Table of Contents

1. Define variable. Explain the different types of variables with suitable examples used in social research. .................... 2
2. (CT)What is understood by measurement? Explain the various levels of measurement in social research. .................. 4
3. (CT)Define reliability. Explain the various types of reliability. .......................................................................................... 6
4. (CT)What is validity? Explain the different types of validity in social research. .............................................................. 8
5. (CT)Define Survey Research. Define Survey Design or Survey Research Design. Elucidate the steps involved in the
operation of survey research with a special focus on survey design. ................................................................................. 10
6. How to determine sample size? ....................................................................................................................................... 15
7. Define Experimental Research. Explain the various elements of classical experimental design. .................................. 18
8. Define the questionnaire. Compare the various forms of questionnaires used in survey research. Or, construct a
hypothetical questionnaire by employing the various forms of questions used in social surveys. ................................... 19
# Second Note of Questionnaire (ChatGPT):.................................................................................................................... 24
9. Define data processing. What are the tools and techniques of data processing? ......................................................... 28
10. Descriptive Statistics....................................................................................................................................................... 29
11. What do you understand by inferential analysis? Explain the uses of the chi-square test, t-test, z-test, and ANOVA
in social research. ................................................................................................................................................................. 31

1|Page
1. Define variable. Explain the different types of variables with
suitable examples used in social research.
Variable:
Variable is a very important element of quantitative research.
Variable is a concept that is capable of being measured. When we can measure a concept that can be termed as a
variable.
In quantitative research, a “variable” refers to any characteristic or quantity that can vary depending on time and
situation.
[A variable is a characteristic, attribute, or quantity that can be measured, manipulated, or controlled in research.
It represents something that can vary or change, and it is used to study relationships, effects, or patterns in
scientific experiments, statistical analyses, or observational studies.]
A variable represents any characteristic, number, or quantity that can be measured or quantified. The term
encompasses anything that can vary or change, ranging from simple concepts like age and height to more complex
ones like satisfaction levels and economic status.
Categories/Types of variables:
1. Continuous variable
2. Discrete variable
3. Independent variable
4. Dependent variable
5. Extraneous variable
6. Controlled variable
7. Intervening variable
8. Moderating variable
Continuous variable:
A continuous variable is a variable that does not have a minimum size unit and that can assume any value within
a particular range.
For example, Length. Length does not have a limited size unit. It can be 2.2 to 2.5 meters.
Continuous variables are numerical variables that can assume any value within a specified range. They can take
on an infinite number of values between two endpoints.
They often include fractions and decimals.
Discrete variable:
A discrete variable is a variable that does have a minimum-sized unit.
For example, the Number of children in a family. Number of cars.
A discrete variable is a variable that can take on only distinct, separate values that are countable, typically whole
numbers.

2|Page
Independent variable:
An Independent variable is a variable that a researcher wants to adopt to explain the changes in the dependent
variable.
Independent variables can be explained in terms of several characteristics.
An Independent variable is a variable that is manipulated either by the researcher or by nature or circumstance.
These variables are ones that are more or less controlled.
They cause the dependent variable to change.
Dependent variable:
A dependent variable is a variable that can change in response to the change of an independent variable.
A dependent variable is a variable that is affected by an independent variable. It is the presumed effect in a cause-
and-effect relationship.
Independent variable Dependent variable
The variable that doesn’t depend on other The variable depends on other variables.
variables.
Independent variables are controllable. Dependent variables are not controllable.
Explain the cause and effect of changes in the Explained by the independent variable.
variable
Independent variables are influenced by the Changes in the response variable are directly
researcher. caused by the changes in the dependent
variable.

Extraneous variable:
An extraneous variable is any variable other than the independent variable that may affect the outcome of an
experiment, potentially introducing error.
Researchers try to identify and control these variables to ensure the validity of their findings.
It is a variable that is not considered by the researcher within the scope of some research.
Control variable:
The control variable is a variable that is held constant by the researcher. It means that the effect of the control
variable is eliminated by the researcher.
A control variable is a variable that may affect the relationship between the independent and dependent variables,
which is controlled by eliminating the variable, holding the variable constant, or using statistical methods.
Intervening variable or Mediating variable:
It is a variable that links some independent variables with some dependent variables.
In our research, you might find that an independent variable does not have a direct link with the dependent
variable. In this case, a researcher is supposed to use this kind of variable to link the independent variable with
the dependent variable.

3|Page
For example, a researcher wants to determine or find out if there is a relationship between participation and
educational performance. There are two variables one is student participation and the other is the student's
academic performance.
Moderating variable:
A moderating variable is a variable that can influence the direction and strength of two variables. It can modify
the relationship. Especially the direction and strength of the relationship between an independent variable and a
dependent variable.
For example, I want to find out whether there is a relationship between sleep quality and the academic
performance of the student. The relationship between academic performance and sleep quality can be moderated
by some other variable, which is called mental health.
If a person has good mental health, his sleep quality would be better and that consequently can enhance academic
performance.
So, mental health can be a moderating variable.

2. (CT)What is understood by measurement? Explain the various

levels of measurement in social research.
Measurement:
• Measurement is the process of assigning numbers and symbols to the object according to the rule.
• Measurement refers to the process of assigning numerical values to objects or events according to
specific rules or scales.
• Measurement refers to the process of assigning numbers to objects or events in a consistent and
meaningful way.
Note: The goal of measurement is to quantity the attributes of an object or event so that, they can be compared
with other objects or events.
For example: male -> 1, female -> 2.
Here, 1 represents male, and 2 represents female. Now, we can measure the numbers of males and females in a
class, office, etc.
Education can be measured by its attributes—for instance, SSC, HSC, BA, MA, etc.
Level or scales of measurement:
American Psychologist ‘Stanley Smith developed a classification system for measurement scales in 1946. His
system is still used today and is known as Steven’s four-fold classification of scales. These are:

