STEPS OF RESEARCH

1. Selection of Problem

2. Formulation of hypothesis

3. Preparation of research design

4. Sampling

5. Collection of data

6. Analysis of data

7. Generalisation of result

Sampling

The way in which we select a sample of individuals to be research participants is critical.

How we select participants (random sampling) will determine the population to which we
may generalize our research findings.

The sample represents the group of individuals participating in the study, forming the basis
for the research findings. Selecting the correct sample is critical to ensuring the validity and
reliability of any research; the sample should be representative of the population.

What is sampling?

Sampling involves the strategic selection of individuals or a subset from a population, aiming
to derive statistical inferences and predict the characteristics of the entire population. It offers
a pragmatic and practical approach to examining the features of the whole population, which
would otherwise be difficult to achieve because studying the total population is expensive,
time-consuming, and often impossible. Researchers use various sampling methods to collect
samples from a large population to acquire relevant insights. The best sampling strategy for
research is determined by criteria such as the purpose of the study, available resources (time
and money), and research hypothesis.

What are sampling methods or sampling techniques?

 Sampling methods or sampling techniques in research are statistical methods for selecting a
sample representative of the whole population to study the population’s characteristics.
Sampling methods serve as invaluable tools for researchers. It helps in the collection of
meaningful data and facilitating analysis to identify distinctive features of the people.
Different sampling strategies can be used based on the characteristics of the population, the
study purpose, and the available resources. The various sample methods are

Types of sampling methods

There are two most common sampling methods:

1. Probability sampling: A sampling method in which each unit or element in the population
has an equal chance of being selected in the final sample. This is called random sampling,
emphasizing the random and non-zero probability nature of selecting samples. Such a
sampling technique ensures a more representative and unbiased sample, enabling robust
inferences about the entire population.
2. Non-probability sampling: Another sampling method is non-probability sampling, which
involves collecting data conveniently through a non-random selection based on predefined
criteria. This offers a straightforward way to gather data, although the resulting sample may
or may not accurately represent the entire population.

Types of probability sampling

Various probability sampling methods exist, such as simple random sampling, systematic
sampling, stratified sampling, and clustered sampling. Here, we provide detailed discussions
and illustrative examples for each of these sampling methods:

 Simple random sampling: In simple random sampling, each individual has an equal
probability of being chosen, and each selection is independent of the others. Because the
choice is entirely based on chance, this is also known as the method of chance selection. In
the simple random sampling method, the sample frame comprises the entire population.
 Systematic sampling: The systematic sampling approach involves selecting units or
elements at regular intervals from an ordered list of the population. Because the starting point
of this sampling method is chosen at random, it is more convenient than essential random
sampling. For a better understanding, consider the following example.
 Stratified sampling: Stratified sampling divides the population into subgroups (strata), and
random samples are drawn from each stratum in proportion to its size in the population.
Stratified sampling provides improved representation because each subgroup that differs in
significant ways is included in the final sample.

 Clustered sampling: In this sampling method, the population is divided into clusters, and
then a random sample of clusters is included in the final sample. Clustered sampling, distinct
from stratified sampling, involves subgroups (clusters) that exhibit characteristics similar to
the whole sample. In the case of small clusters, all members can be included in the final
sample, whereas for larger clusters, individuals within each cluster may be sampled using the
sampling above methods. This approach is referred to as multistage sampling. This sampling
method is well-suited for large and widely distributed populations; however, there is a
potential risk of sample error because ensuring that the sampled clusters truly represent the
entire population can be challenging.

Uses of probability sampling

Probability sampling methods find widespread use across diverse research disciplines
because of their ability to yield representative and unbiased samples. The advantages of
employing probability sampling include the following:

Representativeness

 Statistical inference

 Precision and reliability

 Generalizability

 Minimization of Selection Bias

Types of Non-probability Sampling

Non-probability sampling methods are further classified as convenience sampling,

consecutive sampling, quota sampling, purposive or judgmental sampling, and snowball
sampling.
 Convenience sampling: In convenience sampling, individuals are recruited directly from the
population based on the accessibility and proximity to the researcher. It is a simple,
inexpensive, and practical method of sample selection, yet convenience sampling suffers
from both sampling and selection bias due to a lack of appropriate population representation.

