Unit-II

Data Collection: Primary data and secondary sources of data, Qualitative v/s
Quantitative data, methods of data collection

Research methods: Field study, Lab study, Survey method, Observational

Method, Existing data based research, Longitudinal Studies, Panel studies

Data Collection:

For any research study data collection using appropriate method is most
important activity, The data can be of two types as;

➢ Primary data
➢ Secondary data

Primary Data :

➢ Primary data is the data which is collected firsthand by the researcher

specifically for the purpose of the study at hand.
➢ It involves the direct collection of information from original sources.
➢ Primary data is considered more reliable and relevant to the research
objective because it has not been previously used or interpreted by
anyone else.

Methods of collecting primary data:

Surveys and Questionnaires: Researchers can design surveys questionnaires to

collect information from a selected group of respondents. Survey can be
conducted in person, over the phone, through email, or online platforms.

1
Interviews: Interviews can be conducted in a structured or unstructured manner,
where researchers ask questions directly to individuals or groups to obtain
detailed insights and perspectives.

Observations: Researchers observe and record behaviors, events, or

phenomena in their natural setting. This method can be particularly useful in
sociological, psychological, or anthropological research.

Focus Groups: In focus groups, a moderator facilitates a discussion among a

small group of participants who share their opinions, experiences, and attitudes
on a particular topic. The group interaction can generate valuable insights about
subject of discussion .

Experiments: Experimental research involves manipulating variables and

observing the effects in a laboratory. Researchers control and manipulate
independent variables to determine their impact on dependent variables. These
methods are useful in scientific research

Document Analysis: Researchers analyse existing documents, such as reports,

public records, diaries, letters, or archival material, to gather primary data
relevant to their research question.

Secondary Data:

Unlike primary data, which is generated by a researcher for their specific study,
secondary data has already been generated by other researchers, organizations,
government agencies, or other sources. This pre-existing data can be analyzed
and interpreted to address new research questions or to support and enhance
primary research findings.

2
Sources of secondary data:

Published sources: Already published information in academic journals, books,

reports, newspapers, and magazines contains data from previous studies or
surveys which can be used in other research.

Government and organizational data: Many governments and institutions

conduct various surveys, censuses, and data-gathering activities for different
purposes. This data is often available in public domain, government or
organisation’s websites and can be used by researchers.

Online databases and repositories: Online platforms are generating lot of data
which is stored in , numerous data repositories and data warehouses, offering a
wealth of information across various fields.

Research data from other studies: Sometimes, researchers may share their raw
data or datasets for the sake of transparency or to encourage others to build
upon their findings.

Advantages and Disadvantages of Secondary Data:

Advantages:

Secondary data is advantageous for a variety of reasons:

cost-effectiveness,

time-saving,

feasibility, or when the required data is not easily obtainable through primary
research.
3
Disadvantages:

However, it's crucial to ensure the quality, relevance, and reliability of the
secondary data, as it may not always perfectly suit or align with the researcher's
specific needs or research objectives.

Proper citation and acknowledgment of the original sources should be ensured

while using secondary data to maintain academic integrity and avoid plagiarism.

Types of Data : Qualitative and Quantitative data

There are two primary types of data used in research and analysis. These have
different characteristics, collection methods, and analysis techniques as
described below

Properties of Qualitative Quantitative

data
Nature descriptive in nature and Quantitative or numerical in
deals with qualities of nature Observations are in
attributes that cannot be terms of measurable quantities
easily measured or of variables. It is objective,
quantified. quantifiable information.
Methods of Collected through collected through structured
Data interviews, focus groups, methods, such as surveys with
Collection: observations, open ended closed-ended questions, lab
questionnaire surveys, or experiments, or numerical
content analysis of text, measurements
audio, or video

4
Data Format In the form of narratives, Data is collected in form of
quotes, themes, or patterns, numbers, and presented in the
or qualitative terms like form of tables , charts or
good, bad, tall, short, high , graphs, making it suitable for
low etc allowing mathematical and statistical
researchers to gain a deeper analysis
understanding of the
subject matter
Data Analysis involves interpreting the data involves applying
meanings, themes, and statistical or mathematical
patterns found within the techniques to summarize,
data. Researchers may use compare, and draw conclusions
techniques like thematic from the data. In business
analysis, content analysis, or research common analysis is to
narrative analysis. find mean, median, variance,
correlation regression, and
hypothesis testing etc .

