You are on page 1of 5

See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/256446902

Sampling: Why and How of it? Anita S Acharya, Anupam Prakash, Pikee Saxena,
Aruna Nigam

Article · January 2013


DOI: 10.7713/ijms.2013.0032

CITATIONS READS
192 233,011

4 authors, including:

Anita Shankar Acharya Anupam Prakash


Lady Hardinge Medical College Lady Hardinge Medical College
84 PUBLICATIONS   989 CITATIONS    124 PUBLICATIONS   1,174 CITATIONS   

SEE PROFILE SEE PROFILE

Aruna Nigam
Jamia Hamdard University
136 PUBLICATIONS   1,164 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

HEALTH ACCOUNTING SCHEME EMPOWERING PEOPLE FOR HEALTH CARE THROUGH MULTI-SECTORAL COORDINATION - AN OPERATIONAL EVALUATION View project

video lectures for healthcare professionals, undergraduates MBBS students and postgraduates in obstetrics and gynaecology View project

All content following this page was uploaded by Anita Shankar Acharya on 30 May 2014.

The user has requested enhancement of the downloaded file.


INDIAN JOURNAL OF MEDICAL SPECIALITIES 2013;4(2):330-333

Symposium
Sampling: Why and How of it?
Anita S Acharya*, Anupam Prakash**, Pikee Saxena#, Aruna Nigam##

Abstract

A ‘sample’ is a subset of the population, selected so as to be representative of the larger population. Since
we cannot study the entire population we need to take a sample. Sampling techniques are broadly classified
into ‘Probability’ and ‘Non-probability’ samples. Probability sampling allows the investigator to generalise
the findings of the sample to the target population. Probability sampling includes Simple random sampling,
Systematic random sampling, Stratified random sampling, Cluster sampling, etc. A sampling frame is crucial
in probability sampling, because if the sampling frame is not drawn appropriately from the population of
interest, random sampling from that frame cannot address the research problem. Generalisations can be
made ‘only’ to the actual population defined by the sampling frame. Non-probability sampling includes
Convenience/purposive sampling, Quota sampling, Snow ball sampling, etc. Each method of sampling has
its own advantages and limitations, however, probability sampling is preferable, since its results can be
generalised.

Key words: Data collection; sampling studies; sampling bias; population; probability; random allocation.

Introduction Probability sample and 2) Non-probability


sample. Probability samples are the gold standard
In any research study, the best strategy is to in sampling methodology and also for ensuring
investigate the problem in the whole population. generalisibility of the study results to the target
But practically, it is always not possible to population. By probability sampling, we mean each
study the entire population. Alternatively, we individual in the population has an equal chance of
study a “sample” which is sufficiently large and being selected in the study. Probability sampling is
representative of the entire population. A sample further classified as:
is a subset of the population, selected so as to be
representative of the larger population [1]. By 1. Simple random sampling
taking a representative sample, we can reduce the 2. Systematic random sampling
costs incurred, the time taken to do the research 3. Stratified random sampling
and also the manpower needed to conduct the 4. Cluster sampling
study. Sample representativeness depends on three 5. Multiphase sampling
factors: 1) Sampling methodology 2) Sample size 6. Multistage sampling
and 3) Response rate. Sampling methods should be
systematic and defined so as to draw valid inferences All the above sampling methods use a random
from the sample. process.

Classification of sampling methods 1. Simple random sampling: In this method, every


individual has an equal chance of being selected
Broadly, sampling methods are classified as 1) in the sample from the population. Data is chosen

Departments of * Community Medicine, **Medicine and #Obs. & Gynae, Lady Hardinge Medical College & SSK Hospital, New Delhi-110001, India.
##
Department of Obs. & Gynae., HIMS&R, New Delhi, India.
Corresponding author: Dr. Anita S Acharya, Associate Professor, Department of Community Medicine, Lady Hardinge Medical College & SSK
Hospital, New Delhi-110001, India. E-mail: anitaacharya29@gmail.com
Received: 17-06-2013 | Accepted: 29-06-2013 | Published Online: 07-07-2013
This is an Open Access article distributed under the terms of the Creative Commons Attribution License (creativecommons.org/licenses/by/3.0)
Conflict of interest: None declared | Source of funding: Nil | DOI: http://dx.doi.org/10.7713/ijms.2013.0032

