UNIT-2

PSYCHOLOGICAL RESEARCH

Q1. What is the importance of sampling in research? Compare Probability and Non-
probability sampling techniques.
Sampling, as it relates to research, refers to the selection of individuals, units, and/or events
to be studied. Sampling can also be defined as the process by which researchers select a
representative subset or part of the total population that could be studied for their topic so that
they will be able to draw conclusions about the entire population.
The importance of sampling in research can be multifarious and varied depending of the type
of research being conducted as well as the personal perceptions of the researcher. Some them
are:
✓ When we sample, we select some cases to examine in detail, and then we use what we
learn from them to understand a much larger set of cases.

✓ The primary use of sampling in quantitative studies is to create a representative

sample (i.e., a sample, a selected small collection of cases or units) that closely
reproduces or represents features of interest in a larger collection of cases, called the
population.

✓ Generalization
Researchers examine data in a sample in detail, and if sampled correctly, they can
generalize its results to the entire population.

✓ Cost-effective
The number of people a researcher contacts is directly related to the cost of a study.
Sampling saves money by allowing researchers to gather the same answers from a
sample that they would receive from the population.

Sampling reduces the cost of their projects, and a study based on samples definitely
costs lower than conducting a census study.

Non-random sampling is significantly cheaper than random sampling, because it

lowers the cost associated with finding people and collecting data from them. Because
all research is conducted on a budget, saving money is important.

✓ Saves Time
Sampling saves time of the researcher or the research team. Many a times the
researcher is going to feel pressurized into completing the research project within a
certain time limit. This is where the sampling approach is likely to come in handy.
 Invariably, some people do not respond to the first effort at contacting them, which
means that researchers have to invest more time for follow-up. Random sampling is
much faster than surveying everyone in a population, and obtaining a non-random
sample is almost always faster than random sampling.

Thus, by reducing the larger population into smaller subsets and then making
inferences for the whole after studying these samples, the researcher often is able to
save a large amount of one of the most critical assets in while doing research i.e. time.

✓ Improves Accuracy
Sampling increases the accuracy and reliability of the obtained results. This increases
chances of obtaining more accurate and reliable results and at the same time
minimizes the amount of sampling error.

✓ Collecting Richer Data

Sometimes, the goal of research is to collect a little bit of data from a lot of people
(e.g., an opinion poll). At other times, the goal is to collect a lot of information from
just a few people (e.g., a user study or ethnographic interview).

Either way, sampling allows researchers to ask participants more questions and to
gather richer data than does contacting everyone in a population.

Sampling, on the basis of representation bias can be classified into probability and
non-sampling.

TYPES PROBABILITY SAMPLING NON-PROBABILITY

SAMPLING
CRITERIA
Probability sampling is based on Non-probability sampling is a
the fact that every member of a sampling technique where the
Definition population has a known and odds of any member being
equal chance of being selected. selected for a sample cannot be
calculated.

In the most basic form of In non-probability sampling, the

probability sampling (i.e., a members of the population will
Chance of being simple random sample), every not have an equal chance of being
selected member of the population has an selected, and in many cases, there
equal chance of being selected will be members of the population
into the study. who have no chance of being
For example, if you had a selected.
population of 100 people, each For example, a person might have
person would have odds of 1 out a better chance of being chosen if
of 100 of being chosen. they live close to the researcher or
have access to a computer.
 Probability Sampling involves Non-probability sampling, on the
randomly selecting individuals other hand, does not involve
Sample selection from subgroups of the population “random” processes for selecting
to ensure those groups are participants. Instead, it is based
appropriately represented in the on the subjective judgement of
study sample. the researcher.
For example, if a researcher For example, if a researcher wants
wants to study the role of gender to conduct a research study on the
stereotypes in a population of unemployment rate, he will
500 males and females. The selectively look for the desired
researcher decides that a sample population in those areas where
of 70 individuals would be abundant chances of significant
sufficient. In order to select a variables will be present.
sample, any researcher using
probability sampling would
either randomly call out people,
divide clusters or strata, or would
systematically select every nth
unit of population.

