SAMPLING IN QUANTITATIVE

RESEARCH
( NON-PROBABILITY AND PROBABILITY
SAMPLING)
Presented by
Dr.K.Latha ., M.Sc(N)., Ph.D

Associate Professor & HOD – Dept of Obstetrics and Gynaecology Nursing

SRM College of Nursing, SRM University
 Outlines
 Terminologies -definition
 Purpose of sampling
 Stages in the selection of a sample
 Types of sampling in quantitative researches
 Pros and cons of various sampling designs
 Estimation of sample size

Dr.K.LATHA 3/5/2017 2
 Introduction
 Sampling is a not a new development, but in recent times it

is used by people in all fields, even in day today life to get

an understanding about societies, opinions or situations.

 There is no hope of making scientific statements about a

population based on the knowledge obtained from a sample,

unless we are circumspect in choosing a sampling method

Dr.K.LATHA 3/5/2017 3
 Introduction
 Quantitative researches – select samples for
Generalization
 Qualitative Researches-Are not concerned
about Generalisability, But with
Understanding of the phenomenon

Dr.K.LATHA 3/5/2017 4
 Sampling Terminologies
 Sample
 Population or universe
 Target population
 Accessible population
 Census
 Sampling frame
 Sampling scheme
 Strata
 Sampling Terminologies
POPULATION
 It
is the entire aggregation of cases that meet a designated set of
criteria
◦ Any complete group
 People
 Sales territories
 Stores

Eg:-All woman in 60 years of age who are in

Psychiatric care in Chennai
or
 Entireaggregation of elements in which the researchers are
interested in
Eg:- Not only living things also Animals, books, records, buildings etc
6
Dr.K.LATHA 3/5/2017
Sampling Terminologies (Cont)
 Target Population
Is the aggregate of cases about which the
Researcher would like to make
generalization
E.g: All the Diabetic patients in Chennai is the
target population
 Accessible Population

It is the aggregate of cases that confirm to

the designated criteria and that are accessible
to the researcher as a pool of subjects for the
study
E.g:- The Diabetic patients who are available
in selected settings Chennai would be the
accessible population
Dr.K.LATHA 3/5/2017 7
 Sampling Terminologies (Cont)
 Strata- Refers to a mutually exclusive segment of a population
established by one or more characteristics

It is used for representativeness of the sample

E.g:- All registered Nurse population can be divided into 2

strata's (males & females)
 Sampling bias- Refers to the systematic over population or
under population of some segment of the population in terms of a
characteristics relevant to the research
 This can be prevented by homogenicity

Dr.K.LATHA 3/5/2017 8
Sampling Terminologies (Cont)
Sampling frame - A sample frame is the listing of all
population elements from which the sample will be
drawn.
 Census -A census is a
count of all the elements
in a population.
Sampling Terminologies (Cont)
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:
- It consists of subset of the units that compose the
population
 Sampling Terminologies (Cont)

Sampling…
It is the process of selecting a portion of the
population to represent the entire
population

Dr.K.LATHA 3/5/2017 11
 The need to sample ????

Sampling- a valid alternative to a census

when

 A survey of the entire population is impracticable

 Budget constraints restrict data collection

 Time constraints restrict data collection

 Results from data collection are needed quickly

The purpose of sampling…
 To gather data about the population in order to
make an inference that can be generalized to the
population

Dr.K.LATHA 3/5/2017 13
 Indispensable of sampling in
Research
Saves lot of time
Provides accuracy
Controls unlimited data
Studies individual
Reduces cost
Gives greater speed /helps to complete in stipulated
time
Assists to collect intensive and exhaustive data
Organizes conveniences

Dr.K.LATHA 3/5/2017 14
Disadvantages of Sampling (Defective
Sampling)
1. If sampling is biased, or not representative, or too
small, the conclusion may not be valid and reliable.
2. In research, the respondents to a study must have a
common characteristics which is the basis of the
study.
3. If the population is very large and there are many
sections and subsections, the sampling procedure
becomes very complicated.
4. The researcher always does not possess the necessary
skill and technical knowledge in sampling procedure.