Examples Responses to open-ended Daily, weekly or monthly Sales

survey questions about analysis, Marks analysis of
feelings or experiences, students , Profit and loss
interview transcripts analysis, market trend analysis
exploring opinions or etc
beliefs, observational notes

5
 about behaviours or
interactions.

In research, both qualitative and quantitative data have their unique advantages
and purposes.

Qualitative data is valuable for exploring complex phenomena, gaining in-depth

insights, and understanding the context.

Quantitative data is useful for trend analysis, testing hypotheses, making

predictions, and generalizing findings to a larger population. In some cases,
researchers may use both types of data to complement each other and provide
a comprehensive understanding of the research topic.

Methods of Data Collection :

There are various methods of data collection, and researchers often choose the
most appropriate ones depending on their research goals, available resources,
and the nature of the data they want to collect. Here are some common methods
of research data collection:

Surveys: Surveys involve asking respondents a set of standardized questions to

gather information about their opinions, attitudes, behaviors, or demographics.
Surveys can be conducted in person, over the phone, through mail, or online
google forms or on printed forms.

Interviews: Interviews are one-on-one interactions with participants to gather

in-depth information about their experiences, beliefs, or knowledge. Interviews
can be structured with (predetermined set of questions), semi-structured (a mix

6
of predetermined and open-ended questions), or unstructured (open-ended
questions only).

Observations: Observational research involves watching and recording the

behavior of individuals or groups in natural or controlled settings. It can be
participant observation in which the researcher is actively involved or non-
participant observation where the researcher remains only an observer.

Experiments: Experiments involve manipulation of variables and observing their

effects on participants to establish cause-and-effect relationships. They are often
conducted in controlled settings to ensure the validity of the results.
Experiments are normally conducted in scientific labs or on agricultural fields

Case Studies: Case studies involve in-depth data analysis of an individual, group,
or situation. Researchers use various data sources, such as interviews,
observations, and documents, to gain a comprehensive understanding of the
subject. Secondary data can also used in case studies

Content Analysis: Content analysis is used to systematically analyze textual,

visual, or audio content. Researchers identify themes, patterns, or trends in the
data, by applying coding schemes to categorize information.

Focus Groups: When a small group of focussed participants is collected discuss

a specific topic. The researcher facilitates the discussion to gain insights into
participants' perspectives and experiences.

Secondary Data Source: Researchers uses secondary data which have already
been collected for other purposes. This can include government reports, publicly
7
available datasets, or data collected by other researchers, data available on
online platforms

Action Research: Action research involves researchers collaborating with

participants to identify and address problems in real-world settings. Data
collection methods in action research can vary depending on the specific project.

Ethnography: When a researchers spend extended periods in the field or

community , observing and interacting with participants to gain deep insights
into their behaviors, beliefs and culture.

Online Analytics: In online analytics data is collected from online sources, such
as social media platforms, website traffic, or user interactions, to understand
online behavior and trends.

Some times researchers can use a combination of these methods to triangulate

data and enhance the validity and reliability of their findings. The choice of data
collection methods depends on the research objectives, the available resources,
ethical considerations, and the nature of the study population.

Sampling:

When it is not possible to study the entire population and analyze its data, a
small sample is taken form the population for study. It is assumed that
characteristics of the sample represent the characteristics of the entire
population

Thus sampling is a crucial aspect of research methodology that involves selecting

a subset of individuals, items, or elements from a larger population to study and
draw inferences about the entire population.
8
The process of selecting this subset is known as the sampling method, and the
selected subset is referred to as the sample. Proper sampling is essential to
ensure that the results of the study are representative of the population and
applicable to the entire population from which the sample was drawn.

There are various sampling methods, the most appropriate one based on the
research objectives, the nature of the population, available resources, and the
desired level of precision.

Sampling Methods :

Random Sampling: In random sampling, each member of the population has an

equal chance of being selected for the sample. This method ensures that the
sample is unbiased and representative of the population, allowing for
generalization of the findings.