The goal of probability sampling The goal of non-probability

is to obtain a sample that is sampling is to indicate if a
Goal representative of the population particular trait or characteristic
of interest, so that the results of exists in a population or not.
the study may be generalized to Researchers use this method when
the population. they aim at contemporary
qualitative research like pilot
studies, or exploratory research.

Probability sampling is an ideal Non-probability sampling is not

choice for quantitative studies in ideal for quantitative research
Quantitative which the goal is to use because results from non-
Research statistical analysis to draw probability samples cannot be
conclusions about the generalized to the larger
population. population as confidently
compared to probability samples.

Probability sampling may be Non-probability sampling is well-

less appropriate for qualitative suited for many types of
studies in which the goal is to qualitative research. This is
describe a very specific group of because qualitative research is not
people and generalizing the always concerned with
Qualitative Studies results to a larger population is generalizing the results to a larger
not the focus of the study. population. Qualitative
researchers often use purposive
sampling, a non-probability
sampling technique in which the
researcher chooses participants
because they have specific
 expertise or insight regarding the
phenomenon of interest.

Probability Sampling is usually Getting response using non-

costlier and more time- probability sampling is faster and
consuming than non-probability more cost-effective than
Cost and Time sampling due to the probability sampling because, the
methodologies used, assignment sample is known to the
of random number to different researcher, the respondents
items of population and complex respond quickly as compared to
analysis and interpretation people who are randomly
stages. selected.
An inference can be that it is An inference would be that in a
relatively difficult to find right non-probability sampling it is
and study-appropriate easier to find respondents.
respondents in a probability
sampling.
Probability Sampling uses lesser In non-probability sampling, the
reliance over the human researcher, to a large degree is
judgment which makes the aware of the characteristics of the
Biasness overall process free from over- sample being selected, and with
biasness. an increased indulgence into
For example, if a researcher human judgement, researchers
wants to study the effect of using this method are at high risks
environmental conditions on of getting exposed to biasness.
behaviour, and he randomly For example, if a researcher wants
selects samples from the to study the effect of
population, then he will not environmental conditions on
engage in making subjective behaviour and he selects only
judgements and presumptions those samples from the
about the behaviour of samples population that are known to
according to the environment. him/her, then it is very likely, that
the researcher’s subjective bias
comes into play (assuming that he
is already irritable every time).

With a probability sampling In non-probability sampling,

strategy, researchers try to create sample does not accurately
an accurate representative represent the population. The
sample that has mathematically sample can seriously misrepresent
Accuracy and predictable errors. the features of a population.
Sample For example, if a researcher An example of such sampling is
Representativeness wants to study the effects of the person-on-the-street interview
punishment in the early conducted by television programs.
development of a child Television interviewers tend to
(preferably 3-6 years old), then pick people who look “normal” to
the researcher will randomly them and avoid people who are
select x number of children unattractive, disabled,
within the specified age group impoverished, elderly, or
for viewing the results, inarticulate.
 irrespective of the gender, height
appearance, etc.

Probability sampling is used Non- probability sampling is used

Variability in when the population is diverse or when the population has similar
population is heterogeneous in nature. traits or is homogeneous in
For example, a population nature.
having people of different ethnic, For example, a population having
language, gender or regional individuals of similar ethnic,
affiliations will be suited for language, gender, or regional
probability sampling. affiliations will be suited for non-
probability sampling.

Q2. Discuss Various types of Probability and Non-Probability sampling with

relevant examples.

PROBABILITY SAMPLING

A probability sampling method is any method of sampling that utilizes some form of
random selection, and therefore, following the principle of randomization or chance.
There are several different ways in which a probability sample can be calculated, the
method chosen depends on a number of factors, such as: available sampling frame,
how costly it is to survey members of the population, how spread out the population
is, how users will analyse the data.

Types of probability sampling

1. SIMPLE RANDOM SAMPLING

Simple random sampling is a random probability sampling method

which includes any method of sampling that utilizes some form of
random selection.

