SAMPLING TECHNIQUES

Satya Shrestha
Lecturer
KUSMS
TERMINOLOGIES (1/8)
Population:
 Refers to the entire group of individuals or items of
interest having some common characteristics about which
you want to draw conclusions.

 On the basis of countability, population can be classified

into:
 Finite Population
 Infinite 2
Population
 TERMINOLOGIES (2/8)
Finite population: Individuals of the population are countable. Eg.

 Number of NNC registered nurses

 Number of students in a certain college etc.

Infinite Population: Individuals of the population are uncountable.

Eg.

 Number of bacteria in the pond

 Number of fishes in the sea etc..

3
TERMINOLOGIES (3/8)
 Target Population: The entire population in which the
researchers are interested and to which they would like
to generalize the research findings.

 Accessible Population: The aggregate of cases that

conform to designated inclusion or exclusion criteria and
that are accessible as subjects of the study.
TERMINOLOGIES (4/8)
 Sample population: All those subjects who are actually
being participated in the research study, representing the
target population.
TERMINOLOGIES (5/8)

•Sample:
•Sample is a subset or portion of the total population
selected to participate in a research study.
•Eg: few drops of blood sample is sufficient to check
infections in the whole body.

6
TERMINOLOGIES (6/8)
Sampling:
 Sampling is the process of selecting representative units
from an entire population of a study.

 It is the selection of some part of the whole on the basis of

which judgment or inferences about the aggregate is made.
TERMINOLOGIES (7/8)

 Sampling unit: Each member of a population, selected

as a sample is called sampling unit.

 Sampling frame: It is the list of all elements or subjects

in the population from which the sample is drawn. Eg:

phone book, immunization records, hospital records etc.

TERMINOLOGIES (8/8)

 Parameter: The o Statistic: The

summary value of the summary value of the

variable for a population. variable for a sample.

 µ - mean  - mean

 σ - standard deviation  s – standard deviation

 P- proportion  p - proportion
PURPOSE OF SAMPLING (1/2)
1. Economical: Sampling helps the researchers to save a
lot of time, money and resources for the study.

2. Improved quality of data: With the researcher handling

the information from only a part of the study population
(i.e. the sample), it is easier to maintain the quality of the
research work.
PURPOSE OF SAMPLING (2/2)
3. Quick study results: Dealing with the sample rather
than the entire population will save a lot of time for data
collection and analysis. Hence, the research can generate
the study results faster.

4. Precision and accuracy of data: maintaining the

precision of voluminous data is not an easy task. It is
always easier to establish better rapport with a sample
and thus, to collect more accurate data.
1. IDENTIFYING AND DEFINING THE
TARGET POPULATION (1/1)
 Target population consists of the total group of people or
objects about which the researcher would generalize the
information. Hence, it is important first to identify and
define the target population.
2. DESCRIBING THE ACCESSIBLE
POPULATION AND ENSURING THE
SAMPLING FRAME (1/1)
 The researcher must establish a description about the
accessible population, which is readily available for the
research.

 Then a sampling frame has to be identified to select the

samples.
3. SPECIFYING THE SAMPLING UNIT (1/1)
 Next, the researcher must establish the specified
inclusion and exclusion criteria to select a particular
sampling unit from the sampling frame.
4. SPECIFYING SAMPLE SELECTION
METHODS (1/1)
 The researcher has to specify whether the sample will be
drawn from the population by using probability or non-
probability sampling methods.

 Selection of the sampling method depends on several

factors such as type of population, phenomenon under
study and availability of resources and knowledge of the
researcher.
5. DETERMINING THE SAMPLE SIZE (1/1)
 It is important to determine the sample size for planning
the sampling process.
6. SPECIFYING THE SAMPLING PLAN (1/1)
 Before starting the actual selection of samples, the
researcher has to specify the sampling plan in detail so as
to avoid any undue problems.
7. SELECTING A DESIRED SAMPLE (1/1)
 Finally, the researcher draws a representative sample
from the accessible population, which requires the
implementation of the sampling plan.
CRITERIA OF A GOOD SAMPLING (1/1)
1. The sample to be obtained should be representative of the
population.
2. The sample size should be adequate.
3. The sampling procedures should be practical and feasible.
4. The sampling design should be economic and efficient.
5. A good sample should be free from sampling bias and
errors.
6. A sample is said to be good if once a subject is selected for
the study, it is neither replaced nor is it incomplete in any
aspect of researchers’ interest.
ADVANTAGES OF SAMPLING (1/1)