Dr.K.LATHA 3/5/2017 15
 Principles of Sampling

Two important principles

 Principles of Statistical Regularity-random
(sufficient representative of the sample),
 Principles of Large Numbers-(steadiness , stability
and consistency)
Principles are referred to as the laws of sampling

Dr.K.LATHA 3/5/2017 16
Dr.K.LATHA 3/5/2017 17
Dr.K.LATHA 3/5/2017 18
Dr.K.LATHA 3/5/2017 19
 in
QUANTITATIVE
STUDIES
Dr.K.LATHA 3/5/2017 20
 QUANTITATIVE SAMPLING
IMPORTANT ISSUES
◦ Representation – the extent to which the sample is
representative of the population
◦ Generalization – the extent to which the results of the study
can be reasonably extended from the sample to the
population
◦ Sampling error - the chance occurrence that a randomly
selected sample is not representative of the population due
to errors inherent in the sampling technique
Dr.K.LATHA 3/5/2017 21
TYPES OF SAMPLING DESIGNS
Sampling
Techniques

Probability Non-Probability
Sampling Sampling

•Probability sample – a method of sampling that uses of

random selection so that all units/ cases in the population
have an equal probability of being chosen.
• Non-probability sample – does not involve random
selection and methods are not based on the rationale of
probability theory.
Dr.K.LATHA 3/5/2017 23
 Non Probability sampling technique
 Subjects does not have equal chance to be selected because the ,elements
are chosen by choice not by chance through non random sampling techniques
 It is likely to produce a biased samples than random methods
 Restricts generalization
 Researchers are bound by limitations – time, money, and workforce –
employ non random sampling techniques
 Downside – Unknown proportion is not sampled. Samples may or may not
represent the entire population accurately. There fore cannot be used for
generalization

Dr.K.LATHA 3/5/2017 24
 Uses of Non Probability Sampling

IT IS USED IN FOLLOWING SITUATIONS

 Utilized when the researcher targets to make a qualitative,
pilot or exploratory study.
 When the population is limitless and when randomization is
not possible
 When the research does not aim to produce results that could
not be used for generating generalizabilty
 Researcher has got limited budget, time and work force
Dr.K.LATHA 3/5/2017 25
 Non-Probability Sampling Designs

NON-PROBABILITY

Purposive /
Convenience/ Quota
Judgment
Snowball Sampling
Sampling
Sampling

Sunil Kumar
 1. Convenience Sampling
 It uses the most conveniently available samples in the study
 Subjects may be atypical of the population with regard to the
critical variables being measured
 Common of all sampling techniques
 Considered – fast, inexpensive, easy and the subjects are readily
available (Convenient accessibility and proximity to the researcher)
 Individuals may not be known to the researchers

Eg:- Teachers distributing Questionnaires in the class

Dr.K.LATHA 3/5/2017 27
Convenience sampling (Cont..)
the process of including whoever happens to be
available at the time
…called “accidental” or “haphazard”
sampling

Dr.K.LATHA 3/5/2017 28
Variant - Snow ball Sampling
Another type is
SNOW BALL SAMPLING OR NETWORK
SAMPLING-
It refers to that the early sample members are asked to identify
other people who meet the eligibility criteria who must be
otherwise difficult to identify
E.g. :-Obtaining subjects for study to observe a rare disease.
It is often considered as expedient but runs with the risk of
sampling bias
Dr.K.LATHA 3/5/2017 29
 Types of Snow Ball Sampling
 Linear Snowball Sampling
Subject refers only one other subject

 Exponential Non-Discriminative
Snowball Sampling
Subject gives multiple referrals and each
referral gives some more until required
sample size is reached.