Example: Asking questions randomly in the market without any bias from any
one about any topic

Stratified Sampling: In stratified sampling population is into subgroups or strata

based on specific characteristics (e.g., age, gender, location). Then, a random
sample is taken from each stratum in proportion to its representation in the
overall population. This method ensures representation from various subgroups
and can increase the precision of estimates for each stratum.

Example: 1. Same question can be asked to 20 males, 20 females

2. Data collected from age groups 20 to 30, 30 to 40, 40 to 50

9
Systematic Sampling: In systematic sampling, researchers select every nth
individual from a list of the population. The starting point is chosen randomly,
and then every nth element is taken in the sample. This method is efficient and
straightforward but may introduce some bias if there is a periodic pattern in the
list.

Cluster Sampling: Cluster sampling involves dividing the population into clusters
(e.g., geographical areas or institutions) and randomly selecting some clusters to
include in the sample. Researchers then collect data from all or a random subset
of elements within the selected clusters. This method is useful when it is
impractical to obtain a list of all individuals in the population.

Example: Questions can be asked in 10 metropolitan cities (Each is a cluster)

Convenience Sampling: In convenience sampling participants who are readily

available or easily accessible are selected. While this method is convenient, it
may introduce bias because it may not represent the entire population
accurately.

Purposive Sampling: Purposive sampling, also known as judgmental or selective

sampling, involves handpicking participants based on specific criteria, expertise,
or knowledge relevant to the research question. This method is common in
qualitative research or when studying a specific group.

Snowball Sampling: Snowball sampling is used study hard-to-reach or hidden

populations. The researcher starts with one or a few participants and then asks

10
them to refer other potential participants, who, in turn, refer more participants,
creating a "snowball" effect.

The choice of sampling method should be guided by the research goals, the
nature of the population being studied, and considerations of feasibility and
resources available.

What should be Sampling Size ?:

The optimal sample size of observation depends on various factors, i.e research
objectives, the nature of the population, the sampling method, the desired level
of precision, and the available resources. While there is no unique one-size-fits-
all answer to what the sample size should be, but here are some guidelines and
considerations to help determine an adequate sample size:

Population Size: The larger the population size, the larger the sample size
needed to achieve a representative sample.

Confidence Level: The confidence level refers to the level of certainty desired in
the results. The achieve higher confidence level, the larger the sample size
required.

Margin of Error: The margin of error indicates the degree of precision needed in
the estimates. A lower margin of error requires a larger sample size.

Variability of the Population: For homogeneous population, a smaller sample

size may be sufficient. However, if the population is diverse, a larger sample size
is needed to capture that variability.

11
Cost and Resources: Practical limitations of time, budget, and available
resources, may influence the sample size. There must be a balance in the the
desired sample and the resources available.

Study Design and Analysis: The complexity of the study design and analysis also
influence the sample size. Certain statistical tests and analyses may require
larger samples to achieve valid results.

Type of Research: Different types of research (e.g., qualitative, quantitative or

experimental) may have different considerations to determine sample size. For
qualitative research, data saturation may be a consideration.

Ethical Considerations: In human research, researchers need to ensure that the

sample size is large enough to produce meaningful results while respecting the
rights and well-being of participants.

The goal of determining the appropriate sample size is to strike a balance

between obtaining meaningful results and practical constraints, ensuring that
the study is well-powered to draw reliable conclusions from the data collected.

Research Methods:

Field Study

Field Research is a method of collecting qualitative data with the aim to

understand, observe, and interact with people in their natural setting. It
requires specialized market research tools. Field research is used most in the

12
field of social science, such as anthropology and health care professions, as in
these fields it is vital to create a bridge between theory and practice.

There are 4 main methods of conducting field research, and they are as follows:

➢ Ethnography,
➢ Qualitative Interviews,
➢ Direct Observation &
➢ Participants Observation

Ethnography:

Ethnography is a kind of fieldwork with aims to record and analyse a particular

culture, society, or community. This method is used in social anthropology, and
it usually involves the complete involvement of an anthropologist in the culture
and everyday life of the community they are trying to study.