4|Page
 Ratio
scale

Interval
scale
Nominal scale

Nominal Scales:
Nominal scales of measurement are the simplest and lowest scales of measurement that only assign numbers or
symbols to the attributes of objects or events.
We cannot order, rank, or specify the interval of the number because numbers or symbols are only used for
assigning.
For example:
Each football player has his own jersey number which represents the player. Like, Rolando has 07, Messi has
10, etc. Number 07 represents Rolando and number 10 represents Messi.
Here number is the attribute.
Suppose, in India, Muslims represent ‘A’, Hindu represents ‘B’, and Buddha represents ‘C’ for data collection.
Here, the A, B, and C symbols are the attributes.
Characteristics:
1. Used to label, categorize, or classify variables.
2. Mutually exclusive sub-classes.
3. It has no quantitative value.
4. There is no inherent order or ranking among the categories.
5. It is the lowest level of measurement.
Ordinal Scale or Ordinal level of measurement:
The ordinal level of measurement involves the assignment of numbers to the attributes of a variable in order to
order them where numbers do not have any quantitative meaning.
We can’t calculate the mean, but we can calculate the median.
We can say meaningful distance.
For example:
The numbers 1, 2, and 3 might be used to represent three levels of satisfaction with a product: very good, good,
not good.
Suppose, Love can be leveled as not very much-> very much -> moderate -> a lot.
Characteristics:
1. It is the second level of measurement.
2. It has no quantitative value.
3. This level is used to categorize and also used to arrange in order.
5|Page
 4. It supplies more information than the nominal scale.
5. This scale data can be ordered but the difference between two data is not meaningful.
Interval Scale or Interval level of measurement:
Interval level of measurement involves the assignment of numbers to the attributes of a variable in order to
specify the meaningful distance or equal distance.
For example: temperature can be measured.
This scale has an arbitrary zero point, not an absolute zero point.
An example of meaningful distance is, suppose, temperature increases from 20° to 30°, 40° to 50°. We can say
the distance is 10° but we can’t calculate the ratio. We can’t say that temperature increases 2 times if the
distance for temperature is 20° to 40°.
Arbitrary zero: here, zero doesn’t mean zero. For example, temperature zero doesn’t mean no temperature.
Temperature zero means 0° temperature.
Characteristics:
1. The interval scale is a quantitative scale.
2. In this scale, there are equal intervals between the events or variables that are ordered.
3. It has arbitrary zero (0) but not absolute zero.
4. ‘+’, ‘-’ operation is possible.
5. ‘*’, ‘/’ operation is not possible.
Ratio level of measurement:
The ratio level of measurement has an absolute zero point.
It is the most complex and powerful tool of measurement. Because, we can measure height, weight, income,
are, etc.
In the ratio level of measurement, zero means zero.
Characteristics:
1. Ratio level of measurement is the most complex and powerful scale of measurement.
2. In this scale, variables can be ordered.
3. It has a true zero point.
4. ‘+’, ‘-’ operators and ‘*’, ‘/’ operators are possible.

Two things are very important. These are:

i. Reliability in measurement
ii. Validity in measurement

3. (CT)Define reliability. Explain the various types of reliability.

Reliability:

6|Page
Reliability refers to how consistently a method measures something. If the same result is produced by using the
same methods under the same circumstances, the measurement is considered reliable.
For example, if I use a scale to measure something’s weight and I find there are 4 kg. After some time, I measured
the same thing in a scale, and now I find 4.14 kg. So, that means there are some problems in this scale. This scale
is not reliable because it does not produce similar results if it is repeatedly used.
On the other hand, if I use a thermometer to measure the temperature of a liquid multiple times, and it consistently
shows the same result each time, the thermometer is considered reliable.
Reliability in the measurement can be categorized into 3 types:
1. Stability reliability
2. Representative reliability
3. Equivalence reliability
Stability reliability:
Stability reliability (sometimes called test or re-test reliability) is the agreement of measuring instruments over
time. It addresses the question of whether the measure delivers the same result when applied at different times.
Being a researcher, we can apply test or re-test methods.
Test or re-test method:
That method is used to determine the reliability or stability of the result over time. In the test/retest method, we
need to apply the instrument once under the same or similar conditions. The ratio between test and re-test scores
determines whether our instrument is reliable or not. The ratio between the result of the test and the re-test
determined reliability.
In this context, the greater the value of the ratio, the reliability of the instrument.
Suppose the test score is 50 and the re-test score is 50
Both are the same, if we divide 50/50, the ratio is 1
It means that it has the highest level of stability and reliability.
If our test score is 40 and the re-test score is 50
In this case, it is less than 1. It means it has less reliability compared to the previous one.
Another way is, if we deduct the re-test score from the test score and we get 0, that indicates the highest level of
reliability.
Representative reliability:
Representative reliability is reliability across sub-populations or different types of cases. It addresses the question
of whether the indicator delivers the same answer applies to different groups.
Suppose a researcher wants to know the age of participants, so he/she puts a question about age. But if a female
wants to hide her actual age, she gives the wrong answer, and if a male also gives a wrong answer about their age,
the researcher’s instrument doesn’t have reliability.

7|Page
The measure needs to give accurate information about every age group. So, now we can solve this kind of problem
through analysis that can help to address the problem. It’s called sub-population analysis.
Sub-population analysis:
It verifies whether or not representative reliability has or not.
Suppose we want to know a person’s education. So, we put questions like What is your education level? What is
your educational grade? And what is your knowledge? Here, I put 3 questions to measure a person’s education.
So, I conducted a sub-population analysis to see whether the question equally works for men and women and to
check the errors they make in answering the question.
If I find that men and women make equal mistakes/errors in answering questions, that means my question may
be reliable and equally applicable to men and women.
Equivalence reliability:
It actually applies when we have multiple indications. So, we have developed a set of indicators to measure
particular concepts or construct particular behaviors. Equivalence reliability will address the question of whether
the measure produces consistent results across different indicators.
If several indicators measure the same construct and produce the same result, then it has equivalence reliability.
How can we determine equivalence reliability? There are two tests.
1. Cron Bach’s alpha test [It is very useful to determine our scale equivalence or not]
2. Split-half method

These two tests are used to assess the internal consistency of a measurement instrument.

4. (CT)What is validity? Explain the different types of validity in

social research.
Validity in measurement:
Validity means the extent to which a measure is free from errors or the truthfulness and trustworthiness of the
measure.
It is the degree to which some instruments are free from errors. It addresses the question Am I measuring what I
am supposed to measure.
It means that I have developed a scale. If my scale is valid, it is supposed to measure some concept accurately.
So, validity is the degree to which a researcher measures what he is supposed to measure.
Validity is the degree to which a method measures what it is intended to measure.
Validity can be classified into various types:
1. Face validity.
2. Content validity.
8|Page
 3. Criterion validity.
i. Concurrent validity
ii. Predictive validity
4. Construct validity.
i. Convergent validity
ii. Discriminant validity
Face validity:
It is a subjective judgment of how well a test or assessment appears to measure what it’s intended to measure.
It is a validity where the validity of an instrument is determined by the subjective evaluation of some academics.
Content validity:
It addresses the question of whether the instrument reflects the true meaning of a particular context. It refers to
the extent to which a measurement instrument, such as a questionnaire or test, comprehensively assesses or
represents all facets of the construct it is intended to measure.
Criterion validity:
It uses some criteria to determine the validity of an instrument. Criterion validity refers to the extent to which a
measure is related to an outcome.
It refers to the degree to which a test or measurement accurately predicts or correlates with an established outcome.
There are two types of criteria validity:
1. Concurrent validity.
2. Predictive validity.
Concurrent validity:
It is a type of criterion-related validity that assesses the correlation between a new measure and an established
measure taken at the same time.
Predictive validity:
It is the ability of a test to predict future outcomes, such as performance or behavior.
Construct validity:
It is applicable when we have multiple indicators to measure a particular concept or construct. It addresses the
question of whether the measure is valid to the various indicators operated in a consistent manner. It’s similar to
equivalence validity.
There are two types of construct validity. These are:
1. Convergent validity.
2. Discriminant validity.
Convergent validity:
It applies when multiple indicators convert or are associated with each other.