 Exponential discriminative
Snowball Sampling
Subject refers multiple people but only
one is chosen as sample
Dr.K.LATHA 3/5/2017 30
 Snow ball sampling
Referral system helps find Potential sampling
samples quickly bias
Low Chance of lack of co-
cost solution
Works for hesitant subjects operation
Peer network might
Secretive groups can be
not exist
identified easily

Merits Demerits
Dr.K.LATHA 3/5/2017 31
 Merits – Convenient sampling
 Considered easiest, cheapest and least time consuming,
 Over riding principle is Researchers convenience and the
respondent availability

Demerits- Convenient sampling …

 difficulty in determining how much of the effect (dependent
variable) results from the cause (independent variable)
 Samples are not representative of the entire population
 Obvious criticism is sampling bias (can be reduced by
homogeneity)
 Samples may under represented or over represented or
completely left out

Dr.K.LATHA 3/5/2017 32
Dr.K.LATHA 3/5/2017 33
2. QUOTA SAMPLING

 It uses strata of the population established by one or

more characteristics, where the researcher determine

the proportion of the element needed from the various

segments of the population

 Method could yield more representative samples than

convenience sampling
 QUOTA SAMPLING (Cont… )

It is two staged restricted judgmental sampling

◦ I stage - Consists of developing control categories

or quotas of the population elements (quota can be –

age, gender, education, race , religion and SES )

◦ II stage – Sample elements are selected from

quotas based on convenience or judgment
Eg:- Out of 1000 samples only 200 is sample size, if male proportion

is high then the there will be too many in the sample

Dr.K.LATHA 3/5/2017 35
 Merits – Quota sampling
 The investigator can ensure segments are ensure that the diverse
segments are represented in the sample
 It avoids over representation or under representation of strata's
from the population
 It determines the correct number of samples from each strata
 It is less costly
 It does not require any sophisticated skills
 Administratively easy
 Reduces sampling error
Dr.K.LATHA 3/5/2017 36
 De-merits – Quota Sampling
 Though the technique seems to be wholly representative
of the population, it should be kept in mind that only the
chosen traits of the population were taken into
consideration while forming the sub groups
 Not possible to estimate errors
 Bias is possible , as investigator / interviewer can select
persons known to him.

Dr.K.LATHA 3/5/2017 37
Dr.K.LATHA 3/5/2017 38
 3. Purposive/ judgmental
sampling
The process whereby the researcher selects
a sample based on experience or knowledge
of the group to be sampled… called
“JUDGMENTALSAMPLING”

Dr.K.LATHA 3/5/2017 39
 PURPOSIVE SAMPLING (Cont)..

 The researcher might decide purposively to select the widest

possible variety of respondents or might choose subjects who are
judged to be atypical of the population in question
 E g:- Studying lived in experiences of post disaster depression
requires samples affected with the disaster depression

Or
A newly developed instrument can be tested with the purposive
sampling of diverse types of people
Eg. A scale on Post partum depression
 Merits – Purposive sampling

Merits
 A Sample of areas or persons can be chosen purposively

because they have certain typical characteristics

 Simple to draw sample and useful in explorative studies
 Saves resources, requires less field work

Dr.K.LATHA 3/5/2017 41
 Demerits – Purposive sampling

Demerits
 Always not reliable
 Human minds have difficulty in recognizing typical items
 Requires knowledge about population which he usually does not possess.
 Potential for inaccuracy in the researcher’s criteria and resulting in

inaccurate sample selections

 People who are less accessible (more difficult to contact, more reluctant

to participate) are under-represented

Dr.K.LATHA 3/5/2017 42
PROBABILITY SAMPLING DESIGNS

Dr.K.LATHA 3/5/2017 43
 RANDOM SAMPLING

Selecting subjects so that all members of a population have an

equal and independent chance of being selected

Probability Sampling

Simple Stratified Cluster

Random Random Systematic
Sampling
Sampling Sampling Sampling

Proportionate Dis Proportionate

One- Two Multi-
Stage Stage Stage
Sunil Kumar
 1. SIMPLE RANDOM SAMPLING
 Most pure and basic probability sampling design
 Each element of the frame thus has an equal
probability of selection.
• 2 Essential Pre-requisites to implement
 Homogeneous population
Sampling frame - it is the technical name for the actual list of
the sampling units or elements
 Can calculate the probability –
sample size (n) and population size (N)
Therefore, the probability is = n/N

