Sampling Methods

Community Medicine Unit

International Medical School
Management and Science University
 Learning Objectives
1. Define population, samples, sampling frames
and sampling units
2. Describe the non-probabilistic sampling
methods
3. Describe the four probabilistic sampling
methods
4. Discuss advantages and disadvantages of
sampling methods
5. Conduct simple random sampling from a
given sampling frames
2
 Population vs Sample
Population: The
entire set of
individuals or
Population: other entities to
a set which includes all which study
measurements of interest findings are to be
generalized
to the researcher
(The collection of all responses,
measurements, or counts that are of
interest)

Sample:
A subset of the population
 Definition of sampling

Procedure by which some members

of a given population are selected as
representatives of the entire population
 Sampling

Population

Sampling to Inferential
make data statistics used
collection to generalize
manageable back

Sample
Why sampling?

Get information about large populations

 Less costs
 Less field time
 More accuracy i.e. Can Do A Better Job of Data
Collection
Target Population:
The population to be studied/ to which the
investigator wants to generalize his results
Sampling frame
List of all the sampling units from which sample
is drawn
Sampling Unit:
smallest unit from which sample can be
selected
Sampling scheme
Method of selecting sampling units from
sampling frame
 Sampling Methods

Sampling
Methods

Probability Non-
Sampling probability

Systematic Stratified Snowball

Convenience

Simple Quota
Cluster
Random
 Sampling Methods
• Probability Sampling
– Only sampling method that allows to draw valid
conclusions about population
– Subjects randomly chosen - every potential
subject has equal chance of being chosen

• Nonprobability Sampling
Subjects chosen by nonrandom methods; most
commonly used in clinical studies
 Strengths of Probability Sampling
(Random Sampling)
• Allows for estimation of sampling error.
• Sampling error is the difference between sample
statistics and population parameters.
• In other words, the variation between the average of
values found in the sample and those found in the
population
• Random sampling gives the greatest confidence in the
validity of findings in a study.
• Removes the possibility of bias in selection of subjects
• Ensures that each subject has a known probability of
being chosen
• Allows application of statistical theory
 Random selection vs Randomization
Random selection
• Randomly selecting subjects to ensure that they are
a representative sample of the general population or
of a specific group of patients (Random sampling)
 Non-probability Sampling
1. Convenience sampling
2. Quota sampling
3. Snowball sampling
 Convenience Sampling
• Also called haphazard or accidental sampling
• The sampling procedure of obtaining the
people or units that are most conveniently
available

Advantages Disadvantages
 Very low cost  Variability and bias cannot
 Extensively be measured or controlled
used/understood  Projecting data beyond
 No need for list of sample not justified.
population elements
 Quota Sampling
• Ensures that the various subgroups in a population are
represented on pertinent sample characteristics to the
exact extent that the investigators desire
• It should not be confused with stratified sampling.
Advantages Disadvantages
 moderate cost  Variability and bias
 Very extensively cannot be measured or
used/understood controlled (classification
 No need for list of of subjects0
population elements  Projecting data beyond
 Introduces some sample not justified.
elements of stratification
 Snowball Sampling
• A variety of procedures
• Initial respondents are selected by probability or non-
probability methods
• Additional respondents are obtained from information
(or referrals) provided by the initial respondents

Advantages Disadvantages
 low cost  Bias because sampling

 Useful in specific units not independent

circumstances  Projecting data beyond

 Useful for locating rare sample not justified.

populations
 Probability Sampling
1. Simple random Sampling
2. Systematic Sampling
3. Stratified Sampling
4. Cluster Sampling
 Four Probability Sampling Methods
• Simple random Sampling
Simple Random
– The most basic statistical
sampling method.
– each person in the
Systematic
population has an equal
probability of being
selected
Stratified
– Select at random
– Dice, Card, Random number
Cluster generator (Random Number
Table, calculator,
Randomly Excel) Systematically
select the select random
Develop Assign each
Define required numbers until it
sampling unit a
population amount of meets the
frame number
random sample size
numbers requirements
 Simple random sampling
Advantages
– Simple
– Sampling error easily measured

Disadvantages
– Need complete list of units
– Does not always achieve best representativeness
– Units may be scattered and poorly accessible
 Simple random sampling
Example: evaluate the prevalence of tooth
decay among 1200 children attending a school

• List of children attending the school

• Children numerated from 1 to 1200
• Sample size = 100 children
• Random sampling of 100 numbers between 1
and 1200

How to randomly select?