9|Page
For example, if different tests designed to measure the same concept produce similar results, they have good
convergent validity.
Discriminant validity:
It is the opposite of convergent validity. According to discriminant validity, if ‘a’ and ‘b’ are two different concepts
or different constructs, the indicators that we develop measuring the ‘a’ construct or ‘b’ construct [construct means
concept] should not be associated with each other. The measures of ‘a and ‘b’ should not be associated with each
other.
For example, conservatism and urbanism, or modernism, are two different concepts. The indicators that we
measure for conservatism and on the other hand, the indicators that we measure for urbanism, should not be
associated with each other.
###Define survey design or survey research design. Explain the various types of survey designs

5. (CT)Define Survey Research. Define Survey Design or Survey

Research Design. Elucidate the steps involved in the operation of
survey research with a special focus on survey design.
Survey Research:
A survey is one kind of quantitative research method. A survey is a type of quantitative research method that
examines a larger population with the help of statistical tools.
•Survey is a type of research to obtain information by gathering data from a particular sample of a given
population, through personal or impersonal means, to study its characteristics.
•Survey research means collecting information about a group of people by asking them some questions and
analyzing the results.
•The term survey refers to techniques for collecting data from the target population of interest.
•While surveys are generally associated with human populations (for example, opinion polls, consumer surveys,
demographic and economic surveys), surveys of other types of populations (such as geological surveys and
administrative records) are also common.
•In most situations, however, data are collected from only a subset of the population, in which case the survey is
called a sample survey.
There are some objectives or purposes of survey research:
This research is conducted to address four purposes
1)Survey research is conducted or survey research method is applied in order to investigate or study socio-
economic conditions.
2)It Is conducted to identify the social and demographic characteristics of the population.

10 | P a g e
3)Survey research is conducted to address a social problem affecting some society or that affects a particular
community area or society.
4)Survey research is applied in order to identify or explore the opinions, and views of the people about a certain
issue.
There are two kinds of survey
1)census survey
2)sample survey
Census Survey
When data is collected from each and every case of the population, that is called a census survey.
Census is the complete enumeration of the population. In a census survey, data is collected from each and every
unit of the population.
We use a census when we want accurate information for many subdivisions of the population. Such a survey
usually requires a very large sample size, and often, a census offers the best solution.
Advantages of census survey
1. Data is more reliable because data is collected from the whole population.
2. There is no sampling bias or error.
3. Data for sub-populations may be available, assuming satisfactory response rates are achieved.
4.(Because of the above reasons) detailed cross-tabulations may be possible.
A disadvantage of census survey
1. It is a costly method since the statistician closely observes each and every item of the population.
2. It is time-consuming.
3. It requires a lot of manpower to collect the data.
Sample Survey
A Sample Survey is sometimes known as a social Survey.
A sample is a representative part of the population.
A sample is not the complete enumeration of the population. in a sample survey, data is not collected from each
and every unit of the population but rather is collected from some part of the population.
In a sample survey, data is collected from a sample that is down from the population. Therefore, a Sample Survey
is the logical estimate of the characteristics of the population. Population characteristics based on the sample.
Advantages of sample survey
1. Reduces cost - both in monetary terms and staffing requirements.
2. It reduces the time needed to collect and process the data and produce results as it requires a smaller scale of
operation.
11 | P a g e
3. (Because of the above reasons) enables more detailed questions to be asked.
4. Enables characteristics to be tested which could not otherwise be assessed.
5. Importantly, surveys lead to less respondent burden, as fewer people are needed to provide the required data.
6. Results can be made available quickly
A Disadvantage of Sample Survey
1. It can produce sampling error.
2. It is very difficult to produce a representative sample. if our population is too small and heterogenous, a sample
survey is not suitable.
3. To conduct a sample survey, training is required.
4. May have difficulty communicating the precision (accuracy) of the estimates to users.
5. Data for small geographical areas also may be too unreliable to be useful.

Survey Design or Survey Research Design:

Survey design is the planning of conducting a survey.
Survey design refers to the process of creating a set of questions and a methodology for collecting data from a
sample population. It involves determining the objectives of the survey, identifying the target population, selecting
an appropriate sampling method, and designing survey questions that are clear, unbiased, and relevant to the
research goals.
Effective survey design is crucial for obtaining reliable and valid data. The design process involves several key
steps, including defining the research question, determining the appropriate survey type (such as a cross-sectional
or longitudinal survey), selecting the survey method (such as online, phone, or in-person), and designing the
survey instrument (including the format, content, and response options)
Overall, survey design aims to produce a survey that accurately captures the opinions, attitudes, behaviors, or
other characteristics of a population of interest, and enables researchers to draw meaningful conclusions based on
the data collected.
Different forms of survey design
There are several types of survey designs that researchers can use to collect data, depending on their research
questions, the population of interest, and available resources. Here are some of the most common survey designs:
There are mainly two types of social survey Designs that are widely used in research.
1. Cross-sectional.
2. Longitudinal study
1. Cross-sectional: Cross-sectional study is a type of Survey design where data is collected from the sample or
various individuals at one point in time.
Key features of a cross-sectional study

