COER UNIVERSITY ROORKEE

Department of Artificial Intelligence & Machine Learning

Research Methodology
ASSIGNMENT-4

1. Define primary and secondary data with suitable examples?

Ans:- Primary Data:

Definition:

Primary data is the original data collected directly by a researcher or investigator for a
specific purpose or research project. It is first-hand data that has not been previously
collected or published.

Examples:
• Data collected from a survey conducted by a researcher to study
consumer preferences.

• Responses gathered from interviews or questionnaires.

• Observations made during an experiment in a lab.

• Measurements taken during field research.

Secondary Data:

Definition:

Secondary data is information that has already been collected, processed, and published by
someone else. It is used for analysis by researchers who were not involved in the original data
collection.

Examples:

• Data from government reports or census data.

• Information obtained from books, journals, or newspapers.

• Statistics from research articles or published studies.

• Historical records or databases available online.

Key Difference:

• Primary data is collected first-hand for the current research.

• Secondary data is pre-existing and originally collected for a different purpose.

2. Explain the different methods of primary data collection?

Ans:- Methods of Primary Data Collection:

Primary data is collected directly from the source for a specific research purpose. There
are several methods used to collect primary data, depending on the nature and objective
of the study. The main methods are:

1. Observation Method:

Description:
This method involves directly observing the behavior or events as they occur.

Types:

• Structured Observation: Uses a predefined set of rules or checklists.

• Unstructured Observation: More flexible and open-ended.

Example:
A researcher observes customer behavior in a retail store to understand purchasing habits.

2. Interview Method:

Description:

Data is collected by directly asking questions to the respondents, either face-to-face or

over the phone.

Types:

• Structured Interviews: Use pre-set questions.

• Unstructured Interviews: More conversational and flexible.

Example:
A journalist interviews people to understand public opinion about a new policy.
3. Questionnaire Method:

Description:

A set of written questions is provided to respondents to fill in their answers. It can

be conducted in person, by post, or online. Advantages:
Cost-effective and can cover a large audience.

Example:
An online survey sent to college students to collect feedback on teaching methods.

4. Survey Method:

Description:

A large-scale data collection method where a group of people is asked about their
opinions, behaviors, or characteristics.

Features:
Often combines questionnaires and interviews.

Example:
A national health survey conducted to assess the fitness level of the population.

5. Experimental Method:

Description:

Involves conducting experiments under controlled conditions to study cause-and-effect

relationships.

Example:
Testing the effect of a new fertilizer on crop yield in a laboratory or field setting.

6. Focus Group Discussion:

Description:

A small group of people discusses a topic under the guidance of a moderator. It is used to
gather detailed opinions and attitudes.

Example:
A company organizes a focus group to get feedback on a new product design.
3. Discuss the advantages and disadvantages of secondary data?

Ans:- Advantages of Secondary Data:

1. Saves Time:
The data is already collected, so you don't need to spend time gathering it.

2. Low Cost:
It is usually free or cheaper than collecting new data.

3. Easy to Find:
Available in books, websites, newspapers, and government reports.

4. Useful for Comparison:

Helps to compare past and present trends.

Disadvantages of Secondary Data:

1. May Be Old:
The data might not be up to date.

2. Not Exact for Your Topic:

It may not fully match what you are researching.

3. Not Always Reliable:

You may not know how it was collected or if it's correct.

4. Missing Information:
Some important details might not be included.

4. Compare and contrast observation and interview methods?

Ans:-
Point of Observation Method Interview Method
Comparison
Collecting data by watching Collecting data by asking
Definition
people or events. questions directly.

Mainly behavioral and nonverbal

Data Type Verbal and opinion-based data.
data.
 No direct interaction with Involves direct communication
Interaction
participants. with participants.

Nature of Data Objective (what is seen). Subjective (what is said).

Studying real-time actions and Understanding thoughts,

Best Suited For
habits. feelings, and opinions.

Checklists, cameras, or field Questionnaires, audio

Tools Used
notes. recorders, or written notes.

Usually time-consuming, may Can be quick, but might need

Time and Cost
require fewer resources. more preparation.
Less biased (if observer is Can be biased due to the
Accuracy neutral). interviewer's influence.

5. What are the different sources of secondary data?

Ans:- Sources of Secondary Data
Secondary data comes from information that has already been collected and published by
someone else. These sources can be classified into two main types: internal and external
sources.

1. Internal Sources:

These are data sources found within an organization or institution. Examples:

• Company sales records

• Financial statements

• Customer databases

• Employee records

• Previous internal research reports

2. External Sources:
These are data sources found outside the organization and are publicly or
commercially available. Examples:

a. Government Publications:
• Census data

• Economic surveys

• Labour and employment statistics • Reports by ministries and departments

b. Books and Journals:

• Academic books

• Research journals

• Newspapers and magazines

c. Reports and Publications from Institutions:

• Reports by WHO, IMF, World Bank, etc.