Ethnographers spend significant time living and interacting with the community
which they are studying. They become active participants in the daily life of the
group, participating in rituals, ceremonies, and other activities to understand
the culture as a whole rather than focusing on isolated aspects. They examine
various dimensions, including beliefs, values, traditions, social structures,
language, rituals, and practices

Ethnography involves conducting research in the field (the community or

cultural setting under study) rather than a controlled laboratory environment.
Fieldwork can last from a few months to several years.

Qualitative Interviews:

13
Qualitative interviews are a valuable research method for gathering in-depth
insights and understanding human experiences, behaviors, and perspectives.
There are several types of qualitative interviews, each with its own specific
characteristics and purposes. Some of the common types of qualitative
interviews include:

Structured Interviews:

Structured interviews involve a predetermined set of standardized questions

asked to all participants. The questions are typically closed-ended or have
limited predefined response options. This format allows for efficient data
collection and straightforward data analysis. However, it may not capture the
richness and complexity of participants' responses as much as other interview
types.

Semi-Structured Interviews:

Also known as conversational interviews, semi-structured interviews use a

flexible yet guided approach. The interviewer has a general outline of topics to
cover but probe the interviewee with questions which are further generated
depending on the participant's responses. This type of interview allows for in-
depth exploration while maintaining some structure.

Open-Ended Interviews:

In open-ended interviews researcher ask broad, non-directive questions that

give participants the freedom to respond in detail and share their thoughts,
experiences, and feelings. This type of interview is particularly useful for
exploring complex or sensitive topics and gaining rich, narrative data.

14
Unstructured Interviews:

Unstructured interviews are the most flexible type, where there is no

predetermined set of questions or specific agenda. The interviewer engages in
a free-flowing conversation with the participant, allowing them to discuss topics
of importance to them. Unstructured interviews require skilled interviewers to
maintain focus and extract meaningful insights from the unscripted responses.

Group Interviews (Focus Groups):

In group interviews several participants discuss a topic together, often guided

by a facilitator. Focus groups are a common form of group interviews used to
gather diverse perspectives and stimulate group discussions. Group interviews
can be especially helpful for understanding social dynamics, collective opinions,
and shared experiences.

Narrative Interviews:

In narrative interviews participant share his life stories or experiences, narrating

events chronologically or provide a detailed account of significant moments.
This approach emphasizes the participants' individual perspectives and allows
for a deeper understanding of the context and meaning behind their
experiences.

Direct Observation:

In direct observation method, researchers systematically observe and record

behaviors, events, or phenomena as they naturally occur in real-life settings.
15
Data is collected through firsthand observation rather than relying on self-
reports or secondary sources. Direct observation is commonly used in various
disciplines, including psychology, anthropology, sociology, education, and
environmental studies, to study human behavior, interactions, and natural
processes.

Some of the key features of the direct observation method are:

Natural Setting: Observations take place in the natural environment where the
behaviors or events occur, without the intrusion of artificial conditions or
experimental manipulation.

Objective and Unobtrusive: Observers strive to remain objective and impartial

and minimize their influence on the observed subjects to avoid altering their
behavior (i.e., observer bias). Often the subjects may not be aware that they
are being observed.

Systematic Recording: Researchers follow a structured plan for observation,

specifying what is to be observe, how to record data, and the criteria for
identifying and categorizing behaviors or events.

Quantitative and Qualitative Data: Direct observation can yield both

quantitative data (e.g., frequency, duration of behaviors) and qualitative data
(e.g., descriptions, contextual information).

Event or Time Sampling : Researchers may use event sampling to record specific
behaviors or events of interest when these occur, or use time sampling, where
observations are made at predetermined intervals and record the presence or
absence of certain behaviors.

16
Ethical Considerations: Ethical guidelines must be followed when observing
human subjects, ensuring informed consent, confidentiality, and protection of
privacy.

Real-Time Data: Observations provide data in real-time, capturing the natural

context and sequence of behaviors or events.

Direct observation is particularly useful when studying behaviors that might be

subject to bias or distortion in self-reporting or when researchers want to gain
a more accurate and detailed understanding of a phenomenon in its natural
context.

For example, researchers might use direct observation to study children's play
behavior, animal interactions in the wild, classroom dynamics, or consumer
behavior in a retail setting.