12 | P a g e
1. A cross-sectional study is that here sample observation of data is collected from the sample at one point in time.
2. Cross-sectional study includes a variety of groups or individuals at a single time.
For example, a cross-sectional survey might be used to gather information about the prevalence of smoking among
college students. A random sample of college students will be surveyed, and their smoking status will be recorded.
2. Longitudinal study: A Longitudinal study is a type of survey design where data is collected from a sample or
population at different or several time periods.
For example, a longitudinal survey might be used to gather information about changes in a population's health
behaviors over time. A sample of individuals would be surveyed at multiple points in time, and changes in their
health behaviors would be recorded.
There are 3 types of longitudinal studies-
1. Trend analysis- Trend analysis is a type of longitudinal study that examines the changes within a population
over time. It is a type of longitudinal study where some characteristics of the population are mentioned at different
points in time to examine the changes.
For example, a census is conducted for trend analysis. Census can trend how social & demographic characteristics
of a population change over time.
2. Cohort study: A cohort study is a type of longitudinal study where a subgroup or subpopulation is studied over
time.
A cohort is a group defined by age. 20-25-year-old women are a cohort group. So, cohort means a particular age
group.
For example, researchers might survey a group of people who lived near a toxic waste site during a specific time
period and follow them over time to understand the health effects of their exposure.
3. Panel study: A panel study is a type of longitudinal study where the panel is collected from the same sample
where data is collected from the same population, same persons, or same individuals of the same group.
For example, a panel survey might be used to gather information about changes in a population's voting behavior.
The same group of individuals would be surveyed at multiple points in time, and changes in their voting behavior
would be recorded.

Phrases of Survey Research

There are a number of steps involved in survey research
1. Selecting the title of the Survey: depends on our research problem.
2. Identifying the purpose of the survey: why the research is intended or accomplished.
Purpose means the overall intent of the research and what the research is intended to do or accomplish. the purpose
of research should be narrowed down to some specific objective.
Example: To know the Socioeconomic conditions of the Chakma people or some group.

13 | P a g e
3. Reviewing relevant literature: In this step, our main objective is to orient ourselves with the available
literature related to our study.
We have to familiarize ourselves with the methods or techniques used in the previous study concerning our survey.
We have to find the research gap. what was done before and what has not been done. This is our gap. We have
identified our gap.
4. Designing Sample:
A. To identify the target population -The target population that we want to focus on in our research is called the
target population.
B. To determine sample size- there are some formulas on how to determine sample size.
5. Determining survey design
There are several steps
A. to determine the type of design-
They are the two types of design
1. Cross-sectional
2. Longitudinal
There are three types:
a. Trend analysis
b. panel study
c. Cohort study.
6. The construction of a questionnaire or interview schedule
A questionnaire is normally structured. So, in survey research, we actually employ a structured questionnaire.
there are options in the questionnaire, and the respondent ticks the right one, which reflects his opinion.
It can be semi-structured; some questions are structured, and some questions are non-structured.
7. Pre-testing questionnaire and interview schedule
to test the questionnaire whether is correct or complete to solve the researcher the problem to fulfill our research
purpose.
Is there any error in the Research question, and do we need to revise or modify it? you will pre-test the
questionnaire. You will apply the questionnaire to the 10-12 respondents so that we can understand what type of
changes you require.
the purpose of pretesting is to revise the questionnaire or update the questionnaire.
8. Finalize questionnaire after pre-test
Based on pre-test findings, we need to finalize the questionnaire.
9. collecting data or data collection
14 | P a g e
collect data through interviews or questionnaires. Whenever we collect data through interviews. It is very
important to follow some ethical rules. We don't need to pressure our respondents; we have to build rapport, and
we have to be very Honest in our research.
you have to express the objective of the research before beginning the data collection.
you have to share the purpose of the research with the respondents so that they can understand what the research
is about. you have to maintain the confidentiality and anonymity of the respondents.
10. data processing
The data we collect from the field. this is Raw in nature. we call its raw data processing involves
A. Data editing- after collecting our data, we have to look at the data very carefully to see whether there is any
error or missing data or whether any question is not answered by the respondents, we have to find out and then
edit.
B. Data coding- DC assigns the numbers to the responses so that they can analyze with the help of the software
or computer.
C. Classification- Classification is the arrangement of the data according to some common characteristics.
D. Tabulation- Tabulation is the representation of data through a table.
11. Data analysis and interpretation:
there are two types of statistics
A. Descriptive b. inferential
12. Report writing
The main objective of report writing is to share our research findings or survey findings among the readers,
academicians, policy maker
1. abstract
2. Title of the survey
3. Introduction
4. Purpose of the survey or study
5. Review of literature
6. Definition of the key terms
7. Methodology
8. Presentation of research findings or data analysis & interpretation
9. Conclusion

6. How to determine sample size?

Sample Size Determination
15 | P a g e
This is a very important part of the submarines when we are conducted server research, we are supposed to select
the sample from the population.
whenever we select a sample, a question arises about our desired sample size.
this is a very important problem in our research.
so, we don't know what our sample size.
Usually, when we select our desired sample size, there are some common rules or principles we have to follow-
1. Sample size should be representative:
Sample should reflect the true character of the population. if the sample does not reflect the true character of the
population. There is no representative sample.
2. The more samples, the better results:
3. If the population is homogenous, a small sample can serve the purpose.
4. If the population is heterogonous, a large sample is required.
There are four basic principles for determining sample size, but they are not good enough to select the sample
size.
However, there are some formulas that can help us select our desired sample size.
There are some concepts that are associated with or required for the sample size determination:
1. Level of Precision or margin of errors-
This concept is very much in use today. The level of Precision is a range within which the true value of the
population or population parameter will fall. So, the level of precision means the margins of error. Margins
between sample statistics and population parameters.
For example, 70% of the total respondents are illiterate, but we don't know if this is correct or not in the context
of the population; maybe it is true in the context of the sample. margins of error indicate how closely a sample’s
statistics reflect the true characteristics or true value of the population. Normally, our margins of error are 5%
/0.05
2. Confidence level:
The confidence level is the likelihood that a sample’s statistics will fall within a certain range.
In social research, we actually take a 95% confidence level. 95%, which means we have 05 chances out of 100
that the sample statistics will fall within a certain range.
we are 95% confident that our sample result is true.
that is very closely associated with a significant level. These two concepts are connected to each other but not the
same. when we have a 95% confidence level, we have a 5% significance level.
Therefore, the level of significance means the level of risk. A researcher is going to accept in his research we have
a 95% confidence level, which means we have a 5% significance level.
3. level of variability

