug
addicted people)
Reduced sample size and costs

Disadvantages/limitations/challenges:
Bias can enter
Known and unknow people can weaken research
position

20
 19-Sep-20

B(5) Extreme Sampling

•It is the selection of unusual cases that fall
outside general patterns

•These extreme cases may be very interesting

theoretically even though they are by definition
not representative of the population.

Continued
Advantages/strengths:
Can draw attention of policymakers for an extreme case
Almost opt out topic can be researched through this
method

Disadvantages/limitations/challenges:
Will not be representative
Requires serious justification supported by strong
materialistic evidences

21
 19-Sep-20

B(6) Self-selected Sampling

•This method is applicable in such case where the
sampling area and sample unit are not fixed.

•Samples offer themselves for selection

Continued
Advantages/strengths:
Convenient for the researcher
Where may be you cannot use any other method in
that case only you can use

Disadvantages/limitations/challenges:
Still can have chance of bias
Acceptability can be low

22
 19-Sep-20

Choosing sampling technique

There is no single principal to choose a particular sampling technique.
It depends on a variety of issues. Some significant issues are as follows

1. On the nature of the research problem and research objectives.

In many instances these two automatically eliminate some techniques.
Hence, researcher(s) then comparatively considers the advantages and
disadvantages of the remaining techniques and try to choose one or a
combination.

Continued
2. Mostly on availability of resources and materials

3. On certain statistical and administrative considerations

4. On duration and timing of research

5. On desired level of precision (correctness) of results

6. What sampling frame or list is available?

23
 19-Sep-20

Continued
7. How big a sample is desirable?

8. What sampling procedure is to be used?

Depending on the above criteria and consideration, the

researcher can choose one single technic or a suitable
combination of some techniques for the research.

Sampling in qualitative research

•Researcher of the qualitative research also use sampling.

•In some cases, it is clear requirement, and in some other cases it is not
that straightforward.

•Sampling in qualitative research looks like as follows:

1. Not towards large number of respondents rather towards typical

cases;

2. Not towards fixed samples but towards flexible samples in size and
types of subjects

24
 19-Sep-20

Continued
3. Not towards random sampling but towards purposive sampling;

4. Not towards representativeness but towards suitability;

5. Not towards strictly defined size but towards convenient and

suitable size;

6. Not towards mechanical sampling but towards theoretical sampling;

7. Not even towards choosing a sample before the study, but often
while the study is in progress.

Determining sample size

•We will discuss the following issues under this sub-topic;

1. Consideration in sample size

2. Mathematical formulas for sample size

3. Determining sample size in given estimated proportions

4. Determining sample size through tables

25
 19-Sep-20

Consideration in sample size

• There are a good of issues need to be taken care of while
determining sample size. These are as follows

1. The size of the population: very large, large or small

2. Nature of population: homogeneous or heterogenous

3. Purpose of study: descriptive, exploratory or explanatory

Continued
4. Nature of the study: Qualitative or quantitative

5. Accessibility of the respondents: Difficult or easy to get

access at a time

6. Availability of resources: constraint or sufficient

7. Variability of respondents required: similar or of various

groups

26
 19-Sep-20

Continued
8. Desired accuracy or confidence level: High accuracy large sample,

9. Sampling error or accepted risk level: representativeness & error

are proportionately related

10. Stratification of samples required: More strata requires more

samples

Mathematical formulas for sample size

• 2 available formulas

1) 1st formula by Taro Yamne (1970:886-87)

n= N/1+N(e)2

Here,
n = sample size
N = Population
e = value of error against confidence level

27
 19-Sep-20

Continued
Calculation

Suppose total population is 500, and confidence level is 95%.

Hence, N = 500, e = (100-95)/100 = .05

n = 500/1 + 500(.05)2
= 500/1+500(.0025)
= 500/1+1.25
= 500/2.25
= 222

Continued
2) 2nd formula by Fink and Kosecoff (1995:62)
n = (z/e)2 (p) (1-p)
Here,
n = sample size
Z = standard score corresponding to a given confidence level
e = value of error against confidence level
p = estimated percentage to population

28
 19-Sep-20

Continued
Calculation
Suppose percentage of population is 25%, and confidence level is
90%.
Hence, z = 1.96 (given from the table), e = (100-90)/100 = 0.10, p
= 0.25

n = (1.96/0.10)2 x (0.25) x (1-0.25)

= (19.6)2 x (0.25) x (0.75)
= 384.16 x 0.1875
= 72.03
= 72

Determining sample size when estimated proportions are given

Example: suppose, the estimate given by other studies on drug
abuse among college students is that 10% students take drugs, then
for a new study, the sample size would be:

n = pqz2/E2

Here,
p = 10
q = 100 - p = 100 - 10 = 90
z = 1.96 (at 95% confidence level)
E = 100 - 95 = 5

29
 19-Sep-20

Continued
Calculation
n = 10 (90) (1.96)2/52
= (900 x 3.84)/25
= 3457.44/25
= 138.29
= 138
If we change the value of p, q, z, and E, the sample size can be
flexibly determine. However, E is inversely related with sampling
size.

Determining sample size through tables

If you have a given population, using the table of Eckhard (Social
Research Methods, 1978:400) you can find the value of your
sample size. This table allows you to find a combination of p, E
and Z.

Here,
p = population estimate
E = error
Z = confidence level

30
 19-Sep-20

Continued; a part of the table

Population 1% 2% 3% 4% 5% 10%
10,000 5,000 2,000 1,000 588 385 99
25,000 7,143 2,273 1,064 610 394 100
50,000 8,333 2,381 1,087 617 397 100
1,00,000 9,091 2,439 1,099 621 398 100

Reference for this session

Ahuja, Ram, Research Methods, Rawat
Publications, Jaipur and New Delhi, Second
Reprint, 2005, PP 155-190.
&
Aminuzzaman, M. Salahuddin, Essentials of
Social Research, Osder Publications, Dhaka, 2011,
PP 100-110.

31
 19-Sep-20

Sample questions from this session

1) Define sampling. Identify and explain other relevant concepts
of sampling. What is the necessity of sampling in research?

2) What are the major identified advantages and utilities of

sampling?

3) What are the major forms of sampling. Discuss the differences

between probability and non-probability sampling.

4) Comparatively discuss the popular methods of probability

sampling.

Continued
5) Comparatively discuss the popular methods of non-probability
sampling.

6) What are major consideration of choosing a sampling technic for a

social research?

7) Portray the major features of sampling in qualitative research.

8) Identify and discuss the major consideration for fixing sample size.

9) How many techniques you know to fix sample size? Elaborate any
two techniques with example.

32
 19-Sep-20

Thanks 

33