330 Indian Journal of Medical Specialities, Vol. 4, No. 2, July - Dec 2013
Sampling: why and how of it?

using random number table or computer generated on, till we complete the requisite sample size of 50.
list of random numbers. It can also be done by
lottery method, using currency notes, etc. 3. Stratified random sampling: Data is divided
into various sub-groups (strata) sharing common
In this method, a sampling frame is required. All characteristics like age, sex, race, income,
the individuals in the study population have to be education, and ethnicity. A random sample is taken
enumerated either in ascending or descending order. from each strata. The advantages are- it assures
The advantages of this method are that minimal representation of all groups in the population
knowledge of the population is required, the internal needed. The characteristics of each stratum can
as well as external validity is high and it is easy to be estimated and comparisons can be made. It also
analyse data. However, the limitations are that the reduces variability from systematic sampling. The
cost is high, a sampling frame is required. They tend limitations are that it requires accurate information
to have large sampling errors and less precision than on proportions of each stratum; also stratified lists
stratified samples of the same size [2]. are expensive to prepare.

Example- Let us say there are 200 participants Example- In studying the prevalence of diabetes in
in a conference and we would like to select 50 an adult population, it would be possible to stratify
participants by simple random sampling. The list of the population according to gender and then having
all the 200 participants would be available which equal number of subjects from both males and
constitutes the sampling frame. The 50 participants females. This would yield sex-wise prevalence of
can now be selected by either using random number diabetes. The sample could also be stratified by
table or by lottery method. Once a participant has place of residence such as urban, rural or peri-
been selected, that particular number is struck urban which would give us area-wise prevalence
off from the random number table. This method of diabetes with equal representation from each
is known as sampling without replacement. In this group.
way 50 participants are selected.
4. Cluster sampling: A cluster random sample is a
2. Systematic random sampling: In systematic two-step process in which the entire population is
sampling, the selection of the first subject is done divided into clusters or groups, usually geographic
randomly and then the subsequent subjects are areas or districts like villages, schools, wards, blocks,
selected by a periodic process. A systematic random etc. It is more commonly used in epidemiologic
sample is one in which every kth item is selected; k research than in clinical research. It is most
is determined by dividing the number of items in the practical to be used in large national surveys. The
sampling frame by the desired sample size. An initial clusters are chosen randomly. All individuals in the
starting point is selected by a random process, and cluster are taken in the sample. Usually it requires
then every kth number on the list is selected. The a larger sample size. Cluster sampling is very useful
advantages of this sampling is that it has moderate when the population is widely scattered and it is
usage, moderate cost, internal and external validity impractical to sample and select a representative
is high, it is simple to draw and easy to verify. The sample of all the elements [3].
disadvantage is that technically only the selection
of the first subject is a probability selection since Example- A sample could be taken from first year
for subsequent selections there would be subjects college students measuring their knowledge of
who will have zero chance of selection. Human papilloma virus (HPV) and cervical cancer.
Suppose all the colleges in Delhi are clusters.
Example- If we take the same example as above, We select 20 colleges either by simple random
N=200 and n=50, therefore, k = N/n, i.e. 4, which or systematic random sampling, and then every
becomes the sampling interval. Now we select a college becomes a cluster. We may then interview
random number between 1 to 4. Suppose it is “3”, all the students or randomly select students in each
so number “3” participant is our first subject. Then selected cluster for their knowledge on HPV.
we go on adding “4” to this number. Our subsequent
subjects would be 7, 11, 15, 19, 23, 27, 31 and so 5. Multiphase sampling: Multi-phase sampling

Indian Journal of Medical Specialities, Vol. 4, No. 2, July - Dec 2013 331
Anita S Acharya and others