16 | P a g e
The level of variability means the direction of attributes in the population. if there is a homogenous population,
the level of variability will be low. in the case of a heterogeneous population, the level of variability will be high.
In social research, the level of variability is 50% or 0.5
Mathematical calculation of sample from the population
There are two formulas
1. Cochran’s formulas (1977)
1. The infinite population.
Sample size, no=[Link]/ (e)2
Where Z= Critical or table value of the desired conference level.
Confidence level - critical value 95% -1.96, 90%- 1.645, 99%- 2.58
P=the distribution of attributes in the population (level of variability)
P=0.5, q=1-0.5=0.5
E= marginal error (0.05), no=sample size for infinite population (this formula is required when we don't know
the population size and the population is infinite)
2. When we know the population
N=no/1+(no-1)/N
Where,
n=sample size for finite population
N= population size
no=384
2. Yamane’s formula (1967)
Yamane's formula is a statistical formula used to determine sample size from a given population. It's particularly
useful when you want to ensure that your sample size is statistically significant and representative of the
population.
It is much easier compared to Cochran’s formula.
n=N/1+N(e2)
Where,
N= population size
n=desire sample size
e= margins of error. (0.05)

17 | P a g e
7. Define Experimental Research. Explain the various elements of
classical experimental design.
Experimental Method
The phrase experimental research has a range of definitions. The experimental method is a scientific and
systematic approach to research in which researchers manipulate one or more variables and controls and measure
any change in other variables. In the strict sense, the experimental method is what we call a true experiment.
• The experimental method is a quantitative research method that involves verifying hypotheses under a
controlled environment or laboratory.
• The experimental method aims to explore/investigate/find out the cause-effect relationship between an
independent variable and a dependent variable under study in a controlled situation.
Basic Elements of Experimental Research
1. Subjects: Subjects are the participants who are involved in experimental research.
2. Treatment: Treatment means an independent variable. It is a stimulus or an external condition that is
applied to the experimental group. The aim of a researcher is to determine the effect of treatment on the
dependent variable.
3. Dependent Variable: Dependent Variable is a variable of which we want to change. As a researcher, we
want to change the dependent variable as a result of treatment.
4. Experimental Group: Experimental Group is the group which receives some treatment.
5. Control Group: Control Group is the group which does not receive any treatment.
6. Random Assignment: Random assignment is the process of assigning the cases into the formation of two
groups on the basis of randomization.
7. Pre-test: Pre-test is the measurement of the dependent variable before some treatment.
8. Post test: Post test is opposite to pre test. It is the measurement of a dependent variable after the treatment.

Types of Experimental Design

There are 3 types of experimental designs that are used in Quantitative Research.
These are:
1. The Classical Experimental Design
The classical experimental design is also known as the true experimental design. It is a very ideal type of
experimental design, and it’s very popular. Classical experimental design includes all the basic elements/ parts of
experimental research, and that’s why it is called true experimental design. It contains all the basic elements of
an experimental design, including subjects, treatment, dependent variable, experimental group, control group,
random assignment, pre-test, and post-test.
Types of Classical Experimental Designs
18 | P a g e
 I. Pre-Test Post-Test Control Group Design: It is a type of true experimental design where test units are
randomly allocated to an experimental group and a control group. Both groups are measured before and
after the experimental group is exposed to a treatment. In this type, we need all the basic elements of
classical experimental research. For example, I am a scientist. I have innovated a vaccine to control the
COVID-19 disease. I want to know whether my vaccine is effective or not. How can I apply this type of
design? I want to carry out the experiment among my students. First, I had to form 2 groups randomly.
Secondly, I will assess the physical condition of the participants of the two groups. Thirdly, I have to apply
the treatment to the experimental group. The control group doesn’t receive any treatment. After doing
these, I am supposed to check whether the participants of the experimental group were infected with
COVID-19. Suppose, out of 20 students, 5 are affected with corona in the experimental group. On the
contrary, in the control group, 15 out of 20. Finally, I will compare the results.
II. Solomon Control Group Design: Sometimes, it happens that some effect of pre-testing still exists in the
post-testing. This creates a problem in getting reliable data. Richard Solomon developed the Solomon
four-group design to solve this kind of problem.
1. Pre-test, treatment, post-test
2. Pre-test, no treatment, post-test
3. Treatment, post-test
4. No treatment, post-test
The effectiveness of the treatment can be evaluated by comparing groups 1 and 3 and between groups 2 and 4.
For example, A mental health worker wants to know the coping skills of his clients. He designed a training method
for this. He formed four groups on random assignment. Two groups received pre-tests, and two groups did not.
Of those who received the pre-test, one group received the training method, and the other did not. The
experimental group received the new training method; a control group received the older model. All four groups
gave a post-test. If the experimental group and control group have similar results, then he may conclude that his
pre-test does not have any effects. If the two groups differ from each other, then the researcher can conclude that
his new training method is effective.

8. Define the questionnaire. Compare the various forms of

questionnaires used in survey research. Or, construct a hypothetical
questionnaire by employing the various forms of questions used in
social surveys.

Questionnaire construction
QC is an important part of Survey research.

19 | P a g e
Questionnaire: simply Questionnaire is an instrument or a tool for data collection. Broadly, a questionnaire is a
document that contains a set of questions that are asked to a group of respondents who provide information on a
particular topic or a research problem.
According to Goode and Halt:
A questionnaire refers to a device for securing answers to questions by using a form that the respondent fills in
himself.
A question involves a series of questions pertaining to psychological, social, educational, or any such issue that
is sent to an individual or a group with the aim of obtaining relevant information on a particular topic of research.
How do we construct a good questionnaire?
Whenever we are supposed to construct a questionnaire, we have to consider a number of things or points if we
want to construct a good questionnaire.
1. Align with Research Objectives
o Ensure all questions directly address the study’s goals.
2. Relevance & Completeness
o Verify questions are accurate, necessary, and cover all research aspects.
3. Simple Language
o Use clear, easy words. Avoid jargon, abbreviations, and slang.
4. Respondent’s Capability
o Questions must match respondents’ understanding level.
5. Avoid Ambiguity
o Be specific (e.g., "What is your monthly income?" instead of "What is your income?").
6. Respect Privacy
o Exclude sensitive/personal questions (e.g., about sex life).
7. No Threatening Questions
o Avoid questions that may intimidate respondents.
8. Avoid Double-Barreled Questions
o Don’t combine two questions (e.g., "Is poverty and crime Bangladesh’s main issue?").
9. No Leading Questions
o Avoid guiding answers (e.g., "You don’t smoke, right?").
10. Eliminate Bias
o Prestige bias: Don’t favor high-status opinions (e.g., "Doctors say smoking causes cancer. Do you
agree?").

