Sampling Methods

Recall…
Statistics is a tool for converting data into information:
Statistics

Data Information

But where then does data come from? How is it gathered?

How do we ensure its accurate? Is the data reliable? Is it
representative of the population from which it was drawn?
This chapter explores some of these issues.
Methods of Collecting Data…
There are many methods used to collect or obtain data for
statistical analysis. Three of the most popular methods are:
• Direct Observation
• Experiments, and
• Surveys.
 Surveys…
A survey solicits information from people; e.g. marketing
surveys of insecticides, pre-election polls etc.

Sample Survey : Sample survey is a method by which

detailed information on the population characteristics are
collected on the basis of sample elements. Population
parameters such as mean, standard deviation etc. are estimated
through sample survey.

Pilot Survey : Small scale surveys are sometimes conducted

in order to get quick primary information before census. Such
a survey is known as pilot survey.
The Response Rate (i.e. the proportion of all people selected
who complete the survey) is a key survey parameter.

Surveys may be administered in a variety of ways, e.g.

Personal Interview,
Telephone Interview,
Self Administered Questionnaire, and
Internet
 Population and Sample
• A population can be defined as including all people
or items with the characteristic one wishes to
understand.
• Because there is very rarely enough time or money
to gather information from everyone or everything
in a population, the goal becomes finding a
representative sample (or subset) of that
population.
Sample: A sample is a small representative fraction of a population. For
example, to investigate the adaptation status of a new insecticide of all
the farmers of the country, some farmers are selected to collect
necessary data, selected farmers constitute a sample of the population
of farmers.
Important statistical terms
Population:
a set which includes all
measurements of interest
to the researcher
(The collection of all responses,
measurements, or counts that are of interest)

Sample:
A subset of the population
Sampling:
Sampling is the process/method of selecting a sample from a
population
Target Population:
The population to be studied/ to which the investigator wants
to generalize his results
Sampling Unit:
Smallest unit from which sample can be selected
Sampling frame
List of all the sampling units from which sample is drawn
Sampling scheme/sampling plan/sampling method
Method of selecting sampling units from sampling frame
 Why sampling?
Get information about large populations
 Less costs
 Less field time
 More accuracy
 When it’s impossible to study the whole population
When might you sample the entire population?
When your population is very small
When you have extensive resources
When you don’t expect a very high response
 Classification of Sampling
Methods
Sampling
Methods

Probability Non-
Samples probability

Systematic Stratified Convenience Snowball

Simple
Cluster Judgment Quota
Random
Procedure for Drawing a Probability Sample

Step 1 Define the Target Population

Step 2 Identify the Sampling Frame

Step 3 Select a Sampling Method

Step 4 Determine the Sample Size

Step 5 Select the Sample Elements

Collect the Data from the

Step 6 Designated Elements
Simple random sampling
In this method each population unit has the same
probability for being included in the sample.

Probability of Selection = Sample Size/Population Size

How to select a simple random sample
1. Select a suitable sampling frame
or
List all the subjects/individuals in a population
2. Each element is assigned a number from 1 to N
3. Pick numbers n (sample size) from a list of random
numbers between 1 and N
4. Put the corresponding subjects in the sample.
Simple random sampling
Advantages :
◙ In simple random sampling each population element has
equal probability to be included in the sample.
◙ Investigator's personal bias is completely eliminated in this
sampling process.
◙ Most representative sample can be obtained if the
population is homogenous.
Disadvantages :
◙ For very large population the process of numbering the
population elements may be difficult and complicated.
◙ If the population is not homogeneous, sample drawn in this
method may not be representative.
◙ If the sample units are geographically far away from each
other, data collection often becomes expensive and
troublesome.
 Systematic sampling
This is a mixed sampling procedure. In this method, only one
sample unit is selected randomly and other units are selected
following specific system.
Pick a random case from the first k cases of a sample; select
every kth case after that one.
Sampling fraction
Ratio between sample size and population size
SYSTEMATIC SAMPLING

• 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/sample
size).
• 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 simple example would be to select every 10th name
from the telephone directory (an 'every 10th' sample,
also referred to as 'sampling with a skip of 10').
Systematic sampling
Advantages :
◙ This method is comparatively easier than both simple random
sampling and stratified random sampling. It requires less time and
labour.
◙ If the frame is randomly arranged, this method is more efficient than
simple random sampling.
◙ Proper implementation and supervision is easier in comparison with
other methods.