Dr.K.Latha
 1.Simple random sampling (Cont.)
• Several ways of selecting a simple random sample

Humans have long practiced various forms of random

selection, such as picking a name out of a hat, or
choosing the short straw.
i. Lottery draw: The name or
identifying number of each item
in the population is recorded on a
slip of paper and placed in a box
- shuffled – randomly choose
required sample size from the
box.
 1.Simple random sampling (Cont.)

ii. Random
Number
Tables

Each item is
numbered and a
table of random
numbers is used
to select the
members of the
sample
Prepared by: Mdm. Nor Azian Abu Asan Unit 6: Business Decision
Dept. of Maths & Stats Making
Example 2: Simple Random Sampling (SRS)
Suppose your college has 500 students (population)
and you need to conduct a short survey on the quality
of the food served in the cafeteria. You decide that a
sample of 70 students (sample) should be sufficient for
your purposes.
In order to get your sample, you;

a. Assign a number from 001 to

500 to each students,
b. use a table of randomly
generated numbers (Random
Number Tables)
 1.Simple random sampling (Cont.)
iii. Soft ware programs – There are many software
programs, such as MINITAB and Excel that have routines that
will randomly select a given number of items from the
population

Image source: http://ccelearn.csus.edu/wasteclass/images/randomSampling.jpg

http://www.shebreathes.com/.a/6a00d8341c627a53ef010536e1e60b970b-450wi
 Simple random sampling - Advantages & Disadvantages
Advantages Disadvantages
i. Easiest method & commonly i. Make no use of auxiliary info.
used. Not subjected to bias ii. Time consuming
ii. Requires minimal knowledge iii. Laborious process
about the population in advance iv. Can be expensive and
iii. Free from sampling errors unfeasible for large population
iii. No chance of personal (to identified and reach) or if
preferences the personal interview required.
iv. Chance of selecting the v. not be representative of the
markedly deviant sample is whole population
low ,but the probability decreases vi. Doesn’t make use of the
as the size of the sample knowledge about a population
increases. which researcher may already
have
Dr.K.LATHA 3/5/2017 51
 2. Stratified Random Sample

 Used for heterogeneous population

 The population is divided into two or more groups

called strata, According to some criterion, such as

type of registered nurses, grade level, age, or income,
and subsamples are randomly selected from each
strata.

Sunil Kumar
 Example : Stratified (Random) Sampling
Using the same survey problem from Example (1): SRS

If you were select a simple random sample of 70 students

from the frame, you might be end up with just a little over
350 female students in your college, since they account for
more than half of a % of the whole college students
population).

a. Stratifying the population by gender. (Male and Female)

b. Calculate the exact sample size from each strata;
Male = (150/500)*70 = 21 male students
Female = (350/500)*70 = 49 female students
Give the total sample = 21 + 49 = 70 students
 2. Stratified Random Sample (Cont)
Stratified random sampling can be classified in to
a. PROPORTIONATE STRATIFIED SAMPLING
If the researcher might decide to select the subjects in
proportion to the size of the population then it is referred as
Proportionate stratified sampling
b. DISPROPORTIONATE STRATIFIED SAMPLING
Proportionate representation is not given to strata
It necessarily involves giving over representation to some strata
and under representation to other referred as Dis-Proportionate
stratified sampling
Example of Proportionate & Dis Proportionate stratified random sampling
Proportionate Dis-Proportionate
STARTUM A B C A B C
Population size 100 200 300 100 200 300
Sampling fraction 1/2 1/2 1/2 1/2 1/4 1/6
Final sample size 50 100 150 50 50 50
 Eg :Proportionate &Dis proportionate Stratified Sampling