There are varied steps that a simple random sampling technique

follows:
At first, the researcher specifies the population and target population
and identifies its specific sampling elements.

Next researchers create an accurate sampling frame and they then use a
true random process to pick elements from the sampling frame.

Next, a researcher must locate the specific sampled element selected

by a random process.
To select elements from a sampling frame, the researcher will need to
create a list of random numbers that will tell him/her which elements
on it to select.
The researcher will need as many random numbers as there are
elements to be sampled. The random numbers should range from 1 (the
first element on the sampling frame) to the highest number in our
sampling frame.
For example: If the sampling frame lists 15,000 households, and we
want to sample 150 from it, we need a list of 150 random numbers.

There are two main ways to obtain a list of random numbers:

One, is the traditional way of obtaining a random-number table. In this
a list of numbers that has no pattern and that researchers use to create a
random process for selecting cases and other randomization purposes.
The numbers are generated by a pure random process so that any
number has an equal probability of appearing in any position.

Second, is the contemporary method of using computer programs to

produce lists of random numbers. Such programs are readily available
and often free.

Simple Random Sampling can either be present:

With replacement- that is, replacing an element after sampling it so it
has a chance to be selected again.
Without replacement- in this, the researchers toss out or ignore
elements already selected for the sample. For almost all practical
purposes in social science, random sampling is without replacement.

The set of many different samples is called as a sampling distribution.

It is a distribution of different samples. It reveals the frequency of
different sample outcomes from many separate random samples. The
sampling distribution pattern tells us that over many separate samples,
the true population parameter is more common than any other
outcome.
Some samples may deviate from the population parameter, but they
are less common. When researchers plot many random samples, the
sampling distribution always looks like a normal or bell-shaped curve.

The Central Limit Theorem is a mathematical relationship that states

when many random samples are drawn from a population, a normal
distribution is formed, and the centre of the distribution for a variable
equals the population parameter.
For a huge number of random samples, the sampling distribution
always forms a normal curve, and the midpoint of the curve will be the
population parameter.
 The central limit theorem lets the researchers estimate the chance that a
particular sample is unrepresentative or how much it deviates from the
population parameter. It lets them estimate the size of the sampling
error.
Random sampling does not guarantee that every random sample
perfectly represents the population. Instead, it means that most random
samples will be close to the population parameter most of the time. In
addition, researchers can calculate the precise probability that a
particular sample is inaccurate.

Confidence intervals are range of values, usually a little higher and

lower than a specific value found in a sample, within which a
researcher has a specified and high degree of confidence that the
population parameter lies.

The sampling distribution is the key idea that tells a researcher the
sampling error and confidence interval. Thus, no researcher can
conclude, “This sample gives a perfect measure of the population
parameter,” but they can say, “We are 95 percent certain that the true
population parameter is no more than 2 percent different from what
was have found in the sample.”

Some advantages of simple random sample are that: it is easier to use,

there is no need of prior information of the population required, and
there is an equal and independent chance of selection for each and
every sampling element.

However, this method also comes clubbed with potential

disadvantages, that are: the method becomes impracticable if the
sampling frame is large and the method does not represent a proper
proportionate sample.

An example of the simple random sampling would be to select 50

samples randomly from a population of 200 individuals.

2. SYSTEMATIC SAMPLING

The Systematic Sampling Method is a random sample in which a

researcher selects every nth (e.g., third or twelfth) case in the sampling
frame using a sampling interval.

In this method, a researcher first calculates the sampling interval to

create a quasi-random selection method. The sampling interval tells a
researcher how to select elements from a sampling frame by skipping
elements in the frame before selecting one for the sample.
 For instance, we want to sample 300 names from 900. After a random
starting point, we select every third name of the 900 to get a sample of
300. The sampling interval is 3.
A sampling interval is easy to compute and is the inverse of sampling
ratio.

In most cases, a simple random sample and a systematic sample yield

equivalent results. One important situation in which systematic
sampling cannot be substituted for simple random sampling occurs
when the elements in a sample are organized in some kind of cycle or
pattern.