Disadvantages :
◙ The main demerit of this method is that it is not a random sampling
method in the true sense.
◙ If the correct and complete sampling frame is not known, sampling
in this method is not possible.
◙ If population size is not a multiple of the sample size -
(i) Resulting sample may not be of the required size.
(ii) Sample mean will not be an unbiased estimate of the
population mean.
5.18
Stratified Random Sampling
If the population is not homogeneous (the population elements
are not similar) in respect of the characteristic under study, a
simple random sample may not properly represent the
population. In such cases, the whole population is divided into
a number of more or less homogeneous subdivisions, these
subdivisions are called strata. From each of these subdivisions,
separate random selections of elements are made to constitute
a sample. This method of sampling is known as stratified
random sampling.
The strata should be such that -
◙ Elements included in each stratum should be as far as possible of
homogeneous nature, and
◙ Elements of different strata should be as far as possible of different
nature.
Stratified Random Sampling…

A stratified random sample is obtained by separating the

population into mutually exclusive sets, or strata, and then
drawing simple random samples from each stratum.
Strata 2 : Age Strata 3 : Occupation
< 20 professional
Strata 1 : Gender 20-30 clerical
Male 31-40 blue collar
Female 41-50 other
51-60
> 60
We can acquire about the total population,
make inferences within a stratum
or make comparisons across strata

5.20
Stratified Random Sampling…
After the population has been stratified, we can use simple
random sampling to generate the complete sample:

If we only have sufficient resources to sample 400 people total,

we would draw 100 of them from the low income group…

…if we are sampling 1000 people, we’d draw

50 of them from the high income group.

5.21
Advantages :
◙ Sample units are selected from different strata of the
population on the basis of relative importance, so the sample
drawn in this method is more representative compared to the
sample obtained by other methods.
◙ Administration of stratified random sampling is more
convenient than simple random sampling.
◙ Sampling unit selection is less expensive and less time
consuming in stratified random sampling compared to simple
random sampling.
◙ Supervision is comparatively easier in stratified random
sampling.
Disadvantages :
◙ Stratum selection sometimes may become complicated.
Improper stratification leads to reduce the reliability of the
collected information.
◙ It is not easy to determine the sample components of
different strata without previous experience.
◙ Sampling is not possible if sizes of the different strata are
not known.
  Types of Stratified Samples
 Proportional Stratified Sample:
 The number of sampling units drawn
from each stratum is in proportion to
the relative population size of that
stratum
 Disproportional Stratified Sample:
 The number of sampling units drawn
from each stratum is allocated
according to analytical considerations
e.g. as variability increases sample
size of stratum should increase
 Types of Stratified Samples…


 Optimal allocation stratified sample:

 The number of sampling units drawn from
each stratum is determined on the basis of
both size and variation.
 Calculated statistically
Determination of Number of Sample Units :
In stratified random sampling, there rare two different methods for
determining the number of choosable sample units from each stratum :
(i) Proportional allocation.
(ii) Optimum allocation.
Proportional Allocation : If the number of sample units for each
stratum are determined according to the same ratio, the method is known
as proportion allocation; 10% sample unit selection from each strata is
an example.
Optimum Allocation : Basic principle of this method is to select
sample component from each stratum such that variance of the estimate
becomes the minimum. This method is very useful if there exists wide
difference in the standard deviations of different strata. This method is
popularly known as Neyman's Optimum Allocation.
Proportionate & Disproportionate Stratified
Sampling
• Proportionate sampling plan, the sample will have specified
characteristics in exact proportion to those same characteristics
which are distributed in the population.
• E.g.: students in college of social work, taking year of study in the
college as basis of stratification.

