2. Why Do We Need Sampling?
• A critical question in any research enquiry is: Who are our ‘people’ or what is our population of interest? Since each and every individual in
the population cannot be observed directly, we are studying how we collect data from a subset of individuals (a sample) and use those
observations to make inferences about the entire population.
• Sampling is used when:
large population can be conveniently covered
we do not want to use up the cases (for example, when testing electric bulbs to see how long they last, we take a bulb and leave it on
until it burns out. We cannot test all the bulbs this way, because our whole objective is to ‘sell’ the bulbs and not ‘burn’ them out)
it is not necessary to survey all cases (for most purposes, taking a sample yields estimates that are accurate enough)
we have limited time, money and energy
units of area are homogenous
accuracy percentage is not acquired
the data are unlimited
3. What are the Advantages and Disadvantages of
Sampling?
Advantages
• Economical: significantly less costly than the entire population
• Increased speed: Less time-consuming than the population to collect, analyse and interpret data
• Greater scope: easier data handling and management, comprehensive scope and flexibility
• Accuracy: accurate and authentic results of the analysis, possibility of drawing valid inferences or generalisations
• Practical method: very practical method when the population is infinite
• Rapport: establishes adequate rapport with the respondents.
Disadvantages
• Biased: possibility of biased selection resulting in erroneous conclusions
• Difficulty in selecting truly a representative sample for complex topics
• Need for subject-specific knowledge
• Changeability of sampling units: unscientific method for heterogeneous units of population
• Impossible to select a representative sample in case of small or too heterogeneous population
4. Sampling Terminology
• Sampling constitutes an important component of any piece of research because of the significant impact that it can have
on the quality of your results/findings.
• Some of the important sampling terms used frequently include population, units, sample, sample size, sampling frame,
sampling techniques and sampling bias, among others.
• The most likely problems with obtaining a sampling frame are that:
– individual databases are often incomplete
– the information held about organisations in databases is sometimes inaccurate
– the information held in databases soon becomes obsolete or out of date
• Prior to selecting a sample, you have to define a sampling frame. A sample frame is a list of all the units of the population
of interest (e.g., names of individuals, telephone numbers, residential addresses, census tracts and so on). Researchers
have to make sure that your sampling frame is as complete, accurate and up to date as possible.
5. How Can You Select a Sample Unit?
• The choice of sample unit must be appropriate. A sample unit is the actual unit that we include in our
sample. Usually this unit refers to an individual person, but it could be an organisation, a company, a school
or a neighbourhood, depending on what you are measuring and how you are measuring it.
• It is important to avoid sampling bias when selecting a sample unit. Sampling bias occurs when the units
that are selected from the population for inclusion in our sample do not reflect the population. Put in
simple terms, it is any systematic failure of a sampling method to represent its population.
• For this reason, you have to think very carefully the types of sampling techniques that you will use for
selecting units for your sample. There are some sampling techniques, such as convenience sampling which
is a type of non-probability sampling (the university library example earlier), which are vulnerable to
greater bias than probability sampling techniques
6. What are Different Sampling Techniques?
The purpose of sampling techniques is to help the researcher select units (sample elements from population) to be
included in his or her sample. Broadly speaking, there are two groups of sampling technique:
Probability or representative sampling: Probability samples are those samples in which every unit in the
population has an equal chance or chance greater than zero of being selected in the sample. To achieve this,
the researcher utilises some form of random selection. A typical feature of a probability sample is the
absence of both ‘systematic’ and ‘sampling bias’.
Non-probability or judgemental sampling: Non-probability samples are those which a researcher selects
using his or her subjective judgement.These are samples where some elements of the population have ‘no’
chance of selection or where the probability of selection cannot be accurately determined. Since the
selection of elements is non-random, non-probability sampling does not allow the estimation of sampling
errors.
7. How Can You Select a Sampling Technique
and the Sample?
Researchers use several methods of sampling when doing a research study and it is important to select the most
appropriate sampling technique or method to obtain a representative sample.
• Probability Sampling: Among all the methods, probability sampling is the most preferred and the best overall group
of methods of sampling. The following five key probability sampling methods are the most widely used techniques
in social sciences research studies.
Simple random: In this method, each and every unit of the population has an equal chance of being selected or
included in the sample. In random sampling, the researcher considers the complete list of the universe but he
or she selects cases from this list at random. Random selection does not mean haphazard selection. What it
means is that the process of selecting a sample is independent of human judgement. Within this method, the
most commonly used methods are:
Lottery method
Random numbers table
8. How Can You Select a Sampling Technique
and the Sample? (Contd.)