1. It entails lower cost.

2. It takes shorter time.
3. Better quality of information can be collected.
4. More comprehensive data can be collected.
FACTORS INFLUENCING SAMPLING
PROCESS (1/1)
 Nature of the researcher  Circumstances
 Inexperienced investigator  Lack of time
 Lack of interest  Large geographic area
 Lack of honesty  Lack of cooperation
 Intensive workload  Natural calamities
 Inadequate supervision

 Nature of the sample

 Inappropriate sampling
technique
 Sample size
 Defective sampling frame
TYPES OF SAMPLING

Sampling

Non-probability
Probability Sampling
Sampling
PROBABILITY SAMPLING
 Probability sampling is the method in which each
member of the population has a known chance
(probability) of being selected.

 Probability samples are more accurate than non-

probability samples .

 They remove conscious and unconscious sampling

bias.

24
TYPES OF PROBABILITY SAMPLING
1) Simple random sampling
2) Systematic sampling
3) Stratified sampling
4) Multistage sampling
5) Cluster sampling
SIMPLE RANDOM SAMPLING (1/5)
 Each element of the frame thus has an equal probability
of selection.

 Applicable when
 Population is small, homogeneous & readily available.
 A list of all the members of the accessible population is
available.

26
27
SIMPLE RANDOM SAMPLING (2/5)
 The first step of simple random sampling method is to
identify the accessible population and obtain a list of all
the elements/members of the population. Then a sample
is drawn from the sampling frame by one of the
following methods:
• Lottery method
• Table of random number method
• Computer generated table of random numbers

28
Lottery method
RANDOM NUMBER TABLES:

30
Computer generated table of random numbers
SIMPLE RANDOM SAMPLING (3/5)
 Sampling schemes may be without replacement ('WOR' -
no element can be selected more than once in the same
sample) or with replacement ('WR' - an element may
appear multiple times in the one sample).

 For example, if we catch fish, measure them, and

immediately return them to the water before continuing
with the sample, this is a WR design, because we might
end up catching and measuring the same fish more than
once. However, if we do not return the fish to the water
(e.g. if we eat the fish), this becomes a WOR design.
32
SIMPLE RANDOM SAMPLING (4/5)
Advantages:
1. It is a scientific method; so there is less personal bias
since every member is given equal chance of being
selected.

2. Reliable and accurate result is obtained at the limited

cost, time and labor.

3. It is more ideal for statistical purposes.

4. It requires minimum knowledge about the population in

advance. 33
SIMPLE RANDOM SAMPLING (5/5)
Disadvantages
1. If sampling frame is large, this method is impracticable.

2. It needs complete updated list of units of population,

which may not be available.

3. Numbering the population is too much tedious and

time consuming.

4. Expensive to conduct.

34
SYSTEMATIC SAMPLING (1/3)
 Systematic sampling relies on arranging the target
population according to some ordering scheme and then
selecting elements at regular intervals through that
ordered list.

 Systematic sampling involves a random start and then

proceeds with the selection of every kth element from
then onwards.

 In this case,
k=population size (N)/sample size (n)

35
SYSTEMATIC SAMPLING (2/3)
 It is important that the starting point is not
automatically the first in the list, but is instead randomly
chosen from within the first to the kth element in the
list.

 A list of subjects in the target population is obtained,

then the first subject is chosen at random between 1
and K. Then every kth subject is chosen from the
sampling frame.