• Distribution of students according to year in college

year population Proportion of each
class
Bsw 1 50 .25
Bsw2 40 .20
Bsw 3 30 .15
Msw 1 40 .20
•Msw 2 40 .20
Total 25-05-2025 200 1.00
(Lal Das D.K,2008) 27
Now, if the investigator has decided to draw a sample of
suppose 60 he will include the number of students from each
stratum in similar proportion. The procedure to be followed :

No.students from Bsw1 =60 X .25 = 15

No.students from Bsw2 =60 X .20 = 12
No.students from Bsw 3 =60 X .15 = 9
No.students from Msw =60 X .20 = 12
No. students from Msw2 =60 X.20=12
_________
60
25-05-2025
(Lal Das D.K,2008) 28
Disproportionate Stratified Sampling
• It is similar to the proportionate stratified sampling except
that the sub samples are not necessarily distributed
according to their proportionate weight in the population
from which they were drawn.
• E.g.: Distribution of Students by Sex
• sex No.Students Persentage

Male 160 80
Female 40 20
Total 200 100
(Lal Das D.K,2008)
• Now, if the investigator has decided to draw a sample of
suppose 60. then he has draw 30 from each sub-strata.
Disproportionate sampling gives equal weights to each 29
Cluster Random Sampling
In this method the population is divided into a required
number of mutually exclusive groups or classes; these groups
or classes are known as clusters. Then some clusters are
randomly selected and data are collected from all the units
included in these selected clusters.
Cluster Random Sampling
1. Take the population map.
2. Divide it into equal clusters
3. Assign each cluster a random number.
4. Select the cluster/s on the basis of a pre-decided rule.
5. Divide the selected cluster into sub-clusters.
6. Repeat steps 2-4 until you get a manageable sub-cluster.
 Cluster sampling
Section 1 Section 2

Section 3

Section 5

Section 4
 Advantages
 Low cost/high frequency of use
 Requires list of all clusters, but only of
individuals within chosen clusters
 Can estimate characteristics of both cluster and
population
 For multistage, has strengths of used methods

 Disadvantages
 Larger error for comparable size than other
probability methods
 Multistage very expensive
Types of Cluster Sampling
Cluster Sampling

One-Stage Two-Stage Multistage

Sampling Sampling Sampling

Simple Cluster Probability

Sampling Proportionate
to Size Sampling
Difference Between Strata and Clusters
• 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

35
 Non probability samples

 Convenience samples
sample is selected from elements of a population that
are easily accessible
 Snowball sampling (friend of friend….etc.)
 Purposive sampling (judgemental)
•You chose who you think should be in the
study
Quota sample
Non-Probability Sampling
Methods
 Convenience Sample
 The sampling procedure used to obtain
those units or people most conveniently
available
 Why: speed and cost

 External validity?

 Internal validity

 Is it ever justified?
 Judgment or Purposive Sample
 The sampling procedure in which an
experienced research selects the sample
based on some appropriate characteristic
of sample members… to serve a purpose
  Advantages
Moderate cost
 Commonly used/understood

 Sample will meet a specific objective

 Disadvantages
Bias!


 Projecting data beyond sample not

justified.
  Quota Sample
 The sampling procedure that ensure that
a certain characteristic of a population
sample will be represented to the exact
extent that the investigator desires
  Advantages
moderate cost

 Very extensively used/understood

 No need for list of population elements

 Introduces some elements of

stratification
 Disadvantages
 Variability and bias cannot be measured
or controlled (classification of subjects0
 Projecting data beyond sample not
justified.
  Snowball sampling
The sampling procedure in which the
initial respondents are chosen by
probability or non-probability methods,
and then additional respondents are
obtained by information provided by the
initial respondents
  Advantages
low cost

 Useful in specific circumstances

 Useful for locating rare populations

 Disadvantages
 Bias because sampling units not
independent
 Projecting data beyond sample not
justified.
Sampling Error…
Sampling error refers to differences between the sample and
the population that exist only because of the observations
that happened to be selected for the sample.