Systematic Random Sampling: Systematic sampling is another random sampling technique. Researchers use this
technique for its simplicity and its periodic quality. Compared to random sampling, systematic sampling technique is a
simple technique. The other advantage of systematic random sampling is an evenly sampled population.
Stratified random: Stratified sampling can be proportional or non-proportional. In proportional sampling, the
participants are chosen in proportion to the number in each subgroup. Non-proportional sampling occurs when the
response weight of the subgroup is not a factor. There are two types of stratified random sampling: proportionate and
disproportionate.
Cluster Sampling: Cluster sampling refers to selection of respondents in groups (‘clusters’). In this method, instead of
selecting all the subjects from the entire population right off, we take several steps in gathering our sample population.
Multi-stage Sampling: Multi-stage sampling is a sampling technique that breaks the sampling process into several steps.
9. Non-probability Sampling
• Non-probability sampling, a valuable group of sampling technique, is commonly used in qualitative, mixed methods and even
quantitative research designs. In this sampling, the probability of any particular element of the population being chosen is unknown and
the basis of selection of units is quite arbitrary because researchers rely heavily on personal judgement.
• The method can be used when
the aim is to show that a particular trait exists in the population
the aim is to do a qualitative, pilot or exploratory study
it is impossible to do randomisation due to infinite size of the population
it is unlikely to generate results that will be used to create generalisations pertaining to the entire population
the resources (budget, time and workforce) available are limited
the aim is to significantly diminish the potential for researchers to study certain types of population, such as those populations that
are hidden or hard to reach (e.g., drug addicts, prostitutes) or where a list of the population simply does not exist
10. Non-probability Sampling (Contd.)
There are five types of non-probability sampling techniques:
• Convenience sampling: In convenience sampling, the sample is chosen on the basis of the convenience of the
investigator. The method may help the researcher in gathering useful data and information that would not have been
possible using probability sampling techniques, which require more formal access to lists of populations.
• Purposive or judgemental sampling: In this type of sampling, the researcher chooses subjects as part of the sample
with a specific purpose in mind. He or she believes that some subjects are more suitable for the research compared to
others. Purposive sampling technique is both less costly and less time-consuming. It ensures adequate presentation,
prevents entry of unnecessary and irrelevant items into the sample per chance, allows an in-depth analysis of selected
items and offers better results if the researcher is unbiased.
11. Non-probability Sampling (Contd.)
• Quota sampling: Quota sampling ensures that certain characteristics of a population sample will be represented to the exact
extent that the researcher desires. This technique demands the researcher to first identify relevant categories of people and
only then decide how many subjects are to be included in each category. Thus, the number of people in various categories of
the sample remains fixed. The major advantages of quota sampling are speed of data collection, lower costs and convenience.
Quota sampling becomes necessary when a subset of a population is under-represented and may not get any representation if
equal opportunity is provided to each.
• Self-selection sampling: Self-selection sampling is appropriate when the researcher wants to allow units or cases, whether
individuals or organisations, to choose to take part in research on their own accord. As a sample strategy, self-selection
sampling can be used with a wide range of research designs and research methods. The self-selection sample involves two
simple steps:
Advertising your need for units (or cases)
Verifying and checking the relevance of units (or cases) and either inviting or rejecting them
12. Non-probability Sampling (Contd.)
• Snowball Sampling: Snowball sampling (also called network, chain referral or reputational sampling) is usually done
when there is a very small population size or when the population you are interested in is hidden and/or hard to reach.
You can use this technique when you do not have access to sufficient people with the characteristics you are seeking.
These include populations such as drug addicts, homeless people, individuals with HIV/AIDS, prostitutes, unemployed
people, minority ethnic residents and so on.
• In this type of sampling, the researcher initially contacts a few potential respondents, interviews them and then asks if
they know of anybody else with the same characteristics he or she is looking for. A serious limitation of using a snowball
sample is that it is hardly representative of the population.
13. How Can You Determine the Sample Size?
• The key factors which govern the choice of the sample size are:
The confidence we need to have in your data
The margin of error that we can tolerate
The types of analyses we are going to undertake
• The determination of an ideal size of sample from the population is a complex process involving several concepts and
mathematical equations. The calculation process has been explained by considering the following five steps to make sure that
your sample accurately estimates your population.
Step 1: Knowing your population
Step 2: Degree of accuracy
Step 3: How big a sample do you need?
Step 4: How responsive will people be?
Step 5: So how many people do I send it to?