 This type of sampling method can be done for home

visiting in community field.
37
SYSTEMATIC SAMPLING (3/3)
ADVANTAGES:
 Sample easy to select

 Less cumbersome, time-consuming and cheaper than

simple random sampling
 Sample evenly spread over entire reference population

DISADVANTAGES:
 Sample may be biased if hidden periodicity in
population coincides with that of selection.
 If first sample is not randomly selected, it becomes a
non-random sampling 38
STRATIFIED SAMPLING (1/8)
 This method is followed when the population is not
homogenous.

 In this random sampling technique, the whole population

N is first divided into non-overlapping(mutually
exclusive) subpopulations called strata(sub-group) and
then units are selected randomly from each stratum.

 Stratification is the process of grouping members of the

population into relatively homogenous subgroups before
sampling.
39
STRATIFIED SAMPLING (2/8)
 The subgroups are based on some predetermined criteria
such as geographic location, size, or demographic
characteristics(age, sex, social status, occupation etc).
 Then, random or systematic sampling is applied within
each stratum

40
STRATIFIED SAMPLING (3/8)

 Advantages
• Minimal knowledge of population needed
• Easy to analyze data

 Disadvantages
• High cost; low frequency of use
• Requires sampling frame
• Does not use researchers’ expertise

• Not useful when there are no homogenous subgroups.

43
STRATIFIED SAMPLING (4/8)
Types of stratified sampling
1. Proportionate Stratified Random Sampling

2. Disproportionate Stratified Random Sampling

44
STRATIFIED SAMPLING (5/8)
1.Proportionate Stratified Random Sampling:
 The sample size of each stratum in this technique is
proportionate to the population size of the stratum when
viewed against the entire population. This means that the
each stratum has the same sampling fraction.

45
STRATIFIED SAMPLING (6/8)
 For example, you have 3 strata with 100, 200 and 300
population sizes respectively. And the researcher chose a
sampling fraction of ½. Then, the researcher must
randomly sample 50, 100 and 150 subjects from each
stratum respectively.

Strata A B C

Population Size 100 200 300

Sampling Fraction 1/2 1/2 1/2
Final sample size 50 100 150

46
STRATIFIED SAMPLING (7/8)
 The important thing to remember in this technique is to
use the same sampling fraction for each stratum
regardless of the differences in population size of the
strata.

 It is much like assembling a smaller population that is

specific to the relative proportions of the subgroups
within the population

47
STRATIFIED SAMPLING (8/8)

2.Disproportionate Stratified Random Sampling

 The only difference between proportionate and
disproportionate stratified random sampling is their
sampling fractions. With disproportionate sampling, the
different strata have different sampling fractions.

 The precision of this design is highly dependent on the

sampling fraction allocation of the researcher. If the
researcher commits mistakes in allotting sampling
fractions, a stratum may either be overrepresented or
underrepresented which will result in skewed results.
48
CLUSTER SAMPLING (1/3)
 Cluster sampling is an example of 'two-stage sampling' .

 First stage a sample of areas is chosen;

 Second stage a sample of respondents within those

areas is selected.

 Population divided into clusters of homogeneous units,

usually based on geographical contiguity.

49
CLUSTER SAMPLING (2/3)
 Sampling units are groups rather than individuals.

 A sample of such clusters is then selected.

 All units from the selected clusters are studied.


Often used to evaluate vaccination coverage in EPI

50
CLUSTER SAMPLING (3/3)
 Advantages :
 Applicable where no complete list of units is available
 This can reduce travel and other administrative costs.

 Disadvantages:
 sampling error is higher for a simple random sample of
same size.
 Cluster may not be representative of whole population

52
DIFFERENCE BETWEEN CLUSTER AND
STRATA (1/1)
 Although strata and clusters are both non-overlapping
subsets of the population, they differ in several ways.

 All strata are represented in the sample; but only a subset

of clusters are in the sample.

 With stratified sampling, the best survey results occur

when elements within strata are internally homogeneous.
However, with cluster sampling, the best results occur
when elements within clusters are internally
heterogeneous

 In stratified sampling, a random sample is drawn from

each of the strata, whereas in cluster sampling only the 53

selected clusters are studied.