EPITABLE: random number listing
 EPITABLE: random number listing

Also possible in Excel

Simple random sampling
Four Probability Sampling Methods
 Systematic Sampling
Simple Random  A simplified version of simple random
sampling
 Select a random start, and then go by
equal space
Systematic

Stratified
 Question: how to determine the interval
so that everyone has a chance to be
selected?
Cluster Formula:
Interval = Population size / sample size

Select sample at regular intervals based

on sampling fraction
 1 26 51 76 101 126 151
151 176
176
Systematic sampling 2 27 52 77 102 127 152
152 177
177
3 28 53 78 103 128 153
153 178
178
“Example” 4 29 54 79 104 129 154
154 179
179
5 30 55 80 105 130 155
155 180
180
6 31 56 81 106 131 156
156 181
181
7 32 57 82 107 132 157
157 182
182
8 33 58 83 108 133 158
158 183
183
N = 200 9 34 59 84 109 134 159
159 184
184
10 35 60 85 110 135 160
160 185
185
11 36 61 86 111 136 161
161 186
186
want n = 25 12 37 62 87 112 137 162
162 187
187
13 38 63 88 113 138 163
163 188
188
N/n = 8 14
15
39
40
64
65
89
90
114
115
139
140
164
164
165
165
189
189
190
190
16 41 66 91 116 141 166
166 191
191
select a random number from 1-8: 17 42 67 92 117 142 167
167 192
192
18 43 68 93 118 143 168
168 193
193
choose 3 19 44 69 94 119 144 169
169 194
194
20 45 70 95 120 145 170
170 195
195
21 46 71 96 121 146 171 196
start with #3 & take every 8th unit 22 47 72 97 122 147
171
172
172
196
197
197
23 48 73 98 123 148 173
173 198
198
24 49 74 99 124 149 174
174 199
199
25
25 50
50 75
75 100
100 125
125 150
150 175
175 200
200
24
ExampleExample: systematic
of systematic
systematic samp
sampling
Example: sampling
Example: systematic sampling
Example: sys

Every eighth house is selected

 Systematic sampling
• Advantages
– Simple
– Sampling error easily measured
• Disadvantages
– Need complete list of units
– Periodicity
 Four Probability Sampling Methods
• Stratified Sampling
Simple Random
– Divide the population into subgroups
– Use simple random sampling method
Systematic (or systematic sampling) to select
from each group
– Combine to form one big sample
Stratified

– Think: what is the benefit of using

Cluster
stratified sampling?
• More representative
Divide sampling frame into homogeneous subgroups (strata) e.g. age-group,
occupation;
 Stratified sampling
• Advantages
– Can acquire information about whole population and
individual strata
– Precision increased if variability within strata is less
(homogenous) than between strata

• Disadvantages
– Can be difficult to identify strata
– Loss of precision if small numbers in individual strata
• resolve by sampling proportionate to stratum
population

An Example - Estimate vaccination coverage in a country

Four Probability Sampling Methods
• Cluster Sampling
Simple Random
– Divide the population into subgroups
-- called “clusters”.
– Randomly select some subgroups (not
Systematic
all!)
– In each selected subgroup, use random
Stratified
sampling technique to select sub-
samples
– Combine the sub-samples to form one
aggregate sample
Cluster
– Think: when we use cluster sampling?
(e.g. market research, select towns
first)
 Example: Cluster sampling
Section 1 Section 2

Section 3

Section 5

Section 4
 Cluster sampling
• Advantages
– Simple as complete list of sampling units within
population not required
– Less travel/resources required

• Disadvantages
– Potential problem is that cluster members are
more likely to be alike, than those in another
cluster (homogenous)….
– This “dependence” needs to be taken into
account in the sample size….and the analysis
 Selecting a sampling method

• Population to be studied
– Size/geographical distribution
– Heterogeneity with respect to variable
• Availability of list of sampling units
• Level of precision required
• Resources available
 Choosing Between Nonprobability & Probability
Sampling
Factor Nonprobability Probability

Nature of Research Exploratory Conclusive

Relative Magnitude of Sampling & Nonsampling errors larger Sampling errors larger
Nonsampling Errors

Population Variability Homogeneous Heterogeneous

(low variability) (high variability)

Statistical Considerations Unfavorable Favorable

Operational Considerations Favorable Unfavorable

34
 Summary of methods used
in probability samples
1. Simple, random sampling
• Draw subjects from list with random number
2. Systematic sampling
• Draw every xth subject
3. Stratified sampling
• Take one sample for each stratum
4. Cluster sampling
• Select clusters and then select individuals
 Key issues
• We cannot study the whole population so
we sample it.
• Taking a sample leads to sampling error,
which is measurable.
• Good design and quality assurance ensure
validity and while appropriate sample size
will ensure precision.
• Probability samples are the only ones that
allow use of statistics as we know them.
THANK YOU