 For example if we want to know public opinion about a certain national

issue of a country whose population is divide into four ethnics groups as
under:
Muslims = 40%
Christians = 30%
Hindus = 20%
Buddhists = 10%
 Now, If we want to draw a sample of 1000 persons from the population ,
then, according to the proportional stratified sampling, the composition of
our sample would be:
Muslims = 40/100 × 1000 = 400
Christians = 30/100 × 1000 = 300
Hindus = 20/100 × 1000 = 200
Buddhists = 10/100 × 1000 = 100
In a disproportional stratified sampling we randomly select elements from
the given strata not according to its ratio in the population of interest but
according to our will or need,may be 250 from each ethinic group which is
= 1000 samples.
  Dr.K.LATHA 3/5/2017 55
Stratified random - Advantages & Disadvantages
Advantages Disadvantages
i. Ensure an adequate sample size i. Classification - Problem if
for sub- groups in the population of strata not clearly defined.
interest. ii. Analysis is (or can be) quite
ii.Enhances representativeness complicated.
iii. Almost certainly produce a gain iii. Time consuming, requires
in precision in the estimates of the more labor and expensive
whole population, because a since the samples must be
heterogeneous population is split drawn from multiple
into fairly homogeneous strata. enumerated listings
iv. Easy to carry out iv. Prior knowledge of
v. Higher statistical efficiency composition and of
distribution of population
 3. CLUSTER (RANDOM) / MULTI STAGE
SAMPLLING
 Used for large scale survey
 In cluster sampling – There is successive random

sampling of units consisting of population elements

 It is used if it is impossible to obtain a listing of all the

element
 Sampling method divides the population into contagious

groups called clusters

 Sample are then drawn from the clusters by random

method
 The usual procedure is to sample successively such as

states ,cities ,districts, blocks, and their house holds

Also referred as multi- stage sampling/ this gives all
Dr.K.Latha
clusters a equal chances of being selected
3/5/2017 57
 Example 5: Cluster (Random) Sampling
Imagine that the municipal council of Kanchipuram wants to
investigate the use of health care services by residents.

a. Council requests for electoral subdivision maps that

identify and label each area block.
b. From this maps, the council creates a list of all area
blocks (e.g: Walajabad , Uthiramerur ,Sriperumbudur
Kundrathur ,Thiruporur , Kattankolathur ,Thirukalukundram,
Thomas Malai etc area will serve as the survey sampling
frame.
c. Every household in that area belongs to a area block.
d. Each area block represents a cluster of households.
 Types of cluster sampling
 Single stage cluster sampling- all the
elements of a cluster are selected as a
sample.
 Two stage cluster sampling- we
randomly select some clusters first from
the given population, then some elements
from each cluster are randomly selected. 
 Multi stage units : sampling is done
more than two levels
 Stratified cluster sampling- all the
clusters having similar characteristics are
stratified together and then at least one
cluster is randomly selected from each
stratum. After that either all or some
elements of each selected cluster are
sampled.
Dr.K.LATHA 3/5/2017 59
Cluster sampling - Advantages & Disadvantages
Advantages Disadvantages

