1.

Introduction to Data Collection

What is Data Collection?

Data collection is the systematic process of gathering, measuring, and analysing

information from a variety of sources to answer research questions, test
hypotheses, and evaluate outcomes. It serves as the foundation of all scientific
inquiry, allowing researchers to generate valid and reliable evidence.

Data = Information or facts that you gather.

Collection = the act of gathering or obtaining this information.
Purpose: You collect data to answer questions or solve problems.
Effective data collection ensures that the information gathered is:
Accurate (free from error),
Consistent (repeatable and reliable), and
Relevant (addresses the research problem).

There are two primary categories of data:

Primary Data: Collected directly from the source by the researcher for the
specific study.
Secondary Data: Already collected by others for different purposes but utilized
by the researcher.

2. Techniques and Methods of Data Collection

The methods of data collection vary based on the research design (qualitative,
quantitative, or mixed-methods), the nature of the study, and the research
population. Below is a detailed exploration of each method:
A. Observation Method

Definition: A qualitative method where the researcher observes subjects in their

natural setting without interference. Involves watching and recording the
behaviour or activities of participants in their natural setting without intervening
When to Use: When the researcher wants to observe actual behaviour, rather
than rely on self-reporting.
Example: Observing children’s interaction during recess to study social
behaviour.
Types of Observation:
Participant Observation: Researcher becomes part of the group.
Non-Participant Observation: Researcher observes without interaction.
Controlled Observation: Conducted in a structured, controlled environment.
Uncontrolled (Naturalistic) Observation: Takes place in real-world settings.
Advantages:
Provides real-time, direct data.
Useful in behavioural research.
Limitations:
Subject to observer bias.
Difficult to replicate and analyse statistically.

B. Interview Method
Definition: A verbal technique of data collection where a researcher asks
questions directly to participants.
Description: A method where a researcher interacts directly with the participant
to gather information.
When to Use: When you need in-depth, qualitative data that cannot be captured
by closed-ended questions
Example: An interview with a teacher about challenges in the education
system.
Types of Interviews:
Structured: Pre-planned questions in fixed order.
Unstructured: Open-ended, conversational style.
Semi-Structured: Mix of pre-determined and open-ended questions.
Advantages:
Enables deep insights.
Allows clarification and probing.
Limitations:
Time-consuming.
Prone to interviewer bias.

C. Questionnaire and Schedule Method

i. Questionnaire
Definition: A written instrument containing a series of questions aimed at
gathering information from respondents.
Types:
Closed-ended: Respondents select from given options.
Open-ended: Respondents answer in their own words.
Advantages:
Cost-effective for large samples.
Standardized data.
Limitations:
Limited response depth.
Requires respondent literacy.
ii. Schedule
Definition: Similar to a questionnaire but filled out by a trained enumerator
during a face-to-face interaction.
Advantages:
Suitable for illiterate populations.
Higher response rate due to direct interaction.
Limitations:
More expensive and time-intensive.

D. Case Study Method

Definition: An in-depth, detailed examination of a single subject, group,
institution, or event over time.
Description: A detailed, in-depth analysis of a single case or a small number of
cases.
When to Use: When you need a comprehensive understanding of a particular
subject, group, or event.
Example: A case study of a company’s turnaround strategy.
Types:
Descriptive, Exploratory, Explanatory, and Intrinsic case studies.
Advantages:
Rich, holistic insights.
Useful for exploring complex issues in real-life contexts.
Limitations:
Low generalizability.
Subjective interpretation.
E. Historical Method
Definition: Involves analyzing past records, documents, and other archival
materials to investigate a phenomenon or test hypotheses.
Data Sources:
Official records, diaries, newspapers, letters, legal documents, etc.
Advantages:
Useful for understanding trends over time.
Helps contextualize current issues.
Limitations:
Data authenticity and accuracy may be questionable.
Incomplete records.

F. Experimental Method
Definition: A quantitative method where variables are manipulated to study
cause-effect relationships under controlled conditions.
Description: A controlled method where researchers manipulate one or more
variables to observe their effect on other variables.
When to Use: When the researcher seeks to understand cause-and-effect
relationships.
Example: A laboratory experiment testing the effects of different teaching
methods on student performance.
Key Elements:
Independent Variable: Manipulated by the researcher.
Dependent Variable: Outcome measured.
Control Group: Not exposed to the treatment.
Experimental Group: Receives the treatment/intervention.
Advantages:
High internal validity.
Enables causal inference.
Limitations:
Artificial settings may reduce external validity.
Ethical issues may arise.

G. Survey Method
Definition: A widely used quantitative technique that involves collecting data
from a sample using structured tools like questionnaires or interviews.
Types:
Cross-sectional: Data collected at one point in time.
Longitudinal: Data collected over a period.
Advantages:
Generalizable results (if sample is representative).
Versatile and cost-efficient.
Limitations:
Response biases.
Non-response and sampling errors.

3. Measurement Scales in Data Collection

In social science research, measuring attitudes, opinions, and perceptions is
crucial. This is done using scaling techniques like the Likert Scale and the
Semantic Differential Scale.