20 | P a g e
 11. Avoid Emotional Language
o Keep questions neutral and factual.
12. No Double Negatives
o Poor grammar confuses (e.g., "I don’t have no money?" → "Do you have money?").
Types of question
In the questionnaire, we can find the varieties of types of the question, and let's talk about the types of question
1. Close-ended question -
A close-ended question is a type of question where respondents are offered a set of responses or answers and
asked to choose the one that closely reflects their opinion.
A close-ended question is very easy to understand and is necessary or helpful for quantitative analysis.
For example, what is your monthly income?
A.5000-1000 b.10000-20000
C.20000-30000 D.30000-400000
Advantages:
1. It is easier and quicker for the respondents to answer the questions.
2. We can compare the answers to the questions.
3. Are eligible for statistical analysis.
4. Respondents are more likely to answer sensitive questions.
5. Illiterate respondents can also answer.
Disadvantages:
1. Respondents cannot answer the question freely.
2. Respondents can be frustrated if their desired answer is not among the options.
3. Respondents may be confused if he is offered so many options.
4. If respondents make mistakes in filling out questions, the interviewer may not notice. This can happen in mail
questions or self-administered questions.
5. Marking the wrong answer is possible.
6. Close-ended questions can force the respondents to answer some questions that he is not okay to answer.
2. Open-ended Question: The open-ended question is a type of question where the respondent is free to answer.
So, respondents can answer questions according to their personal understanding & choice.
In the open-ended question, the probable answer is not provided. Rather, a space is given. So, the respondents
write down the appropriate answer according to their own understanding & choice.
For example, "What do you think are the biggest challenges facing our society today?"
21 | P a g e
Advantages:
1. Can answer freely.
2. Permits an unlimited number of possible answers.
3. Can answer the question in detail.
4. Can help the interviewer to produce unanticipated findings.
5. Can allow adequate answers to a complex issue.
6. Can develop the creativity of the respondents.
7. Allow the respondent to apply his own logic and develop the thinking process of the respondents.
Disadvantages:
1. Are very difficult to be quoted.
2. Are quite impossible to analyze with the help of statistical tools.
3. Responses might be irrelevant.
4. Illiterate people may find it difficult to answer.
5. Time-consuming.
6. It requires much physical and mental effort to fill out the open-ended questions.
7. Answers can take up a lot of space in the questionnaire.

3. Mixed Approach: If you employ the mixed approach, you are supposed to produce a mixed questionnaire. It
involves the combination of open-ended and close-ended questions.
For example, what’s your favorite hobby, and how often do you get to do it? (open-ended and closed-ended)
4. Contingency Question: The contingency Question is a type of question that is applicable to some group of
respondents and may be irrelevant to others. And it may not be irrelevant to some groups of respondents.
A contingency question is one kind of close-ended question that applies only to a sub-group of people.
Example: Do you pray, yes or no?
If yes, how many times do you pray? 5times, 3times, 2times.
In this case, the last question is determined by the answer to the previous question.

5. Matrix question: A matrix question is a kind of close-ended question.

A Matrix Question involves a set of questions or statements that uses some rating scale or response categories.
Matrix Question is the method of organizing a set of rating questions that have the same response categories.
Agree Neutral disagree
22 | P a g e
For Example, please rate the following aspects of your recent hotel stay:
Strongly Strongly
Disagree Disagree Neutral Agree Agree
Comfort of the bed
Cleanliness of the
room
Quality of the food
In this matrix question, the respondent is asked to rate multiple aspects of their recent hotel stay on a scale of
strongly disagree to strongly agree. The matrix format makes it easy to answer multiple questions quickly and
compare ratings across different aspects of the hotel stay.
The various types of Questionnaires
1. Mail Questionnaire
The Mail question is the type of question that is needed by the respondents by the host or Email. Mail questions
can be found in Survey Research.
Advantages of Mail Question
1. Low cost- Mail question is very low cost because sending a question by host or email involve very low cost.
At the same time, mail question does not require any trained personnel data collector. Therefore, it could have
money.
2. Reduction in bias- Mail questions can help the researcher to reduce biases. The research process can be
interrupted or hampered by interview biases. So, in this case, the researcher can reduce interview bias by sending
emails to the respondents by mail or host.
3. Gender anonymity- The researcher can ensure gender anonymity. If the questionnaire is sent by Mail or host.
So, the Researcher cannot know the detailed information of the respondents. Such as their name and position.
Respondents’ identities can be kept secret, or anonymity can be ensured if the questionnaire is sent via mail.
4. Accessibility- if the survey is conducted in a huge or large geographical area, respondents are diapered over
the whole geographical area, which could be the easiest process of sending the questionnaire to the respondents.
Disadvantage
1. It requires a very simple question. If the question is very complex, respondents cannot answer.
2. The researcher cannot control the respondents
3. Researchers cannot prove that the respondents may be low responders.
2. Self-Administered Questionnaire-
Mail question is also one type of self-administered question.
A self-administered question is a type of question that is filled out by the respondents themselves. Which does
not require the interviewee to fill out the questions.\
Advantage
1. Respondents can answer the questions without the influence of the researcher or interviewer.

23 | P a g e
2. Respondents can answer the question according to their own time and own convenience.
Disadvantage
1. The question should be very easy to understand. If the questionnaire is very Complex, respondents cannot
answer the question.
2. The researcher cannot clarify the questions. If the question requires more detailed answers, there is no scope
for raising any further questions.
3. Researchers do not have any control over the respondents.
3. Interviewer-administered questionnaire
Interviewer-administered questionnaire This is a type of questionnaire where interviewers ask questions to the
respondents through an interview.
Advantage
1. This is very convincing for the interviewers and the respondents.
2. If the respondents are not educated, if the question requires some explanation, this type of questionnaire is very
helpful for survey research.
Disadvantage
1. it might increase the cost of data collection.
If the interviewer is present or the researcher employs some data collector or interviewers for conduct. This type
of interview might increase the cost of data collection.
3. Internet or Online Questionnaire: Internet or Online Questionnaire is a type of questionnaire which is
fill out by the respondents by using internet technology.

# Second Note of Questionnaire (ChatGPT):

Definition of Questionnaire
A questionnaire is a structured instrument used for data collection, typically consisting of a set of questions
designed to elicit information from respondents. It plays a crucial role in survey research and is widely used in
various fields of social science.
According to Nadlin:
“A questionnaire is a device for securing answers to questions by using a form which the respondent fills in by
himself or herself.”
From this definition, we understand that a questionnaire:
• Contains a set of questions.
• Is filled out directly by the respondent.
• Serves the purpose of collecting specific information.

24 | P a g e
Types of Questionnaires
1. Self-Administered Questionnaire
• Definition: A type of questionnaire that is completed by the respondents themselves, without the
presence or assistance of an interviewer.
• Advantages:
o Ensures anonymity and privacy.
o Respondents can answer at their own pace and based on their own understanding and beliefs.
• Disadvantages:
o Respondents may misunderstand questions.
o Incorrect or incomplete responses are possible.
o Low response rate if distributed without follow-up.