• Research by universities and think tanks

d. Websites and Online Databases:

• Government portals (e.g., data.gov.in)

• Educational and research databases (e.g., JSTOR, Google Scholar)

e. Media Sources:

• News channels

• Documentaries

• Podcasts

f. Commercial Sources:

• Market research firms (e.g., Nielsen, Statista)

• Trade associations and industry reports

6. Explain the concept of measurement and scaling?

Ans:- 1. Measurement:

Definition:

Measurement is the process of assigning numbers or symbols to objects, events,

or phenomena based on a set of rules. This process allows researchers to quantify
the attributes or characteristics of these objects for analysis.
Purpose of Measurement:

• To quantify variables that cannot be directly observed or measured.

• To ensure consistency in data collection and analysis.

• To assign a numerical value to characteristics like height, weight, satisfaction,
or opinion.

Example:

• Measuring the height of a person in centimeters.

• Measuring temperature in Celsius or Fahrenheit.

Measurement is the foundation of data collection because it turns subjective qualities (like
opinions or attitudes) into quantifiable values that can be analyzed statistically.

2. Scaling:

Definition:

Scaling is a technique used to assign numbers or values to represent the intensity,

magnitude, or level of a particular attribute or variable. It involves constructing a scale
that assigns values or categories that help in understanding the extent or degree of a
characteristic.
Purpose of Scaling:

• To represent different levels of a variable or attribute.

• To create a structured approach to measuring things that cannot be directly
observed, such as attitudes, preferences, or satisfaction.
• To allow for comparisons and rankings between different individuals, groups, or items.

Types of Measurement Scales:

There are four primary types of scales used in measurement:

1. Nominal Scale:
• Definition: The simplest scale, used for categorizing data without any order or
ranking.

• Example:
 o Gender (Male/Female) o

Eye color (Blue, Brown,

Green) o Blood type (A, B, AB,

O)

2. Ordinal Scale:
• Definition: This scale ranks the data in a specific order but does not show the exact
difference between ranks.

• Example:

o Ranking in a race (1st, 2nd, 3rd) o Education

levels (High school, Bachelor's, Master's)

3. Interval Scale:
• Definition: This scale not only ranks data but also has equal intervals between
values. However, it does not have an absolute zero.

• Example:

o Temperature in Celsius or Fahrenheit (0°C does not mean no temperature).

o Calendar years (Year 2000 is not "twice as much" as Year 1000).

4. Ratio Scale:
• Definition: The ratio scale is the most advanced scale, with equal intervals and an
absolute zero, meaning zero indicates the complete absence of the attribute being
measured. • Example:

o Weight (0 kg means no weight)

o Height (0 cm means no height)

o Income (0 dollars means no

income)
7. Describe the four levels of measurement with suitable examples.
Ans:- 1. Nominal Scale (Categorical Data) Definition:

The Nominal scale is used for categorizing data into different groups with no order.

Examples:

• Gender: Male, Female

• Blood Type: A, B, AB, O • Colors: Red, Blue, Green Operations:

• You can count how many items are in each category.

• Only the mode (most common category) can be calculated.

2. Ordinal Scale (Ordered Data)

Definition:

The Ordinal scale ranks data in a specific order, but we don’t know the exact
difference between the ranks.

Examples:
• Race Positions: 1st, 2nd, 3rd (We know 1st is better than 2nd, but not how much
better).

• Satisfaction Levels: Very Dissatisfied, Dissatisfied, Neutral, Satisfied, Very Satisfied

Operations:

• You can calculate mode and median (middle value).

• Mean (average) doesn't work because the differences between ranks are not equal.

3. Interval Scale (Ordered with Equal Intervals)

Definition:

The Interval scale orders data with equal differences between values, but zero
doesn’t mean "none." Examples:
• Temperature in Celsius or Fahrenheit: The difference between 10°C and 20°C is the
same as between 30°C and 40°C, but 0°C does not mean "no temperature."
 • IQ Scores: The difference between scores of 100 and 110 is the same as between 110
and 120, but zero doesn’t mean "no intelligence." Operations:

• You can calculate mode, median, and mean.

• Addition and subtraction work, but multiplication and division do not.

4. Ratio Scale (Ordered with Equal Intervals and True Zero)

Definition:

The Ratio scale is the highest level, with equal intervals and a true zero that means
"none."

Examples:
• Height: 0 cm means no height, and someone who is 180 cm tall is 1.8 times taller than
someone who is 100 cm tall.
• Weight: 0 kg means no weight, and someone who weighs 50 kg is twice as heavy as
someone who weighs 25 kg.

Operations:
• You can calculate mode, median, mean, and perform all math operations (addition,
subtraction, multiplication, and division).

Summary of the Four Levels of Measurement:

Level What it does Example

Nominal Categories with no order Gender, Blood Type

Level What it does Example

Categories with order (but unequal spacing) Race Position, Satisfaction

Ordinal
Level

Ordered with equal intervals, no true zero Temperature

Interval
(Celsius/Fahrenheit)
Ordered with equal intervals and true zero Height, Weight, Income
Ratio

8. Explain the process of designing a questionnaire. ?

Ans:- 1. Define the Purpose:
First, know why you are creating the questionnaire. What information do you want
to gather? This will help you create the right questions. Example:
If you want to know how people feel about a new product, your purpose might be to
understand their satisfaction with it.