Increasing the sample size will reduce this type of error.

Nonsampling Error…
Nonsampling errors are more serious and are due to
mistakes made in the acquisition of data or due to the sample
observations being selected improperly. Three types of
nonsampling errors:

Errors in data acquisition,

Nonresponse errors, and
Selection bias.

Note: increasing the sample size will not reduce this type of
error.
Common research designs

• Experimental design
–Subjects are randomly assigned to treatments (=variables) by the
researcher
–Causal inferences are stronger
–Random sampling from the population less important
–Usually laboratory (exc. Moving to Opportunity, MTO)
• Observational design (e.g., surveys)
–Subjects are not randomly assigned to variables
–Random sampling is important.
–Selection bias
–Causal inferences are compromised.
Determining Sample Size
• What data do you need to consider
–Variance or heterogeneity of population
–The degree of acceptable error (confidence interval)
–Confidence level

–Generally, we need to make judgments on all these variables

Determining Sample Size
• Variance or heterogeneity of population
–Previous studies? Industry expectations? Pilot study?
–Sequential sampling
–Rule of thumb: the value of standard deviation is expected to be
1/6 of the range.
Determining Sample Size
Formulas:
Means n = (ZS/E) 2
Proportions n = Z2 pq/ E2
Percentiles n = pc (100 – pc) Z2/ E2

Z at 95% confidence = 1.96

Z at 99% confidence = 2.58
 Errors in sample

Systematic error (or bias)

Inaccurate response (information bias)
Selection bias

Sampling error (random error)

Type 1 error
• The probability of finding a difference with our sample
compared to population, and there really isn’t one….

• Known as the α (or “type 1 error”)

• Usually set at 5% (or 0.05)

Type 2 error

• The probability of not finding a difference that

actually exists between our sample compared to the
population…

• Known as the β (or “type 2 error”)

• Power is (1- β) and is usually 80%

 Sample size

Quantitative Qualitative

2
Z σ 2 Z2 π(1  π)
n n
D2 D2

(σ12  σ 22 )xF 2 P (1 - P) F
n n
D 2
D2
 Problem 1
A study is to be performed to determine a certain
parameter in a community. From a previous study a
sd of 46 was obtained.
If a sample error of up to 4 is to be accepted. How
many subjects should be included in this study at
99% level of confidence?
Answer

2
Z σ 2
n
D 2

2
2.58 x 46 2
n  880.3 ~ 881
42
Problem 2
• A study is to be done to determine effect of 2 drugs
(A and B) on blood glucose level. From previous
studies using those drugs, Sd of BGL of 8 and 12
g/dl were obtained respectively.
• A significant level of 95% and a power of 90% is
required to detect a mean difference between the two
groups of 3 g/dl. How many subjects should be
include in each group?
Answer

(σ  σ )xF
2 2
n 2
1 2
D

(8  12 )x10.5
2 2

n 2
 242.6 ~ 243
3
in each group
Problem 3
It was desired to estimate proportion of anaemic
children in a certain preparatory school. In a similar
study at another school a proportion of 30 % was
detected.
Compute the minimal sample size required at a
confidence limit of 95% and accepting a difference of
up to 4% of the true population.
Answer

Z π(1  π)
2
n 2
D

1.96 x 0.3(1  0.3)

2
n 2
 504.21 ~ 505
(0.04)
Problem 4
In previous studies, percentage of hypertensives among
Diabetics was 70% and among non diabetics was
40% in a certain community.
A researcher wants to perform a comparative study for
hypertension among diabetics and non-diabetics at a
confidence limit 95% and power 80%, What is the
minimal sample to be taken from each group with 4%
accepted difference of true value?
Answer

2 P (1 - P) F
n
D2

2 x 0.55 (1 - 0.55) x7.8

n 2
 2413.2
0.04
Precision
Cost