A. Likert Scale
Definition: A psychometric scale used to measure respondents’ level of
agreement or disagreement with a statement.
Example:
“The government should increase spending on public health.”
1 – Strongly Disagree, 2 – Disagree, 3 – Neutral, 4 – Agree, 5 – Strongly Agree
Characteristics:
Usually 5-point or 7-point scale.
Used in attitude and opinion measurement.
Advantages:
Easy to construct and administer.
Quantifiable and suitable for statistical analysis.
Limitations:
Tendency to choose neutral/middle options.
Does not measure intensity of feeling precisely.

B. Semantic Differential Scale

Definition: A scale that measures people's reactions to stimulus words and
concepts using a set of bipolar adjectives.
Example: Rate “Urban Policy” on the following:
Efficient _ Inefficient
Fair _ Unfair

Advantages:
Reveals subtle attitudes and perceptions.
Flexible and adaptable to various research contexts.
Limitations:
Difficult for some participants to interpret.
Requires careful construction of adjective pairs.

4. Conclusion
Data collection is the cornerstone of empirical research. The choice of data
collection technique must align with the research objectives, research design,
target population, and available resources. Each method—whether qualitative
(like observation or interviews) or quantitative (like surveys or experiments)—
has distinct strengths and limitations. Additionally, the appropriate use of
scaling techniques like the Likert or Semantic Differential Scale allows
researchers to quantify subjective phenomena such as attitudes, beliefs, and
preferences.

A combination of methods, often termed triangulation, can be employed to

enhance the validity and reliability of the research findings.

I. Meaning of Research Design

A research design is the blueprint or plan that outlines how a research study is
to be conducted. It provides a structured framework for collecting, measuring,
and analyzing data in a way that answers the research questions or tests
hypotheses effectively and efficiently.
According to Kerlinger (1986), "Research design is the plan, structure, and
strategy of investigation conceived so as to obtain answers to research questions
and control variance."
Thus, a good research design ensures:
Validity (accuracy of findings),
Reliability (consistency of results),
Objectivity, and Generalizability of the research.

II. Types of Research Design

Research designs are broadly classified into the following categories:
A. Exploratory Research Design
Purpose: To explore a problem or situation when there is limited prior
knowledge.
Characteristics:
Flexible and unstructured.
Used to generate hypotheses.
Methods: Literature review, expert interviews, focus groups, case studies.

B. Descriptive Research Design

Purpose: To describe characteristics of a population or phenomenon.
Characteristics:
Structured and preplanned.
Does not investigate cause-effect relationships.
Methods: Surveys, observational methods, content analysis.

C. Diagnostic Research Design

Purpose: To determine the causes of a phenomenon.
Characteristics:
Focuses on understanding underlying reasons.
Often used in applied fields (e.g., management, psychology).
Methods: Root cause analysis, in-depth interviews, historical review.

D. Experimental (Causal) Research Design

Purpose: To test hypotheses about cause-effect relationships.
Characteristics:
Involves manipulation of variables and control over extraneous factors.
Employs random assignment.
Types:
Pre-experimental design (e.g., one-group pretest-posttest).
True experimental design (e.g., randomized control trials).
Quasi-experimental design (e.g., non-randomized groups).

E. Analytical Research Design

Purpose: To test hypotheses using quantitative data and statistical analysis.
Characteristics:
Focus on relationships and patterns in data.
Methods: Regression analysis, correlation, structural equation modeling.

III. Steps Involved in the Formulation of a Research Design

Formulating a research design is a systematic process involving several steps:

1. Identification and Clarification of the Research Problem

 Clearly define what needs to be studied.
 Ensure the problem is researchable and significant.
2. Review of Literature
 Conduct a thorough review of existing studies.
 Identify gaps in knowledge and relevant theories.
3. Formulation of Objectives and Hypotheses
 State the research objectives (what the study aims to achieve).
 Develop research questions or hypotheses to be tested.
4. Selection of the Research Design Type
 Choose between exploratory, descriptive, diagnostic, or experimental
based on objectives.
5. Identification of Variables
 Independent variables (causes).
 Dependent variables (effects).
 Control variables (factors kept constant).
6. Selection of the Population and Sampling Design
 Define the target population.
 Choose a suitable sampling method (probability or non-probability).
7. Choice of Data Collection Methods
 Select appropriate methods: surveys, interviews, observations,
experiments, etc.
 Decide on tools/instruments (questionnaires, schedules, tests).
8. Operationalization of Concepts
 Define how theoretical concepts will be measured empirically.
9. Data Analysis Plan
 Decide on techniques for analyzing data: descriptive stats, inferential
stats, qualitative analysis.
10. Time Schedule and Budgeting
 Prepare a timeline and estimate financial resources needed.
11. Ethical Considerations
 Ensure ethical standards: informed consent, confidentiality, and
protection from harm.
12. Pre-testing and Pilot Study
 Conduct a small-scale pilot to test feasibility and refine instruments and
procedures.

Conclusion
Research design is a fundamental element of the research process that
determines the overall strategy for data collection and analysis. The choice of
design depends on the research objectives, nature of the problem, and the kind
of data needed. A carefully formulated research design enhances the scientific
rigor and credibility of research findings.