 Consecutive sampling: Participants in consecutive sampling (or sequential sampling) are

chosen based on their availability and desire to participate in the study as they become
available. This strategy entails sequentially recruiting individuals who fulfill the researcher’s
requirements.
 Quota sampling: The selection of individuals in quota sampling is based on non-random
selection criteria in which only participants with certain traits or proportions that are
representative of the population are included. Quota sampling involves setting predetermined
quotas for specific subgroups based on key demographics or other relevant characteristics.
This sampling method employs dividing the population into mutually exclusive subgroups
and then selecting sample units until the set quota is reached.
 Purposive or judgmental sampling: In purposive sampling, the researcher leverages
expertise to select a sample relevant to the study’s specific questions. This sampling method
is commonly applied in qualitative research, mainly when aiming to understand a particular
phenomenon, and is suitable for smaller population sizes.
 Snowball sampling: This sampling method is used when accessing the population is
challenging. It involves collecting the sample through a chain-referral process, where each
recruited candidate aids in finding others. These candidates share common traits, representing
the targeted population. This method is often used in qualitative research, particularly when
studying phenomena related to stigmatized or hidden populations.

Uses of non-probability sampling

Non-probability sampling approaches are employed in qualitative or exploratory research

where the goal is to investigate underlying population traits rather than generalizability. Non-
probability sampling methods are also helpful for the following purposes:

 Generating a hypothesis
 Qualitative research
 Convenience and pragmatism
Probability vs Non-probability Sampling Methods

Characteristics Probability sampling Non-probability sampling

Random selection of participants from Non-random selection of participants

Selection of
the population using randomization from the population based on
participants
methods convenience or criteria

Likely to yield a representative sample May not yield a representative sample

Representativeness of the whole population allowing for of the whole population; poor
generalizations generalizability

Precision and Provides more precise and accurate May have less precision and accuracy
accuracy estimates of population characteristics due to non-random selection

May introduce selection bias if criteria

Bias Minimizes selection bias
are subjective and not well-defined

Suited for statistical inference and

Less suited for statistical inference and
Statistical inference hypothesis testing and for making
hypothesis testing on the population
generalization to the population

Commonly used in qualitative and

Useful for quantitative research where
Application exploratory research where in-depth
generalizability is crucial
insights are the goal
Distinguishing Between a Sample and a Population:

The term population means all members that meet a set of specifications or a specified
criteria. For example, the population of Nepal is defined as all people residing in Nepal. A
population of inanimate objects can also exist, such as all automobiles manufactured in India
in the year 2003. A single member of any given population is referred to as an element. When
only some elements are selected from a population, we refer to that as a sample; when all
elements are included, we call it a census.

What is Census?

Census refers to the complete enumeration of a universe.

A universe may be a place or a specific locality, a group of people, or objects/ elements of

interest from which data is collected. The population of a census can vary according to the
focus or need of the research. A census is an attempt to list all elements/units/members in a
group and to measure one or more characteristics of those elements/units/members. Census is
a quantitative research method and the data collected is analysed with the use of statistics.

Census deals with large numbers. One of the major advantages of census method is accuracy
as each and every unit of the population is studied before drawing any conclusions of the
research. When more and more data are collected the degree of correctness of the information
also increases. Thus the results based on this method are also expected to be less biased.

Data collected through census method allows opportunity for intensive study and analysis
due to the enormity of the data collected. Today, census is well recognised for its advantages
for maintaining national level demographic figures, which are comparable on a periodic
scale, regionally, nationally as well as internationally.

Essential features of Census

 It is a quantitative research method.

 It advocates a systematic, standardized data collection based on predetermined parameters.

 It is defined by a specific universe (which can be a territorial region, a group of people, or

objects or elements like crops, vehicles etc.).

 Census is defined by a point or period of time.

 It covers every element of the universe or population of study, without omission or
duplication.