Various stages of a systematic sampling are:

Defining a population.

Developing a sampling frame.

Deciding the sampling size.

Working out to decide which fraction of the interval of the frame and
the sample size represents.

Selection of sample according to fraction of the interval (eg, if there

are 100 samples from 1000 frames and 10% is the interval size, as a
result every 10th unit will be selected).

After the completion of each and every step, the first unit will be
selected randomly and thereafter, every nth unit (according to the
interval size) will get selected.

3. STRATIFIED SAMPLING

Stratified sampling is a random sample in which the researcher first

identifies a set of mutually exclusive and exhaustive categories, divides
the sampling frame by the categories, and then uses random selection
to select cases from each category.

When researchers use stratified sampling, they first divide the

population into subpopulations (strata) on the basis of supplementary
 information and after that draw a random sample from each
subpopulation.

In stratified sampling, researchers control the relative size of each

stratum rather than letting random processes control it. This guarantees
representativeness or fixes the proportion of different strata within a
sample.

In general, if the stratum information is accurate, stratified sampling

produces samples that are more representative of the population than
those of simple random sampling.

Researchers have fewer errors representing the population and a

smaller sampling error with stratified sampling.

Stratified sampling is used when a stratum of interest is a small

percentage of a population and random processes could miss the
stratum by chance.

For example, researchers draw a sample of 80 from 8000 people living

in a small town using information from the official governmental
authorities. The data was indicative of the fact that there are 150
individuals of 8000 are having mild intellectual disability and for the
researcher’s study, this is an important aspect to be considered. With
simple random sampling, such samples could be missed by chance.
However, with the use of stratified sampling a researcher can obtain a
list of 150 such individuals and select x random samples from it. This
guarantees that the sample represents the population with regard to the
important strata.

In special situations, a researcher may want the proportion of a stratum

in a sample to differ from its true proportion in the population.
For example, the population contains 0.5 percent French, but the
researcher wants to examine French in particular. He/she will
oversample so that French make up 10 percent of the sample.
With this type of disproportionate stratified sample, a researcher
cannot generalize directly from the sample to the population without
special adjustments.

In some situations, we want the proportion of a stratum or subgroup to

differ from its true proportion in the population.

Disproportionate sampling helps the researcher who wants to focus on

issues most relevant to a subpopulation.

4. CLUSTER SAMPLING
Cluster sampling is a type of random sample that uses multiple stages
and is often used to cover wide geographic areas in which aggregated
units are randomly selected and then samples are drawn from the
sampled aggregated units or clusters.

Researchers use cluster sampling to address two problems: the lack of

a good sampling frame for a dispersed population and the high cost to
reach a sampled element.
For example, there is no single list of all software engineers in India.
Even if researchers had an accurate sampling frame, it would cost too
much to reach the sampled engineers who are geographically spread
out in the entire nation. Instead of using a single sampling frame,
researchers use a sampling design that involves multiple stages and
clusters.

A cluster is a unit that contains final sampling elements but can be

treated temporarily as a sampling element itself.

First, researchers sample clusters, and then draw a second sample from
within the clusters selected in the first stage of sampling. A researcher
randomly sample clusters and then randomly sample elements from
within the selected clusters.

Some advantages of using cluster sampling are:

(i) A significant practical advantage is when researchers can create a
good sampling frame of clusters even if it is impossible to create one
for sampling elements. Once a researcher has a sample of clusters,
creating a sampling frame for elements within each cluster becomes
manageable.
(ii) A second advantage for geographically dispersed populations is
that elements within each cluster are physically closer to one another,
which can produce a savings in locating or reaching each element.

Researchers draw several samples in stages in cluster sampling.

In a 3-stage sample,
1st stage- a random sampling of large clusters.
2nd stage- a random sampling of small clusters within each selected
large cluster.
3rd stage- a sampling of elements from within the sampled small
clusters.
An example can be: First, a researcher randomly sample city blocks,
then households within blocks, and then individuals within households.
After selecting a random sample of blocks, he/she counts all
households on the selected blocks to create a sample frame for each
 block. Then he/she uses the list of households to draw a random
sample at the stage of sampling households. Finally, a specific
individual within each sampled household is chosen.