i. More economical -Reduced field i. Clusters may not be

costs representative of whole
ii. Used if the population is population but may be too
particularly large and widely alike.
dispersed ii. Analysis more complicated
iii. Applicable where no complete than for SRS.
list of units is available iii. It contains more sampling
(special errors than simple or stratified
lists only need be formed for sampling
clusters). iv. Researcher can have over
represented clusters and under
represented clusters that can
skew the results of the study
4. Systematic random Sampling
 Variant of simple random sampling
 It involves the selection of every k th case from some list or
group such as every 10th person on the patient list or 100 th
person listed in membership list.
 It is characterized by random start, followed by a
predetermined or systematized order of selection
 It is best used in situation ,unless the population is narrowly
defined , and the sample is non probable in nature
Dr.K.LATHA 3/5/2017 61
 4.SYSTEMATIC SAMPLING (cont)
 If the researcher has the sampling frame, the following procedure can
be used
 Selection of units is based on sample interval, k starting from a
determined point, where k = N/n
 Note-The sampling interval is the standard distance between the
elements chosen for the sample
Determine the sample interval, k = N/n,
k = 300/100 = 3
k=3.
Systematic Sampling
• Similar to simple random sample. No table of random
numbers – select directly from sampling frame. Ratio
between sample size and population size

First unit
Work out Select select by
what fraction according to random
Develop fraction (100 numbers
Define Decide the of the frame
sampling sample from then every
population sample size the sample
frame 1,000 frame nth unit
size then 10% so
represents selected
every 10th unit) (e.g. every
10th)
Sunil Kumar
 Systematic Sampling - Advantages & Disadvantages
Advantages Disadvantages
i. Simple to carry out i. If list has periodic arrangement
ii. Easier to draw, without then sample collected may not be
mistakes. an accurate representation of the
iii. More precise than SRS entire population.
as more evenly spread ii.Sometimes results with biased
over population. samples
iv. Economical iii. Problem is that the list is
v.Staistically more arranged at intervals coinciding
efficient and provides with sampling interval
better representative E .g:- if all the 5th element is Head
Nurse, she will either always or
sample never be included in the study
Dr.K.LATHA 3/5/2017 65
Dr.K.LATHA 3/5/2017 66
Dr.K.LATHA 3/5/2017 67
 Selecting an Appropriate Sampling
Technique
 A researcher has to consider the research objectives first as to
do they call for qualitative or quantıtative research.
 Secondly, available resources should be kept in mind
including the time frame available for conducting the
researcher and making the findings available.
 Researchers should also focus on the need for statistical
analysis and degree of accuracy required with regard to the
research and the expected outcomes.
 Dr.K.LATHA 3/5/2017 69
3.05.2017 Dr.K.Latha
 Sampling & Non-Sampling Errors
 Non-sampling - errors refers to biases and mistakes in
selection of sample.
 Sampling errors - sampling error is the error caused by
observing a sample instead of the whole population.
◦ sampling error is incurred when the statistical characteristics of a
population are estimated from a subset, or sample, of that
population. 

The difference between the sample and population values is

considered a sampling error
Dr.K.LATHA 3/5/2017 71
Methods of reducing Sampling
Errors
Specific problem selection
Systematic documentation of related research
Effective enumeration
Effective pre testing
Controlling methodological bias
Selection of appropriate sampling techniques.

Dr.K.LATHA 3/5/2017 72
Dr.K.LATHA 3/5/2017 73
 WHAT IS SAMPLE SIZE?
 This is the sub-population to be studied in order to make
an inference to a reference population (A broader
population to which the findings from a study are to be
generalized)
 In census, the sample size is equal to the population size.

However, in research, because of time constraint and

budget, a representative sample are normally used.
 The larger the sample size the more accurate the findings

from a study.
 Therefore, an optimum sample size is an essential

component of any research.

74
Dr.K.LATHA 3/5/2017 75
WHAT IS SAMPLE SIZE DETERMINATION

 Sample size determination is the mathematical estimation

of the number of subjects/units to be included in a study.
 When a representative sample is taken from a population,

the finding are generalized to the population.

 Optimum sample size determination is required for the

following reasons:
1. To allow for appropriate analysis

2. To provide the desired level of accuracy

3. To allow validity of significance test.

76
HOW LARGE A SAMPLE DO I NEED?

 If the sample is too small:

1. Even a well conducted study may fail to answer it
research question
2. It may fail to detect important effect or
associations
3. It may associate this effect or association
imprecisely

77
 CONVERSELY
 If the sample size is too large:
1. The study will be difficult and costly
2. Time constraint
3. Available cases e.g rare disease.
4. Loss of accuracy(True value).