2. Mail Questionnaire
• Definition: A questionnaire sent to respondents via postal mail or email.
• Advantages:
o Suitable for respondents who are geographically dispersed.
o Maintains privacy and anonymity.
• Disadvantages:
o Very low response rate.
o Respondents may ignore or discard the questionnaire.
o No assistance if the respondent finds a question unclear.

3. Interviewer-Administered Questionnaire
• Definition: A questionnaire where data is collected by an interviewer, who administers the questions
directly to the respondents.
• Advantages:
o The interviewer can explain unclear questions.
o Leads to a higher response rate.
o Allows for clarification and probing.
• Disadvantages:
25 | P a g e
 o Risk of interviewer bias—the interviewer might influence the responses.
o Lack of anonymity, which might affect honesty.

4. Mixed Questionnaire
• Definition: A questionnaire that includes both open-ended and close-ended questions.
• Purpose: To gain both quantitative data (from close-ended) and qualitative insights (from open-ended).
• Usefulness: Balances structure with flexibility, allowing for in-depth responses alongside measurable
data.

5. Web-Based Questionnaire
• Definition: A digital questionnaire administered online via web platforms, email links, or survey tools.
• Advantages:
o Cost-effective and easy to distribute.
o Can reach a large number of respondents quickly.
• Disadvantages:
o Excludes individuals without internet access.
o Risk of multiple submissions from the same person.

Types of Questions in a Questionnaire

1. Close-Ended Questions
• Definition: Questions with predefined response options.
• Advantages:
o Easy to quantify and analyze.
o Ensures uniformity in responses.
• Example:
Do you use public transportation?
o Yes
o No

2. Open-Ended Questions

26 | P a g e
 • Definition: Questions that allow respondents to answer in their own words.
• Advantages:
o Gathers in-depth information.
o Useful for opinion-based or exploratory research.
• Disadvantages:
o Harder to analyze and compare.
• Example:
What are your views on online education?

3. Contingency Questions
• Definition: Questions that are conditionally applicable based on the response to a previous question.
• Purpose: To filter respondents and avoid irrelevant questions.
• Example:
Q1: Are you married?
o Yes
o No
Q2 (If yes): How many children do you have?

4. Matrix Questions
• Definition: A set of questions presented in a table format, using the same set of answer options or rating
scale.
• Advantages:
o Reduces space and avoids repetition.
o Makes it easier for respondents to compare and rate multiple items.
• Example:
Statement Strongly Agree Agree Neutral Disagree Strongly Disagree
I enjoy using social media. ☐ ☐ ☐ ☐ ☐
Social media improves communication. ☐ ☐ ☐ ☐ ☐

27 | P a g e
9. Define data processing. What are the tools and techniques of
data processing?
Data Processing and Data Analysis
Data processing is a very important step. After collecting data, your task is to process it. Raw data is not suitable
for statistical analysis. Data processing means transforming raw data into a suitable format for statistical analysis.
The data that is collected must be processed or organized for analysis. This includes structuring the data as
required for the relevant analysis Tools.
It involves some steps.
1. Data Editing
Data editing is the process of checking responses to a survey, questionnaire, or other research instrument to
identify mistakes made either by the respondent or interviewer. The purpose is to control the quality of the
collected data. Data editing can be performed manually, with the assistance of a computer, or a combination of
both. This is the task of the lead researcher.
2. Coding
Data coding is the process of assigning numbers to the attributes of a variable. Code is simply a number that is
assigned to a particular attribute of a variable. Sex is a variable. If "1" is assigned to male, then 1 is the code of
male. Coding is necessary to quantify the variables. Computers cannot read qualitative aspects. It can only read
quantitative data. So, it is necessary to code the data to perform statistical analysis.
Concepts related to coding:
I. Coding Procedure: The Coding procedure is a set of rules that is developed by the researcher for
coding/assigning numbers to the variables.
II. Code book: The Coding book is basically important in survey. Coding book guides a researcher
on what to code and how to code. A codebook includes coding procedures and a list of variables
that need to be coded.
Rules of Coding:
● Coding should be consistent across the cases and units of analysis. If 1 is the code of females, 1
should be assigned to each and every female under your study.
● In the case of the ordinal scale, the higher score should receive a higher code. If education is an
ordinal variable, a higher order should receive a higher code. So, the University will receive a
higher code.
● In the case of nominal data, there are no hard and fast rules. Any variable can be assigned any code
as per the researcher’s wish.
3. Data Entry: After the coding, the first task is to input the data into a computer. Data entry means inputting the
data into a computer. Excel, SPSS, and SAS are suitable for data entry.
4. Data Cleaning: Data cleaning is the final check-up in the inputted data for any error, any duplication in the
input data. Data cleaning involves the detection and removal (or correction) of errors and inconsistencies in a data

28 | P a g e
set or database due to the corruption or inaccurate entry of the data. Incomplete, inaccurate or irrelevant data is
identified and then either replaced, modified, or deleted.
5. Classification- simply data
a. geographical location
b. time
c. based on a quantitative variable, sex/height/age/income
d. qualitative attributes. Religion/
6. Tabulation: Summarizing and presenting data in terms of columns and rows
Obj-simply data, compare data,
Important to data processing

10. Descriptive Statistics

Descriptive statistics are used to describe the basic features of the data in a study. Descriptive statistics help to
describe and understand the features of a specific data set by giving short summaries about the sample and
measures of the data. Are concerned with those statistical tools that are used to summarize, categorize, and
describe. E.g. co-relation, central tendency.
Types of Descriptive Statistics
There are 3 types of descriptive statistics. These are:
1. Univariate analysis is the analysis of one (“uni”) variable.
2. Bivariate analysis is the analysis of exactly two variables.
3. Multivariate analysis is the analysis of more than two variables.