MULTISTAGE SAMPLING
 Complex form of cluster sampling in which two or more
levels of units are embedded one in the other.

 First stage, random number of districts chosen in all

states.

 Followed by random number of talukas, villages.

 Then third stage units will be houses.

 All ultimate units (houses, for instance) selected at last

step are surveyed.
54
NON- PROBABILITY SAMPLING (1/3)
Non-probability sampling is a sampling
technique where the samples are gathered in a
process that does not give all the individuals in
the population equal chances of being selected.

55
NON- PROBABILITY SAMPLING (2/3)
When to Use Non-Probability Sampling
 This type of sampling can be used when demonstrating
that a particular trait exists in the population.

 It can also be used when the researcher aims to do a 

qualitative, pilot or exploratory study.

 It can be used when randomization is impossible like

when the population is almost limitless.

56
NON- PROBABILITY SAMPLING (3/3)
 It can be used when the research does not aim to
generate results that will be used to
create generalizations pertaining to the entire population.

 It is also useful when the researcher has limited budget,

time and workforce.

57
TYPES OF NON PROBABILITY
SAMPLING
1) Convenience Sampling
2) Purposive sampling or judgmental sampling
3) Quota sampling
4) Snow ball sampling
1. CONVENIENCE SAMPLING (1/4)
 Sometimes called Haphazard or accidental sampling.
 The researcher using such a sample cannot scientifically
make generalizations about the total population.
 It involves the sample being drawn from that part of the
population which is close to hand.
 A sample that can be taken easily without random
selection.eg. People walking by on the street
59
CONVENIENCE SAMPLING (2/4)

 Examples
 A reporter can take the views of people in certain issues
at street.
 In clinical practice, we might use clients who are
available to us as our sample.

60
 CONVENIENCE SAMPLING (3/4)

 Advantages
 Useful for making pilot studies, particularly for testing
the research instruments.
 Cheap and quick

62
 CONVENIENCE SAMPLING (4/4)

 Disadvantages:
 Difficult to generalize the population from sampling results.
 Result obtained by this method are generally biased and
unsatisfactory, since the result obtained can hardly be
representative of the population parameters.

63
PURPOSIVE SAMPLING (1/4)

 Also called as purposive or deliberate sampling.

 The choice of item in the sample depends entirely upon
the judgment of the investigator.
 It is very useful for situations where we need to reach a
targeted sample quickly.

 It is used primarily when there are limited number of

people that have expertise in the area being researched.

 It is always subject to some kind of bias. 64

PURPOSIVE SAMPLING (2/4)
 It is appropriated when:
i. The researcher uses it to select unique cases that are
especially informative, for e.g. Fee structure of the
patient.

ii. It is used to select members who are difficult to reach,

e.g: study of prostitutes.

iii. To identify particular type of cases.

65
 PURPOSIVE SAMPLING (3/4)
 Advantages

1. This method is useful when a sample of small size is to be

selected.

2. If the investigator is wise enough to choose only those items

which are more representative, the results obtained may be fairly
accurate and reliable.

3. When quick decisions are needed, no other method is more

67

appropriate than judgment sampling is used

 PURPOSIVE SAMPLING (4/4)
Disadvantages

1. The sampling may be biased due to personal beliefs of the

investigator.

2. The samples can be selected in a wrong way if the

investigator wants to obtain certain results to meet only
his/her objective.

3. Though simple, it is not scientific and it is not in general

use. 68
QUOTA SAMPLING (1/4)
 In quota sampling, a population is first segmented into
mutually exclusive sub-groups i.e. quota just as in
stratified sampling.

 Then judgment or convenience sampling is used to select

the subjects or units from each segment based on a
specified proportion.

69
 QUOTA SAMPLING (2/4)
 Hence it is a form of convenient sampling involving
selection of quota(group of accessible sampling units) by
traits such as sex, age, social classes.

 Quota sampling is therefore a method of stratified

sampling in which the selection within the strata is non
random.