Hence, optimum sample size must be determined before

commencement of a study.

78
 Power analysis
 Power analysis is an important aspect of research design. It
allows us to determine the sample size required to detect an
effect of a given size with a given degree of confidence
 The following four quantities have an intimate relationship:

1. sample size
2. effect size – (is the difference between two means /divided
by the standard deviation of the two conditions.
3. significance level = P(Type I error) = probability of finding
an effect that is not there- usually set as 5% or 0.05
4. power = 1 - P(Type II error) = probability of not finding
an effect that is there
Given any three, we can determine the fourth.

Dr.K.LATHA 3/5/2017 79
 Strategy for assessing power
 Determine form of experiment/research
 Decide on hypothesis test
 Carefully consider which effect(s) is (are) important
to detect
 Select desired power (& alpha)
 Determine likely effect size (or range of effect sizes)
 Conduct power analysis
 Consider whether study is feasible as currently
designed

80
 Post hoc - Power Analysis

 When the estimated sample size can not be included

in a study, post-hoc power analysis should be carried
out.
 Post-hoc power analysis is done after a study has
been carried out to help to explain the results if a
study which did not find any significant effects.

81
 PROCEDURE FOR CALCULATING
SAMPLE SIZE.

There are four procedures that could be used for

calculating sample size:
1. Use of formulae
2. Ready made table
3. Nomograms
4. Computer software

82
 Sample Size – Thumb Rules
 According to Uma Sekaran in Research Method for Business
4th Edition, Roscoe (1975) proposed the rules of thumb for
determining sample size, where sample size larger than 30
and less than 500 are appropriate for most research, and
the minimum size of sample should be 30% of the
population.
◦ A large health science trial – 300 subjects in each groups
◦ Mid sized trials – 100 subjects in each group
◦ Small sized trial and PG thesis – at least 30 subjects in each
groups.
◦ Same norms for retrospective or case control studies
◦ Descriptive studies – atleast 200 subjects.
Eventually thumb rule lacks scientific basis

Dr.K.LATHA 3/5/2017 83
 PROCEDURE FOR CALCULATING
SAMPLE SIZE.
There are four procedures that could be used for
calculating sample size:
1. Use of formulae
2. Ready made table
3. Nomograms
4. Computer software

84
1. USE OF FORMULAE FOR SAMPLE SIZE
CALCULATION & POWER ANALYSIS

 There are many formulae for calculating sample size &

power in different situations for different study designs.
 The appropriate sample size for population-based study

is determined largely by 3 factors

1. The estimated prevalence of the variable of interest.

2. The desired level of confidence.

3. The acceptable margin of error.

85
 To calculate the minimum sample size required for accuracy, in
estimating proportions, the following decisions must be taken:
1. Decide on a reasonable estimate of key proportions (p) to be
measured in the study
2. Decide on the degree of accuracy (d) that is desired in the study.
~1%-5% or 0.01 and 0.05
3. Decide on the confidence level(Z) you want to use. Usually 95%
=1.96.
4. Determine the size (N) of the population that the sample is
supposed to represent.
5. Decide on the minimum differences you expect to find statistical
significance.

86
 For population >10,000.

n=Z pq/d 2 2

n= desired sample size(when the population>10,000)

Z=standard normal deviate; usually set at 1.96(or a~2), which
correspond to 95% confidence level.
p=proportion in the target population estimated to have a
particular characteristics. If there is no reasonable estimate, use
50%(i.e 0.5)
q=1-p(proportion in the target population not having the
particular characteristics)
d= degree of accuracy required, usually set at 0.05
level( occasionally at 2.0)
87
 E.g if the proportion of a target population with certain
characteristics is 0.50, Z statistics is 1.96 & we desire
accuracy at 0.05 level, then the sample size is
 n=Z2pq/d2

n=(1.962)(0.5)(0.5)/0.052
n=384.