Cluster sampling is usually less expensive than simple random

sampling, but it is less accurate. Each stage in cluster sampling
introduces sampling errors, so a multistage cluster sample has more
sampling errors than a one-stage random sample.

When using cluster sampling, a researcher must decide the number of

clusters and the number of elements within clusters.
A design with more clusters is better because elements within clusters
tend to be similar to each other.
If few clusters are chosen, many similar elements could be selected,
which would be less representative of the total population.

When researchers sample from a large geographical area and must

travel to each element, cluster sampling significantly reduces travel
costs. As usual, there is a trade-off between accuracy and cost.
NON-PROBABILITY SAMPLING

Non-probability sampling is a sampling technique where the odds of

any member being selected for a sample cannot be calculated. It relies
on the subjective judgement of the researcher and does not involve
random selection. Methods used in the non-probability sampling are
not based on the rationale of probability theory.

Various types of non-probability sampling are:

1. CONVENIENCE SAMPLING

Convenience Sampling is a non-random sample in which the

researcher selects anyone he or she happens to come across. It is also
known as accidental, availability, or haphazard sampling.

In this type of sampling, the primary criteria of a researcher for

selecting cases are that they are easy to reach, convenient, or readily
available.

This sample type may be legitimate for a few exploratory preliminary

studies and some qualitative research studies when the research
purpose is something other than creating a representative sample.

The convenience sampling often produces very nonrepresentative

samples, so it is not recommended for creating an accurate sample to
 represent the population. When we select cases, our sample can
seriously misrepresent features in the entire population.

Convenience samples are easy, cheap, and quick to obtain. It is one of

the reasons why researchers rely on convenience sampling, which is
also the most commonly used type of non-probability sampling.

However, often it happens that the researchers are ignorant about how
to create a good representative sample.
An example is a newspaper that asks readers to clip a questionnaire
and mail it in, a Web site that asks users to click on a choice, or a
television program that asks viewers to call in their choices.
Such samples may have entertainment value, but they easily yield
highly misleading data that do not represent the population even when
a large number of people respond.

Two key ideas to remember about representative samples are that: (i)
self-selection yields a nonrepresentative sample and (ii) a big sample
size alone is not enough to make a sample representative.
By citing the same example as above, If you want to know about
everyone in city XYZ that has a population of 1 million, only some
read the newspaper, visit a Web site, or tuned into a program.
Also, not everyone who is reading the newspaper, visiting the Web
site, or has tuned in is equally interested in an issue. Some people will
respond, and there may be many of them (e.g., 50,000), but they are
self-selected.
Researchers cannot generalize accurately from self-selected people to
the entire population.
Many in the population do not read the newspaper, visit specific Web
sites, or tune into certain television programs, and even if they did,
they may lack the interest and motivation to participate.

In convenience sampling, everyone in a sample might be of the same

age, gender, or background.

2. QUOTA SAMPLING

Quota sampling is a non-random sample in which the researcher first

identifies general categories into which cases or people will be placed
and then selects cases to reach a predetermined number in each
category.

A well-designed quota sampling is an acceptable nonprobability

substitute method for producing a quasi-representative sample.
 In quota sampling, the researchers first identify relevant categories
among the population they are sampling to capture diversity among
units. Next, they determine how many cases to get for each category
this is the “quota.” Thus, a number of cases are fixed in various
categories of the sample at the start.

Quota sampling ensures that a sample has some diversity.

For example: a researcher selects twenty males and twenty females
under age 30 years of age, thirty males and thirty females aged 30 to
60, and ten males and ten females over age 60 for a 120-person
sample. While this is a start as a population’s diversity, it is difficult to
represent all possible population characteristics accurately.