is a complex form of cluster sampling. Here the same sampling unit is sampled multiple times.
population is organised into groups; subsequently
groups are randomly selected and then the Non-probability sampling: Non-probability samples
members are randomly selected in these groups are those in which the probability that a subject is
(an equal number selected per group). A part of the selected is unknown and results in selection bias in
information is collected from whole sample and part the study. They include the most commonly used
from sub-sample. This method of sampling is mostly convenience/purposive sampling, quota sampling,
carried out to increase precision, reduce costs and snowball sampling, etc.
reduce non-response.
1. Convenience/purposive sampling: This is the
Example- In a tuberculosis survey, Mantoux test is most commonly used sampling method. The sample
done in all cases (Phase I); in the next phase, X-ray is chosen on the basis of the convenience of the
chest is done in all Mantoux positive cases (Phase investigator. Often the respondents are selected
II); in the last phase, sputum examination is done in because they are at the right place at the right
all X-ray positive cases (Phase III). time. Convenience sampling is most commonly used
in clinical research where patients who meet the
Survey by this method of sampling is less costly, less inclusion criteria are recruited in the study. The
laborious & more purposeful. advantages are that they are most commonly used,
less expensive and there is no need for a list of all the
6. Multistage sampling: It is a complex form of population elements. However, they are not without
cluster sampling in which two or more levels of limitations; the foremost being variability and bias
units are embedded one in the other. It involves cannot be measured or controlled. Secondly, results
the repetition of two basic steps i.e. listing and from the data cannot be generalised beyond the
sampling. Typically, at each stage the cluster gets sample.
smaller in size and in the end, subject sampling is
done. Sometimes, special terminology is used for Example- Patients coming to the out-patient
various stages of sampling. The first stage sampling department of a hospital and meeting the inclusion
is called as ‘Primary Sampling Unit” (PSU), the criteria, school students, members of a social
second stage called as “Secondary Sampling Unit” organisation, etc.
(SSU), the third stage known as “Tertiary Sampling
Unit” (TSU) and so on till one gets to the “Final” or 2. Quota sampling: The sampling procedure that
“Ultimate” sampling units [2]. ensures that a certain characteristic of a population
sample will be represented to the exact extent that
Example- In a national survey, a random number the investigator desires.
of districts are chosen in all the states followed by
random number of talukas and villages. In the third Example- In a sample of 100, the investigator wishes
stage, houses will be selected. All the houses which to have 40% men & 60% women in the sample. He
are the final units of sampling are surveyed. would stop when 40 men are recruited i.e. ‘quota’
for men is over. The advantages are that the cost is
It is not as robust as true random sampling but moderate, it is very extensively used/understood,
probably helps to resolve the limitations inherent to and there is no need for list of population elements.
random sampling. It is extremely useful as it involves It also introduces some elements of stratification.
multiple stages of randomisations. Multi-stage The limitations are similar to convenience sampling.
sampling is used frequently when a complete list
of all the members of the population does not exist Stratified sampling and Quota sampling are similar in
and is inappropriate. The costs are thereby reduced that in both, population is divided into categories/
as compared to traditional cluster sampling. strata and subjects are selected from each category.
The purpose is to select a representative sample
It is to note that in multi-stage sampling, the and/or to allow sub –group analyses.
sampling units for the different stages are different.
On the other hand, in multi-phase sampling the However, there are certain differences between

332 Indian Journal of Medical Specialities, Vol. 4, No. 2, July - Dec 2013
Sampling: why and how of it?

“Stratification” and “Quota” sampling. In the study, it is important to choose a sound and scientific
former, selection of subjects is by simple random sampling methodology. Ideally, probability sampling
sampling once the categories have been created. methods should be used to ensure representativeness
Call-backs are used to get that particular subject. of the sample and also for generalisibility of the
Stratified sampling without call-backs may not, in results to the target population. If they are not
practice, be much different from quota sampling. In used, caution must be exercised in interpreting the
‘quota’ sampling, interviewer selects first available study results.
subject who meets the inclusion criteria. Thereby,
it is a convenience sampling. A sampling frame is
required for stratified sampling but not for quota Key Points
sampling. More importantly, stratified sampling uses
probability sampling, thus permitting the estimation • The sampling method chosen depends on the
of sampling error which is not possible with quota population of interest.
samples. • Careful planning is the key for generating
reliable results.
3. Snow-ball sampling: In this sampling procedure, • Probability samples are the gold standard in
the initial respondents are chosen by probability sampling methodology.
or non-probability methods, and then, additional • Probability sampling means one can
respondents are obtained by information provided generalise to the population defined by the
by the initial respondents. sampling frame.
• Non-probability sampling means one cannot
Example- In a study on a sample engaging in high- generalise beyond the sample.
risk behaviour or substance abuse, a person who is
engaging in a high-risk behaviour may name other References
persons involved in similar high-risk behaviour
practices, and this continues further till adequate
1. Probability & related topics for making
number of respondents are completed. The
inferences about data. In: Dawson B, Trapp RG
advantages are its low cost, usefulness in specific
(eds). Basic & clinical biostatistics. McGraw Hill,
circumstances and for locating rare populations.
The disadvantages include bias because sampling USA 2004,4th edn; 61-92.
units are not independent and projecting data 2. Choosing the type of probability sampling.
beyond sample is not justified. Available from http://www.sagepub.com/upm-
data/40803_5.pdf Accessed on 16th June,2013.
Conclusion 3. Baridalyne N. Sampling, sample size estimation
and randomisation. Indian J Med Spec
In order to have valid results from any research 2012;3:195-7.

Indian Journal of Medical Specialities, Vol. 4, No. 2, July - Dec 2013 333

View publication stats

You might also like