88
If study population is < 10,000

nf=n/1+(n)/(N)
nf= desired sample size, when study population <10,000
n= desired sample size, when the study population > 10,000
N= estimate of the population size

Example, if n were found to be 400 and if the population size

were estimated at 1000, then nf will be calculated as follows

nf= 400/1+400/1000
nf= 400/1.4
nf=286
89
 SAMPLE SIZE FORMULA FOR COMPARISON OF GROUPS

If we wish to test difference(d) between two sub-samples regarding a proportion

& can assume an equal number of cases(n1=n2=n’) in two sub-samples, the
formula for n’ is

n’=2z2pq/d2
E.g suppose we want to compare an experimental group against a control group
with regards to women using contraception. If we expect p to be 40 & wish to
conclude that an observed difference of 0.10 or more is significant at the 0.05
level, the sample size will be:
n’= 2(1.96)2(0.4)(0.6)/0.12
=184
Thus, 184 experimental subject & another 184 control subjects are required.

90
 2. Sample size using determination
using a table
 Research advisors (2006), based on the sample size
calculation formula by Krejcie & Morgan (1970)
  In Nursing research studies, the following is ideal
 Eg. To study patients satisfaction
• Confidence interval – 95%
• Margin error – 5%
 Thus, if you have 5000 customers and you want to sample a
sufficient number to generate a 95% confidence interval that
predicted the proportion who would be repeat customers
within plus or minus 2.5%, you would need responses from
a (random) sample of 1176 of all your customers. 
 As you can see, using the table is much simpler than
employing a formula. Dr.K.LATHA 3/5/2017 91
92
 3. USE OF NOMOGRAM FOR SAMPLE SIZE CALCULATION
 A nomogram is a chart consisting of three or more lines or curves so
arranged that the required reading can be quickly made by just
moving the ruler.
 Various nomograms have been devised such as
 1) to calculate the sample size in diagnostic studies,
 2) to find the number of clusters required for estimating the
prevalence rate,
 3) to find the number needed to treat in a therapeutic trial against
values of absolute risk in the absence of treatment etc
 The nomogram has four elements to it
◦ (number of patients,
◦ confidence intervals,
◦ prevalence and
◦ sensitivity/specificity).
 If the researcher knows any three of these, they will be able to
estimate the fourth. 93
94
3. USE OF COMPUTER SOFTWARE FOR SAMPLE
SIZE CALCULATION & POWER ANALYSIS

The following software options that can be used for calculating

sample size & power;
 Epi-info
 nQuerry
 Power & precision
 Sample
 STATA
 SPSS (observed power)
 Web based power calculators
 G*Power (consistent with Dr. Jackson’s 2 nd principle of

economics)

95
STILL CONFUSED………………………..

Smart people don’t

do it alone……
Call a statistician
•Sample selection
•Sample size determination
•Analysis of data

96
A Word in Conclusion

Almost all research is a compromise between the ideal

and the possible.
Ideally, we should study the whole population; in practice,
but we don’t have time or capacity.
But care in our sample selection, both size and method, will
ensure that the research does not fall into the traps of
either introducing bias, or lacking precision.

This, in turn, will give it that vital

credibility.
Dr.K.LATHA 3/5/2017 97
 References

 Creswell, J., W. (2012) Educational research:

Planning, Conducting, and Evaluating Quantitative
and Qualitative Research, 4th ed.
 Patton, M.Q. (2002). Qualitative Research and

Evaluation Methods. Thousand Oaks, CA: Sage.

 Suresh K sharma (2016) .Nursing research and

staistics, 2nd edition, Elseivers Publications

Dr.K.LATHA 3/5/2017 98
Questions???

99
Thanks for your attention
Dr.K.LATHA 3/5/2017 10
1