Prof. Dr.

Saraswati Sharma (PhD)

Vice Principal –CON
Member Secretory of IRC
Session Outline
• Basic terms used Sampling technique
• Types of Sampling
• Technique of sampling
Population, Sample and Sampling
Population refers to the material of the study, whether it is
human subjects, animal, plants or objects

The sampling is
process of
selecting part of
population which
will represent
the whole.
Concept on Sampling
 When it is practically not possible to
include all units of the population for
investigation, then we do sampling.
 It is a process of systematically selecting
cases for inclusion in a research project.
 It involves the selection of a number of
study units from a defined study
population
Contd.
Sampling mortality:
• Is the dropout small size of the population. Mortality
is the loss of study subjects from the first data
collection point (pretest) to the second data collection
point (post test).
• Mortality in the sample refers to a problem that arises
when employing a longitudinal design and
participants who start the research process are unable
to complete the process. Essentially, the persons begin
the research process but then fail to complete the
entire set of research procedures.
Inclusion and Exclusion Criteria
 Inclusion and Exclusion Criteria are used for controlling
extraneous variables and for enhancing homogeneity and
power of the study and applicability of the result to the
population.
Inclusion Criteria:
 Are those characteristics that fulfil the needs of the
researcher for the study.
 The Purpose of Inclusion Criteria are to focus on certain
demographic parameters such as age, sex, ethnicity,
occupation, etc., clinical characteristic such as only Type II
Diabetes , geographic consideration such as Kathmandu.
Example: All under 5 year-old children admitted in Paediatric
Ward.
Contd.
Exclusion Criteria:
 Are those characteristics that disqualify prospective
subjects from inclusion in the study.
 The Purpose of Exclusion Criteria are used to ensure the
safety of the individual and quality of the result through
predict and eliminate expected biases like confounder bias,
incomplete data, etc., ethical problems like inability to
give consent.
 But, too many exclusion criteria will limit the
generalizability of the study findings.
Example: Age: 5 years old and more. Students who do not give
assent to participate in the study.
Types of Sampling

Probability sampling
Non-probability sampling
Classification of Sampling Methods
Sample Design

Probability Sampling Non -Probability Sampling

Simple Random Convenience

Systematic Consecutive

Stratified Random Quota

Judgment/Purposive
Cluster
Snowball
9
Probability sampling:
The choice of these sampling depends on the nature
of research problem, the availability of good
sampling frame, money, time, desired level of
accuracy in the sample and data collection
methods.
In probability sampling:
 Every element in the population under study has a
probability of selection to a sample, and every
member of the population has an equal probability
of being selected.
 Sampling Frame is Crucial in Probability Sampling
the population should be finite.
 Simple Random Sampling :
 Each element in the population has an equal
chance of being selected for the sample.
 It requires a good sampling frame (list of all
elements of the population) from which the
sample is selected .
Random selection methods
 Table of random numbers
 Lottery system
 Coin tossing
Contd.

 Shouldknow about the finite

(predetermined) population.

 [Link] taking 20 samples from 100 items,

it should be divided in 5 units, 20 in each
and selected randomly.
Contd.
 Advantages
 Free from classification error
 Requires minimum knowledge of the
population in advance
 Most reliable information can be access at the
least possible cost and save the time and labor
Systematic Sampling
 When the population size is larger it becomes
very tiresome to use table of random numbers to
select the sample.

 Systematic sampling can increase the sample‟s

representativeness when the population elements
can be ordered in some pattern with regard to
the characteristics being investigated.
Contd.

 Systematicsampling involves the selection

of every “Kth” case from some list or group,
such as every 10th person on a student list.
 „K‟ is the ratio of N/n,
 where „N‟ population as a whole
 „n‟ is the sample size required.
Contd.
Steps of systematic sampling
 First,the size of the population is divided by
the size of the desired sample to obtain the
sampling interval width
 The „sampling interval‟ is the standard
distance between the selected elements
 [Link] we want a sample of 50 from a
population of 5000 , our sampling interval
would be 100, ( 5000‟N‟/50 „n‟ = 100)
 Contd.

 Every 100th case on the sampling frame would be

sampled

 If the random number chosen is 73, the people

corresponding to numbers 73, 173,273,373, and
so forth would be included on the sample
Contd.
Advantages
 It is very easy to operate
 Even distribution of sample over the entire listed
population.
 Less expensive and faster than simple random
sampling.
Disadvantages:
 May sometimes result into a biased sample.
Stratified Sampling
 This sampling method is used when we have to
select samples from heterogeneous population

 In stratified random sampling, the population is

divided into two or more groups or strata that are
homogenous between groups with homogenous
characteristics such as age, gender, and
occupation, and then selecting the sample
randomly from each stratum
 Contd.

 There are three Key questions which have to be

addressed while using the stratified sampling:
The base of stratification: What characteristics
should be used to sub-divide the population into
strata?
The number of strata: How many strata should be
constructed, by rule there should be maximum 6
strata
Sample size within strata: How many observations
should be taken in each stratum.
Cluster Sampling

 When the population is too large, it is impossible

to obtain a listing of the elements.

 Total population is divided into cluster or parts.

 These cluster might be schools, colleges,

industries, or even geographical regions.
(In this sample individuals are not selected but the
region/ area is selected.)
Contd.