SURVEY

Survey is a method of collecting information from a sample of the population. A sample is a

subset of units in a population, selected to represent all units in a population of interest. It is a
partial enumeration because it is a count from part of the population. Information from the
sampled units is used to estimate the characteristics for the entire population of interest.
Surveys come in many different forms and have a wide variety of purposes. Unlike a census,
they gather information from only a small sample of people (or farms, businesses or other
units, depending on the purpose of the study). Information is collected by means of
standardized questions so that every individual/unit surveyed responds to and is evaluated for
exactly the same question. Data is gathered to study demographic information, knowledge,
opinions, behavior, attitudes, habits, desires, values and beliefs etc. Survey as a method seeks
to identify principles on the sample design, data collection instruments, statistical adjustment
of data and data processing. A survey can be used for both quantitative and qualitative
research

5.3.1 What is Survey Research?

Traditionally, surveys are recognised as a part of quantitative research, and are vastly used as
such till date. In quantitative research the sample is scientifically chosen so that each
individual in the population has a known chance of selection. In this way, the results can be
reliably projected to the larger public. Thus the sample must be comprehensive in its design
and extensive enough to provide a reliable representation of the whole population. The
objective is not to describe the particular individuals who by chance are part of the sample,
but to obtain a statistical profile of the population. Individual respondents are not identified.
Survey's results are presented in the form of summaries, such as statistical tables and charts.
A survey can be used to acquire demographic and socio-economic profile of a population,
however it can also be used for knowledge, opinions, behavior, and attitudes etc., however it
misses out on personal feelings or thoughts on the situation.

Advantages and Disadvantages of Survey

Research Survey research has the following advantages:

(i) It is inexpensive and less time consuming with in-depth fieldwork-based research.

(ii) It has a wide coverage (i.e. a large section of population can be covered through survey
research) Also, survey research can be carried out in remote locations through e-mails,
telephone etc.

(iii) Many questions can be asked through a survey.

(iv) Since the same set of question is asked to the whole sample, it is possible to draw
comparisons.

Survey research has the following disadvantages:

(i) Sometimes, it is difficult to frame question that are relevant to/ can be answered by
everyone included in the sample.

(ii) It is in- flexible in the sample.

(iii) These is a likelihood that respondents do not answer all the question or not provide
reliable data.

(iv) Socio-economic context is often ignored

1 2 Complete enumeration: Census is a systematic method that requires collecting recording

of data about all the members of the population. Partial enumeration: Survey considers a
subset of the population selected to represent the entire group, in all its characteris

S. Census Survey
N.
1. Census is used for quantitative research. Survey is used in both quantitative and
qualitative research.
2. Complete enumeration: Census is a Partial enumeration: Survey considers
systematic method that requires collecting a subset of the population selected to
recording of data about all the members of represent the entire group, in all its
the population. characteristics.
3. Census is best suited for a population of Survey is better suited for
heterogeneous nature homogeneous social groups.
4. As census accounts for every unit of the There are predictable chances of errors
population, the data when analyzed has
 greater accuracy and reliability. in the results drawn from the sample
5. Census leads to numerical enumerations of Survey can gather information on
the population opinions, behaviour, attitudes, habits,
desires, values and beliefs
6. Census requires considerable resources A survey requires comparatively less
(administrative, financial, logistical, resources as the enumeration of a
managerial etc.) sample is easier than the whole.
7. Census is time consuming. A survey requires comparatively less
time.

Sample Size

The sample size is defined as the number of observations used for determining the
estimations of a given population. The size of the sample is drawn from the population.
Sampling is the process of selection of a subset of individuals from the population to estimate
the characteristics of the whole population. The number of entities in a subset of a population
is selected for analysis.

A well-calculated sample size ensures statistical validity, reducing bias while maximizing
precision. Selecting the right sample size is crucial—too small, and the results may lack
significance; too large, and resources may be wasted on unnecessary data collection.
Researchers use specific formulas, incorporating variables like confidence level, margin of
error, and population variance, to determine the optimal sample size for reliable conclusions