However, direct observation can be time-consuming, requiring trained

observers to avoid errors in data collection and interpretation. The presence of
observers might influence the behavior of those being observed (the Hawthorne
effect).

Participant’s Observation:

Participant observation is qualitative research used in various fields, such as

anthropology, sociology, and ethnography, to study people's behaviour and
social interactions within their natural settings.

In this method researchers immerse themselves in the social context of the

participants they are studying, actively observing and sometimes participating

17
in their activities, and recording detailed notes about their experiences. The
participant observation method normally works in following Steps:

Step1: Identification of Research Site

The researcher chooses a specific location or community for study. This could
be a village, a workplace, a school, a religious gathering, or any other social
setting relevant to the research question

Step 2 : Building Relationships

Researcher try to gain access to the research site and obtain the trust and
cooperation of the participants by building relationships with the community or
key informants who can introduce the researcher and vouch for their intentions

Step 3: Immersion and Observation

Once access is granted, the researcher spends an extended period in the field,
typically weeks or even months, to observe the participants' daily activities,
interactions, and behaviors. The researcher aims to be as objective and
unobtrusive as possible to minimize the observer effect and maintain the
naturalness of the setting.

Step 4: Recording Observations

During the observation, the researcher takes extensive notes, often using a
journal or field notebook, to record their observations, thoughts, and reflections
on what they are witnessing. These notes are used for analyzing the data later

18
Step 5: Participant Roles

In some cases, the researcher may also take on a participant’s roles by actively
participating in certain activities alongside the subjects. This can provide
additional insights and build stronger rapport with the participants.

Step 6: Clarification and Follow-up

Throughout the process, the researcher may ask questions to clarify their
understanding of the participants' behaviors or seek additional information
from key informants.

Step 7: Data Analysis:

After the observation period, the researcher analyse the collected data. This
analysis involves reviewing and categorizing the field notes to identify patterns,
themes, and significant behaviors. Data may also include photographs, videos,
or audio recordings.

Step 8: Research Writing

The final step is the interpretation of the findings and the reporting of the
research results. Researchers often use quotes, anecdotes, and detailed
descriptions to illustrate their conclusions. The research report may also include
the researcher's reflections on their own role in shaping the observations and
interpretations.

19
Participant observation allows researchers to gain a deep understanding of the
participants' perspectives and the social context, providing rich and contextually
embedded data. But it is a time-consuming and resource-intensive method, and
the researcher's presence can still influence the participants' behaviors to some
extent.

Secondary Research:

Secondary Research also known as desk research or literature review or existing

data based research that involves gathering and analyzing existing data,
information, and sources that have been previously collected and published by
others. Instead of collecting data directly from participants or through primary
sources, researchers rely on the work of others to answer their research
questions or investigate a particular topic of interest.

Key features of secondary research:

Sources of Data:

Secondary research data is sourced from academic journals, books, government

reports, market research reports, statistical databases, historical records, and
other published materials. The internet has significantly expanded access to
secondary data, making it easier to find relevant information. Unlike primary
research, secondary research relies on the analysis of data that has already been
collected by others. This could include qualitative data (e.g., interviews, focus
groups) and quantitative data (e.g., survey results, statistical data).

Purpose:

20
Secondary research is conducted for various purposes, such as to gain a
comprehensive understanding of a topic, support or refute hypotheses, provide
context for primary research findings, identify gaps in existing knowledge, or
explore trends and patterns over time.

Advantages of Secondary Data:

1. Cost and Time-Effectiveness: Secondary research is generally quicker and

more cost-effective compared to primary research since data is already
available.
2. Large Sample Sizes: Secondary data often comes from large-scale studies,
providing researchers with access to significant sample sizes.
3. Historical and Longitudinal Analysis: Researchers can analyze data
spanning different time periods, allowing for historical and longitudinal
studies.

Limitations:

Data Availability: The availability and quality of secondary data may vary,
and researchers may encounter limitations in the scope or relevance of the
available information.