2. Know Your Audience:

Think about who will answer your questions. Are they students, employees, or customers?
This will guide how you phrase your questions. Example:
For teenagers, you’d use simple language. For professionals, you might use more
technical terms.

3. Choose the Types of Questions:

You can use two types of questions:

• Closed-ended: Fixed answers (e.g., Yes/No, Multiple Choice).

• Open-ended: Respondents can answer in their own words.

Example:

• Closed-ended: “Do you like this product? Yes/No.”

• Open-ended: “What do you like about the product?”

4. Organize Questions:

Put questions in a logical order. Start with simple questions, then move to harder ones.
Group similar questions together.

Example:
First ask about general info (like age), then ask about product satisfaction.

5. Keep Questions Simple:

Make sure each question is clear and easy to understand. Avoid complex language
or questions that might confuse people. Example:
Instead of asking, “How do you rate the superior quality of our service?” just ask, “How
would you rate our service?”

6. Use Rating Scales:

If you want to know how much someone agrees or likes something, use a scale (like 1 to
5, where 1 means bad and 5 means great).

Example:
“Rate your satisfaction with the product (1 = Very unsatisfied, 5 = Very satisfied).”

7. Test Your Questionnaire:

Before sending it out to everyone, try it with a small group. Ask them if the questions are
clear and easy to answer.

8. Finalize and Distribute:

Once you’re happy with the questions, send the questionnaire to your audience. You can
share it online, via email, or in person.

9. Follow Up:

If you don’t get enough responses, send a reminder to people to fill out the questionnaire.

9. Discuss the importance of data collection in research. ?

Ans:- 1. Provides Evidence for Analysis:

Data collection gives you the information you need to analyze and draw conclusions.
Without data, you can’t prove your ideas or theories.

Example:
If you're studying customer satisfaction, you need data to see if customers are happy with
your product.

2. Tests Hypotheses:
Data helps you test your ideas or predictions (hypotheses). It shows whether
your assumptions are correct or not. Example:

If you think more advertising will increase sales, collecting data will help you see if that's
true.

3. Helps in Decision-Making:
Data helps businesses, researchers, and organizations make better decisions based on facts
rather than guesses. Example:
A company might collect data on customer feedback to decide whether to change their
product or service.

4. Ensures Accuracy and Consistency:

Good data collection ensures your research is accurate and the results can be trusted. It
makes the research stronger. Example:
If you collect data from many sources or use different methods, your findings will be
more reliable.

5. Makes It Possible to Generalize:

By collecting data from a group (sample), researchers can make guesses or
conclusions about a larger group (population). Example:
If you ask 1,000 people about their shopping habits, you can guess what most customers
might do.

6. Finds Patterns and Trends:

Collecting data over time helps you notice trends or patterns that you might miss right
away.

Example:
Tracking sales for a few years could show that sales are higher in certain seasons, helping
the company plan better.
7. Legal and Ethical Responsibility:
In many fields, collecting data the right way is required by law and ethics.
Researchers must ensure privacy and follow rules. Example:
 In medical research, it's important to collect data properly to ensure treatments are safe
and effective.

8. Ensures Fairness:
Data collection helps make sure research results are based on facts and not
personal opinions or biases. Example:
To test a new drug, data needs to be collected in an unbiased way to ensure the results are
valid.

10. Write short notes on:

● a) Likert scale

● b) Focus group method

Ans:- a) Likert Scale
The Likert scale is a common tool used in surveys and questionnaires to measure attitudes,
opinions, or behaviors. It asks respondents to indicate their level of agreement or
disagreement with a statement. The scale usually ranges from 1 to 5 or 1 to 7, where
the numbers represent different levels of agreement.

Example:
A typical Likert scale question might be:
"How satisfied are you with our customer service?"

1. Very Dissatisfied

2. Dissatisfied

3. Neutral

4. Satisfied

5. Very Satisfied

Advantages:

• Easy to administer and understand.

• Provides clear data for statistical analysis.

• Can measure attitudes on a continuum (e.g., from strongly agree to strongly
disagree).
 b) Focus Group Method
A focus group is a qualitative research method where a small group of people (usually 610) are
brought together to discuss a specific topic or product. The discussion is guided by a moderator
who asks open-ended questions to encourage conversation and gather in- depth insights.

Key Features:

• Participants are selected based on specific criteria related to the study.

• Discussions are usually recorded and analyzed for themes, patterns, and
insights.

• Focus groups are often used to gather opinions, experiences, and

perceptions.

Advantages:

• Provides rich, detailed insights.

• Encourages participants to build on each other’s responses.

• Useful for exploring new ideas or understanding deeper motivations.

Example:

A company might use a focus group to discuss customer reactions to a new product before its
official launch.