Quota samples have three weaknesses:

(i) First, they capture only a few aspects (e.g., gender and age) of all
population diversity and ignore others (e.g., race-ethnicity, area of
residence in the city, income level).
(ii) Second, the fixed number of cases in each category may not
accurately reflect the proportion of cases in the total population for the
category. Perhaps 20 percent of city residents are over 60 years old but
are 10 percent of a quota.
(iii) Lastly, we use convenience sampling selection for each quota
category. For example, researchers include the first twenty males under
age 30 that they encounter— even if all twenty are high-income White
lawyers and belong from a well-off family. Nothing prevents us from
sampling only “friendly”-acting people who want us to pick them.

3. PURPOSIVE SAMPLING

Purposive sampling is a non-random sample in which the researcher

uses a wide range of methods to locate all possible cases of a highly
specific and difficult-to reach population.

Also known as Judgemental sampling, purposive sampling is a

valuable sampling type for special situations. It is used in exploratory
research or in field research.

It uses the judgment of an expert in selecting cases, or it selects cases

with a specific purpose in mind. It is inappropriate if the goal is to have
a representative sample or to pick the “average” or the “typical” case.
 In purposive sampling, cases selected rarely represent the entire
population.

Purposive sampling is appropriate to select unique cases that are

especially informative. For example, if a researcher wants to use
content analysis to study magazines to find cultural themes, he/she can
use three specific popular women’s magazines to study because they
are trend setting.

Researchers also use purposive sampling to identify particular types of

cases for in-depth investigation to gain a deeper understanding of
types.

In the purposive sampling method, researchers select the samples

based purely on the researcher’s knowledge and credibility. In other
words, researchers choose only those people who they deem fit to
participate in the research study.

Purposive sampling is not a scientific method of sampling and the

downside to this sampling technique is that the preconceived notions of
a researcher can influence the results. Thus, this research technique
involves a high amount of ambiguity.

4. SNOWBALL SAMPLING

Snowball sampling is a non-random sample in which the researcher

begins with one case and then, based on information about
interrelationships from that case, identifies other cases and repeats the
process again and again.

Snowball sampling, also called network, chain referral, reputational,

and respondent-driven sampling is a method for sampling (or
selecting) the cases in a network.

The method uses an analogy to a snowball, which begins small but

becomes larger as we roll it on wet snow and it picks up additional
snow.
For example, Rohan and Naman do not know each other directly, but
each has a good friend, Susan, so they have an indirect connection. All
three are part of the same friendship network. Researchers represent
such a network by drawing a sociogram.

Snowball sampling is a multistage technique. It begins with one or a

few people or cases and spreads out based on links to the initial cases.
 For example, we want to study friendship networks among the
teenagers in our community. We might start with three teenagers who
do not know each other. We ask each teen to name four close friends.
Next, we go to each set of four friends and ask each person to name
four close friends. This continues to the next round of four people and
repeats again. Before long, a large number of people have been
identified. Each person in the sample is directly or indirectly tied to the
original teenagers, and several people may have named the same
person.

The process stops, either because no new names are given, indicating a
closed network, or because the network is so large that it is at the limit
of what can be studied. The sample includes those named by at least
one other person in the network as being a close friend.
Snowball Sampling helps researchers to find a sample when they are
difficult to locate. Researchers use this technique when the sample size
is small and not easily available.

This sampling system works like the referral programme, once the
researcher finds suitable subjects, he/she asks them for assistance to
seek similar subjects to form a considerably good size sample.

5. DEVIANT CASE SAMPLING

Deviant case sampling is a non-random sample, especially used by

qualitative researchers, in which a researcher selects unusual or
nonconforming cases purposely as a way to provide increased insight
into social processes or a setting.

Also known as extreme case sampling, deviant case sampling is used

when researchers are interested in cases that differ from the dominant
pattern, mainstream, or predominant characteristics of other cases.

The goal of purposive sampling is to locate a collection of unusual,

different, or peculiar cases that are not representative of the whole.

Researchers use a variety of techniques to locate cases with specific

characteristics, and select cases because they are unusual.