1. Univariate Analysis
“Uni” means one. Univariate analysis is the analysis of a single variable. It deals with only one variable of the
sample data. There are different tools and methods here. These are discussed below:
I. Frequency Distribution: A frequency distribution is a list or table that displays the frequency of various
outcomes in a sample. Frequency distribution in statistics is a representation that displays the number of
observations within a given interval. In other words, Frequency Distribution is the arrangement of data
into several classes.
II. Percentage Distribution: Percentage distribution is one form of a frequency distribution in
which individual frequencies are shown as a percentage of the total frequencies.
III. Graphical presentation: Graphical representation refers to the use of charts and graphs to visually
display, analyze, clarify, and interpret numerical data, functions, and other qualitative structures. Graphical
representation is the use of intuitive charts to clearly visualize and simplify data sets. Graphs enable us to
study the cause-and-effect relationship between two variables. Graphs help to measure the extent of
29 | P a g e
 change in one variable when another variable changes by a certain amount. Graphs are also easy to
understand and eye-catching. There are several forms of graphical presentation as follows:
● Histogram: A Histogram is a type of graph of a joint rectangle that represents frequency and class interval.
A histogram is a non-cumulative frequency graph, it is drawn on a natural scale that represents frequencies
of the different classes and class intervals drawn close to each other. Histogram represents continuous
variable. Histogram is two dimensional.
● Bar diagram: Bar graphs are the pictorial representation of data (generally grouped) in the form of
vertical or horizontal rectangular bars. A bar diagram represents discrete variables. A bar diagram is one-
dimensional.
● Pie chart: A pie chart is a circular statistical graphic that is divided into slices to illustrate numerical
proportion. A pie chart represents a percentage distribution in a circular form. The total size of a pie chart
is 100%. Then, the whole chart is divided into parts based on the percentage of different groups.
IV. Central Tendency: In descriptive analysis, it’s also important to find out the Central (or average)
Tendency or response. Central tendency is measured with the use of three averages — mean, median, and
mode. The function of central tendency is to give us a central value.
The mean (Average) represents the sum of all values in a dataset divided by the total number of the values.
The median is the middle value in a dataset that is arranged in ascending order (from the smallest value to the
largest value).
The mode defines the most frequently occurring value in a dataset.
V. Dispersion: Dispersion refers to the spread of the values around the central tendency. There are two
common measures of dispersion: the range and the standard deviation. The range is simply the highest
value minus the lowest value. To get the standard deviation, we take the square root of the variance. Based
on standard deviation, we can examine the consistency of the values of some distributions. The less value
of standard deviation, the less variation among the distribution.
2. Bivariate analysis
Bivariate analysis means the analysis of bivariate data. Bivariate data is when you are studying two variables. It
is one of the simplest forms of statistical analysis, used to find out if there is a relationship between two sets of
values. It usually involves the variables X and Y.
● Correlation: It is used to examine the relation between two quantitative variables. We cannot apply this
to categorical data. When our data is ordinal, then we can use correlation.
● Lambda Test: When 2 variables are nominal, then we can use the lambda test to examine the relation
between two variables.
● Gamma Test: When 2 variables are ordinal, then we can use the gamma test to examine the relation
between two variables.
● Kendal’s Tau-B Test: This is similar to the gamma test, but it provides more refined output than the
gamma test.
3. Multivariate Analysis

30 | P a g e
Multivariate analysis is required when more than two variables have to be analyzed simultaneously. It is a
tremendously hard task for the human brain to visualize a relationship among 4 variables in a graph, and thus
multivariate analysis is used to study more complex sets of data.
• Multivariate Analysis of Variance: Multivariate analysis of variance (MANOVA) extends the analysis
of variance to cover cases where there is more than one dependent variable to be analyzed simultaneously;
see also Multivariate analysis of covariance (MANCOVA).
• Multiple Linear Regression: A linear regression method where the dependent variable Y is described by
a set of X independent variables. An example would be to determine the factors that predict the selling
price or value of an apartment.
• Multiple Linear Correlation: Allows for the determination of the strength of the linear relationship
between Y and a set of X variables.

11. What do you understand by inferential analysis? Explain the

uses of the chi-square test, t-test, z-test, and ANOVA in social
research.
Inferential Statistics
Descriptive statistics describes data (for example, a chart or graph), and inferential statistics allows you to make
predictions (“inferences”) from that data. With inferential statistics, you take data from samples and make
generalizations about a population. For example, you might stand in a mall and ask a sample of 100 people if they
like shopping at Sears. You could make a bar chart of yes or no answers (that would be descriptive statistics), or
you could use your research (and inferential statistics) to reason that around 75-80% of the population
(all shoppers in all malls) like shopping at Sears.
There are several ways to conduct inferential statistics, as given below:
1. Chi-Square Test: This is a hypothesis test. Chi-square is used to determine significant association
between 2 variables. This is not a correlation test. Rather, it is a hypothesis test. This measure is the
association between two categorical variables. E.g., Social Class (Independent) - Educational Level
(Dependent). These are categorical variables.
2. Significant Relationship: In research, we deal with sample data. Our objective is to predict population
characteristics based on sample data. If there exists any relation between two variables in the real world,
that is a significant Association.
3. Z-test/T-test:
These two are considered as similar in SPSS.
When we compare 2 groups to determine whether there is a difference between them, then we use these tests.
Z-tests are a statistical way of testing a hypothesis. If we have a sample size of less than 30 and do not know the
population variance, then we must use a t-test.

31 | P a g e
T-tests are a statistical way of testing a hypothesis when we do not know the population variance and our sample
size is small, n < 30.
4. One-way ANOVA:
Very similar to the t-test. In a test, we compare two groups. The one-way analysis of variance (ANOVA) is used
to determine whether there are any statistically significant differences between the means of three or more
independent (unrelated) groups by using one category or one variable.
For example, if we want to determine any significant difference of income pattern between peoples of different
religion, then we can use one-way anova test.
5. Two-way ANOVA:
In statistics, the two-way analysis of variance is an extension of the one-way ANOVA that examines the influence
of two different categorical independent variables on one continuous dependent variable.
6. Mann-Whitney Test:
The Mann-Whitney U test is used to compare differences between two independent groups when the dependent
variable is either ordinal or continuous but not normally distributed. For example, you could use the Mann-
Whitney U test to understand whether attitudes towards pay discrimination, where attitudes are measured on an
ordinal scale, differ based on gender (i.e., your dependent variable would be "attitudes towards pay
discrimination" and your independent variable would be "gender", which has two groups: "male" and "female").
Another example is, if we want to determine any difference of income pattern between male and female, if a
sample is not randomly drawn and not categorical and the sample size is small, then we can apply the Mann-
Whitney Test.
7. Kruskal-Wallis Test:
Very similar to one-way ANOVA. Anova is a parametric test. Kruskal-Wallis H is Non parametric test. The
Kruskal-Wallis H test (sometimes also called the "one-way ANOVA on ranks") is a rank-based nonparametric test
that can be used to determine if there are statistically significant differences between two or more groups of an
independent variable on a continuous or ordinal dependent variable.
8. Factor Analysis:
Factor analysis is a technique that is used to reduce a large number of variables into fewer factors. This technique
extracts the maximum common variance from all variables and puts them into a common score. As an index of
all variables, we can use this score for further analysis.

32 | P a g e