 The total population is divided into a number of

subsets called cluster.
 Some of these clusters are selected using simple
random sampling technique and the subject
within the selected clusters are studied.
 E.g. Suppose researcher want to study the quality
of nursing education provided by CTEVT, first you
should identify the Nsg colleges affiliated to
CTEVT, then only individual sample is selected
according to simple random sampling.
 Contd.

 Population is divided into groups, usually

geographic or organizational
 Some of the groups are randomly chosen
 In pure cluster sampling , whole cluster is
sampled
 Contd.
Advantages:

 Economic efficiency : faster and less

expensive than simple random sampling

 Does not require a list of all members

of the universe

 The same sample of clusters can be

used again and again for drawing
random sample
 Contd.
Disadvantages:

 Requires large sample size ( double the

sample size of simple random sampling)
 Sometimes differences in cluster may not
meet the sample population
Multistage Sampling
 Multistage sampling is the extension of cluster
sampling and is used when the population units
are arranged in hierarchical order.
 This sampling technique consists of two or more
stages of cluster sampling in sample selection.
 In this sampling large clusters of population are
divided into smaller clusters in two or more
stages in order to take samples in stages using
smaller and smaller sampling units at each
stages.
Contd.

Multistage sampling.
 First of all large population is divided into
cluster (1st stage) and
 then from the cluster, strata is divided (2nd
stage) and
 Then from the strata, individual is selected
(3rd stage) for sample.
This method is widely used in conducting area
survey so the basis of multistage sampling is the
geographical location.
 Multistage Sampling Contd.
 Eventually one or more cluster are chosen at random and
every individual meeting the criteria from the selected
clusters are included in the sample.
Example:
To find out the “Attitude of Adolescents of Nepal on ASRH
(Adolescent Sexual & Reproductive Health ”)
First select the some provinces from 7 provinces
Second select the some districts randomly from selected
provinces
Third some Municipalities from selected districts and
then select the adolescents from selected municipalities
 Merits and Demerits of Multistage
Sampling
Merits:
 Can use existing division and
subdivisions
 It is more flexible
Demerits:
 It may be biased because sample
selected only from selected clusters
 Analysis is more complicated
 Non- Probability Sampling
The sample are selected not by chance,
but by the purpose of the study or by
personal convenience, or the judgment of
the researcher. It is mostly used in
research including nursing research
 Each sampling unit of the population
being studied does not have an equal
chance of being included in the study
(due to the way the sample is selected)
 Convenience Sampling
 Also known as haphazard or accidental
sampling
 Selection of sample is totally based on
ease of accessibility.
 Depend on convenience of the
researcher
 Representativeness highly uncertain.
Consecutive Sampling
 Consecutive sampling, also known as
total enumerative sampling, is
a sampling technique in which every
subject meeting the criteria of inclusion
is selected until the
required sample size is achieved.
 It is very similar to convenience
sampling except that it seeks to include
all accessible subjects as part of the
sample.
Consecutive Sampling Contd.

 This technique is generally used in small-sized

populations.
 For example, if a researcher wants to study the
activity pattern of post kidney-transplant
patient, he can selects all the post kidney
transplant patients who meet the designed
inclusion & exclusion criteria, & who are
admitted in post-transplant ward during a
specific time period.
Merits of Consecutive Sampling

 Little effort for sampling

 Less expensive,
 Less time consuming,
 Ensures more representativeness of the
selected sample.
Quota Sampling
 A quota sampling is one in which the
researcher identifies strata of the population
and specifies the number (the quota)

 The researcher uses knowledge about the

population to build some representativeness
in the sampling plan

 The different between quota sampling and

cluster sampling is that there will be no
randomization in quota sampling
Contd.
 E.g. If you want to study unemployment
status in a VDC, first divide the subjected
people according to male- female, rich-
poor, literate-illiterate then quota
established
Steps of Quota Sampling:
 Deciding the division of population into
segments.
 Deciding what percent of the sample item
from each segment.
 Actual selection of the item.
Purposive Sampling or Judgmental Sampling

 The purposive sampling based on the

assumption that a researcher‟s
knowledge about population can be
used to hand pick the cases to be
included in the sample

 Subjects selected for a good reason

tied to purposes of research
Contd.

Example :
 Researcher wants to study practices of
traditional healers. It is impossible to
list all traditional healers in a
community and sample randomly from
the list. Instead, researcher uses
subjective information (e.g. from
villagers about their location) to
identify a “sample” of healers for
inclusion in his / her research project
Snowball Sampling or Network Sampling
 In sociological studies where a specific
behavior characteristic is present in few
individuals who are used to locate others
with same specific behaviors or similar
characteristics
 E.g. Socially stigmatized disease
condition, illicit drug use, female sex
worker, where respondents are difficult
to identify and contact
 Useful for hidden population
Snow Ball sampling
Snowball Sampling Contd.

 Snowballing begins with a few eligible

participants and then continues on the
basis of participants referrals

 Selecting the cases in a network

 Just as a snowball rolling downhill gets

larger and larger as it picks-up more
snow, a snowball sample picks up more
and more to be in sample over time
Snowball Sampling Contd.
The procedures followed in snowball
sampling are as follows:

 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
county,