Researchers can sometimes learn more about social life by considering

cases that fall outside the general pattern or including what is beyond
the main flow of events.
 For example, If researchers wish to study high school dropouts, and the
previous research suggested that a majority of dropouts come from
low-income, single-parent families who tend to be racial minorities.
The family environment is one in which parents and/or siblings have
low education or are themselves dropouts. In addition, many dropouts
engage in illegal behaviour.
The researchers might seek dropouts who are members of the majority
racial group, who have no record of illegal activities, and who are from
stable two-parent, upper-middle-income families. By looking at
atypical dropouts they might learn more about the reasons for dropping
out.

In the deviant case sampling, a researcher gets cases that are

substantially different from the dominant patter as it used to identify
cases that are unusual or special in some way.

6. SEQUENTIAL SAMPLING

Sequential sampling is a non-random sample in which a researcher

tries to find as many relevant cases as possible until time, financial
resources, or his or her energy is exhausted or until there is no new
information or diversity from the cases.

Sequential sampling differs because researchers continue to gather

cases until the amount of new information ends or a certain diversity of
cases is reached. The principle is to gather cases until we reach a
saturation point.

In economic terms, information is gathered until the marginal utility, or

incremental benefit for additional cases, levels off or drops
significantly. It requires that we continuously evaluate all collected
cases.
For example, researchers locate and plan in-depth interviews of fifty
mentally retarded people over 45 years of age who have been living
without an institutional care or many years. Depending on the research
purposes, getting an additional ten mentally retarded people whose life
experiences, social backgrounds, and worldviews differ little from the
first fifty may be unnecessary.

A sequence of one or more samples is taken from a group. Once the

group has been samples, a hypothesis test is performed to see if
researchers can reach to a conclusion. However, if the researchers
cannot perform the hypothesis test properly, the entire procedure is
repeated, until there is no additional information or new characteristic
in hand.
 In sequential sampling, the sample size is not set in advance because a
researcher does not know at the outset, how many times a procedure
will be repeated.

7. THEORETICAL SAMPLING

Theoretical sampling is a non-random sample in which the researcher

selects specific times, locations, or events to observe in order to
develop a social theory or evaluate theoretical ideas. In theoretical
sampling, what researchers samples (e.g., people, situations, events,
time periods) comes from grounded theory.

A growing theoretical interest guides the selection of sample cases.

The researcher selects cases based on new insights that the sample
could provide.
For example, a field researcher could be observing a site and a group
of people during weekdays. Theoretically, the researcher may question
whether the people act the same at other times or aspects of the site
change. He or she could then sample other time periods (e.g., nights
and weekends) to have a fuller picture and learn whether important
conditions are the same.

8. ADAPTIVE SAMPLING

Adaptive sampling is a nonprobability sampling technique used for

hidden populations in which several approaches to identify and recruit,
including a snowball or referral method, may be used.

Sampling hidden populations (i.e., people who engage in clandestine or

concealed activities) is a recurrent issue in the studies of deviant or
stigmatized behaviour (such as victims of sexual violence, illegal drug
users).

This method illustrates the creative application of sampling principles,

mixing qualitative and quantitative styles of research and combining
probability with nonprobability techniques.

Adaptive sampling is a design that adjusts based on early observations.

For example, researchers ask illegal drug users to refer other drug
users as in snowball sampling. However, they adjust the way that they
trace through the network based on our research topic. They might
identify a geographic area, divide it into sections randomly, and then
select participants in that area through strategies such as random-digit
dialing or by posting recruitment fliers.
Once the researchers identify members of the targeted hidden
population, they use them in a snowball technique to find others.

Through adaptive sampling, researchers get cases based on multiple

basis such as consecutive sampling followed by purposive sampling.

Adaptive sampling is implemented while a survey is being fielded that

means the sampling design is modified in real time, has continuous
data collection based on what has been learned from previous sampling
procedures that have already been completed.

The purpose of adaptive sampling is to improve the selection of

elements during the remainder of sampling, thereby improving the
representativeness of data that the entire sample yields.