70
 QUOTA SAMPLING (3/4)
 Advantages
 Easy to conduct and easy to get the information in short period
of time.
 The cost and time involved in getting information from the
sample will be relatively less for quota sampling.
 It is very useful in public opinion studies, election forecast polls,
as there is not sufficient time to adopt a probability sampling
scheme. 72
QUOTA SAMPLING (4/4)

 Disadvantages
 Does not represent the population.
 Selection of the sample is non random so the samples
may be biased.

73
SNOWBALL SAMPLING (1/5)

 It is also called network sampling or chain sampling.

 Early sample members are asked to identify and refer

other people who meet the eligibility criteria of the

study.

74
SNOWBALL SAMPLING (2/5)
Steps

 Make contact with one or two cases in the population.

 Ask these cases to identify further cases

 Ask these new cases to identify further new cases

 Stop when either no new cases are given or the sample is

as large as is manageable

 Sample is very likely to be homogeneous. 75

 SNOWBALL SAMPLING (3/5)
 In sensitive issues in which people hesitate to expose,
researcher uses this type of sampling design.
 For example: people living with HIV/AIDS, sex workers,
intravenous drug users, prostitution study etc..

76
SNOWBALL SAMPLING

77
SNOWBALL SAMPLING (4/5)
 Advantages:
 It reduces cost and time.

 When utilizing other sampling methods, the chain

referral process permits the researcher to reach
populations that are difficult to sample.

 This sampling technique requires lesser workforce and

little planning as compared to other methods.
SNOWBALL SAMPLING (5/5)
 Disadvantages:
 The researcher has less control over the sampling
method.

 Sample representativeness is not guaranteed.

 There is greater chance of sampling bias because initial

subjects generally nominate people which they are aware
of, who share similar characteristics.
 A QUICK REVIEW
 In a class of 18 students, 6 are chosen for an assignment.

Example Sampling type

Pull 6 names out of a bowl

Select every 3rd student

Divide the class into 2 equal groups according to age

group, randomly select 3 from each group

Divide the class into 6 groups of 3 students each.

Randomly choose 2 groups.

Take the 6 students closest to the teacher

 A QUICK REVIEW
 In a class of 18 students, 6 are chosen for an assignment.

Example Sampling type

Pull 6 names out of a bowl Simple Random

Select every 3rd student

Divide the class into 2 equal groups according to age

group, randomly select 3 from each group

Divide the class into 6 groups of 3 students each.

Randomly choose 2 groups.

Take the 6 students closest to the teacher

 A QUICK REVIEW
 In a class of 18 students, 6 are chosen for an assignment.

Example Sampling type

Pull 6 names out of a bowl Simple Random

Select every 3rd student Systematic

Divide the class into 2 equal groups according to age

group, randomly select 3 from each group

Divide the class into 6 groups of 3 students each.

Randomly choose 2 groups.

Take the 6 students closest to the teacher

 A QUICK REVIEW
 In a class of 18 students, 6 are chosen for an assignment.

Example Sampling type

Pull 6 names out of a bowl Simple Random

Select every 3rd student Systematic

Divide the class into 2 equal groups according to age Stratified

group, randomly select 3 from each group

Divide the class into 6 groups of 3 students each.

Randomly choose 2 groups.

Take the 6 students closest to the teacher

 A QUICK REVIEW
 In a class of 18 students, 6 are chosen for an assignment.

Example Sampling type

Pull 6 names out of a bowl Simple Random

Select every 3rd student Systematic

Divide the class into 2 equal groups according to age Stratified

group, randomly select 3 from each group

Divide the class into 6 groups of 3 students each. Cluster

Randomly choose 2 groups.

Take the 6 students closest to the teacher

 A QUICK REVIEW
 In a class of 18 students, 6 are chosen for an assignment.

Example Sampling type

Pull 6 names out of a bowl Simple Random

Select every 3rd student Systematic

Divide the class into 2 equal groups according to age Stratified

group, randomly select 3 from each group

Divide the class into 6 groups of 3 students each. Cluster

Randomly choose 2 groups.

Take the 6 students closest to the teacher Convenience

THANK YOU