1. Lack of Control: Since the data is collected by others, researchers may

have no control over the methods used, the types of data available, or the
quality of data collection.
2. Bias and Validity: Researchers must critically evaluate the sources of
secondary data to ensure their credibility and potential biases.
3. Ethical Considerations: When using secondary data, researchers should
give proper credit to the original sources and ensure that the data is used
21
 in an ethical manner and mentioned the references in the research
writing.

Secondary research is a valuable approach for gaining insights into a subject

without the need for extensive data collection, it can strengthen the depth and
breadth of research findings When combined with primary research.

Longitudinal Studies:

Longitudinal studies are characterized by their time dimension. Longitudinal

studies are a type of research design to investigate changes and developments
in individuals, groups, or populations over a period of time. Unlike cross-
sectional studies which provide a snapshot of data at a single point in time,
longitudinal studies involve the collection of data at multiple time points,
allowing researchers to observe and analyze trends, patterns, and variations
over time. Data is collected at multiple time points, which could be days, weeks,
months, or even years apart, depending on the research design and objectives.

Data in longitudinal studies can be collected through various methods, including

surveys, interviews, observations, medical tests, and assessments. The same set
of variables is typically measured at each time point to facilitate comparison and
analysis.

Longitudinal studies are valuable for understanding processes of development,

aging, disease progression, educational trajectories, and the long-term effects
of interventions or policy changes. Their ability to capture temporal changes and
individual variations makes them a powerful tool in many areas of research.

Types of Longitudinal Studies:

22
Trend Studies:

These studies involve observing changes in a single group or population over

time. Measurements are made at the same variables at different time points to
identify trends and patterns.

Cohort Studies:

Cohort is a group of individual with a common characteristic. Cohort studies

follow specific groups of individuals, cohorts, over time. Researchers study how
different cohorts experience changes and developments over their lifetimes.

Panel Studies:

Panel studies, also known as panel surveys or longitudinal panel studies, are a
specific type of longitudinal research design that involves repeatedly collecting
data from the same group of individuals, known as the panel, at multiple time
points. The key characteristic of panel studies is the focus on tracking and
observing changes and developments within the same set of participants over
an extended period.

Some of the key features of Panel studies are:

Repeated measures: In panel studies data is collected from the same individuals
at regular intervals (e.g., annually, biennially, or monthly) over an extended
period. These repeated measures allow researchers to observe changes and
trends within each participant and in the group as a whole.

23
Sample selection: In this type of research , the initial sample of participants is
carefully selected based on specific criteria relevant to the research question.
The panel is typically designed to be representative of a larger population or
specific target group.

Data collection: Panel studies use various data collection methods, including
surveys, interviews, observations, medical tests, and other assessments.
Consistency in the data collection instruments is crucial to ensure comparability
across different time points.

Longitudinal Analysis: The primary focus of panel studies is often on

longitudinal analysis, which involves examining how individual participants
change over time and how different variables relate to each other across time.

Types of Panel Studies:

True Panel: In a true panel study, the same group of individuals is followed and
surveyed throughout the entire duration of the study. This provides the most
comprehensive longitudinal data.

Replenishment Panel: In replenishment panel studies, new participants are

added to the panel over time to replace those who drop out. This approach helps
maintain a consistent sample size and account for attrition.

Advantages:

In-Depth Analysis: Panel studies allow for in-depth exploration of individual

changes and trajectories over time.

24
Causality Inference: By observing changes within individuals over time,
researchers can make more robust inferences about causality.

Long-Term Trends: Panel studies are well-suited for studying long-term trends
and developments.

Challenges:

Participant Attrition: Panel retention can be challenging, as some participants

may drop out or become unavailable over the study's duration.

Time and Resources: Panel studies are resource-intensive and require long-term
commitment in terms of time, funding, and personnel.

Sample Bias: Despite efforts to maintain representativeness, the panel's

composition may change over time, leading to potential sample bias.

Data Analysis: Longitudinal data analysis techniques are used to examine

changes and trends over time, such as growth curve modeling, fixed and random
effects models, and event history analysis.

Panel studies are very useful to provide valuable insights into the dynamics of
social, economic, health-related, and behavioral changes over a period of time,
making them a powerful tool for longitudinal research. Their ability to track
individual trajectories and address questions of causality makes them
particularly useful in understanding complex processes and policy effects.

25