Handout 1

Research Data
I. Overview of Research Data
Data is defined as a purposive collection of facts, observations, or information related to a specific
question or problem. In research, data acts as the "raw information" from which statistics are
created to provide interpretation and summary. The objective of data collection is to gather reliable
information to develop new knowledge and improve educational practices.

Quantitative data refers to numerical data and information that can be counted and quantified. On
the other hand, qualitative data refers to descriptive words and images (and sometimes sounds), it
collects feedback, opinions, attitudes, and perceptions that provide contextual information and
cannot be counted.
Quantitative data answers the ‘what’, ‘how often’ and ‘how much/many’ questions,
while qualitative data answers the ‘why’ and ‘how’ questions.
Quantitative data is typically collected through surveys using a representative sample and findings
can be generalised. Qualitative data is usually collected through observations, literature review or
interviews such as key informant interviews or focus-group discussions where there is not any pre-
determined list of answer options. Although findings from qualitative data cannot be generalised,
they offer rich insights and in-depth contextual information that can help explain quantitative
findings. As such, quantitative and qualitative data complement each other (mixed-method)
Quantitative data is quicker and easier to analyse, less open to errors and bias. It identifies trends
and patterns and is usually reported as percentages, averages, or figures. While the analysis
of qualitative data can be lengthier requiring the identification of codes and themes and tends to be
more interpretation-based and subjective to the researcher’s bias.

1
II. Data Collection Methods and Instruments

1. Observation Method

• Definition: The systematic recording of participant behavior using sensory systems, making
judgments on frequency, duration, or latency.
• Characteristics:
o Structured: Uses preset checklists to observe predetermined evidence to support or
refute a theory.
o Non-structured: Allows issues to emerge naturally, often resulting in narrative
qualitative data.
o Direct vs. Indirect: Direct observation involves watching behaviors as they occur,
while indirect observation involves watching the results or reported accounts of
behaviors.
2
 • Advantages: Measures actual behavior rather than reported behavior; does not rely on a
participant's willingness to provide information.
• Disadvantages: Time-consuming; susceptible to researcher bias; cannot capture the
underlying thinking behind observed actions.
• Steps to Conduct:
1. Determine focus based on research questions.
2. Design a system (checklists, guides, or field notes).
3. Select and train observers to ensure consistent data.
4. Schedule observations to capture necessary activity components.
• Data Analysis Procedure: Analysis typically involves coding specific behaviors to
measure frequency or duration, or synthesizing narrative descriptions to identify patterns
in interactions.

2. Interview Method

• Definition: A verbal survey administered individually or in groups to uncover social

meanings, reasoning, and feelings.
• Characteristics: Can be Structured (standardized questions/order), Semi-structured
(flexible guide with open-ended probes), or Non-standardized (conversational and
informal).
• Advantages: Yields rich, in-depth data; allows the interviewer to clarify questions and
explore complex subjects.
• Disadvantages: Expensive and time-consuming; requires highly trained interviewers;
volume of data can be difficult to transcribe and reduce.
• Steps to Conduct:
1. Identify topic and subjects.
2. Prepare an interview guide with logical sequences and probes.
3. Inform respondents of ethical principles (anonymity/confidentiality).
4. Conduct and record the interview in a distraction-free area.
• Data Analysis Procedure: Involves transcribing recordings, followed by thematic
coding where labels are applied to passages of text to identify recurring themes.

3. Questionnaires
Definition: A standardized instrument where respondents provide answers to a set of questions
regarding themselves or a social unit.
• Characteristics: Utilizes Closed-ended questions (multiple-choice for quantitative data) or
Open-ended questions (narrative for qualitative insights).
• Advantages: Efficient for large samples; inexpensive; protects participant privacy.
• Disadvantages: Rigid structure may lack detail; poor construction can lead to biased or
confused answers.
• Steps to Conduct:
1. Define target population and variables of interest.
2. Draft and refine questions, avoiding leading or double-barrelled phrasing.
3. Pilot test the instrument to ensure clarity.
4. Distribute via mail, e-mail, telephone, or face-to-face.

3
 • Data Analysis Procedure: Closed-ended responses are entered into statistical software for
numerical analysis (frequencies, percentages). Open-ended responses are subjected to
content analysis similar to interview transcripts.
4. Tests

• Definition: Tools designed to measure specific samples of human behavior, such as

abilities, skills, or personality traits.
• Types: Achievement tests (measure current knowledge), Aptitude tests (measure potential
success), and Criterion-referenced tests (measure performance against specific objectives).
• Data Analysis Procedure: Focuses on objectivity in scoring, typically using a
standardized key to ensure that any scorer would arrive at the same result.

III. Data Analysis Procedures

Data analysis is the systematic process of inspecting, cleaning, and modeling data to discover useful
information, suggest conclusions, and support decision-making. The procedures for analysis vary
significantly depending on whether the data is quantitative or qualitative.
Analyzing Quantitative Data
Quantitative analysis focuses on numerical data and objective measurements using statistical and
mathematical models.
The Five Stages of Statistical Inquiry:

1. Collection of Data: Gathering information from primary sources (original research) or

secondary sources (existing records).
2. Organization of Data: This involves editing (correcting omissions or errors), classifying
(grouping related facts into homogeneous classes), and tabulating (arranging data into
columns and rows for clarity).
3. Presentation: Systematically displaying data using tables, graphs, and diagrams to make it
intelligible.
4. Statistical Analysis: Applying mathematical measures to understand patterns.
o Descriptive Statistics: Summarize basic features of the data set, such as measures
of central tendency (mean, median, mode) and measures of dispersion (range,
variance, standard deviation).
o Inferential Statistics: Use sample data to make predictions or generalizations about
a larger population, employing techniques like correlation, regression analysis, t-
tests, and Analysis of Variance (ANOVA).
5. Interpretation: Drawing valid conclusions based on the analyzed results.

Analyzing Qualitative Data

Qualitative analysis is non-numerical and focuses on descriptions, opinions, and narrative text to
understand social meanings and "why" phenomena occur.
Standard Steps for Qualitative Analysis:

1. Transcription and Organization: The first stage often involves transcribing verbal
recordings (from interviews or focus groups) into written text and organizing it into
categories.
4
 2. Coding: Researchers identify passages of text and apply labels or codes to them to represent
specific themes.
3. Thematic Grouping: Similar codes are grouped together to identify recurring patterns or
theories.
4. Pattern Synthesis: The researcher interrogates the data to see relationships between themes
and develops explanations.
5. Narrative Description: The final results are presented as a narrative that tells "the story" of
the data, often supported by direct quotations.

Specific Qualitative Designs:

• Grounded Theory: Uses a three-stage coding process: Open coding (segmenting and
labeling data), Axial coding (relating categories to one another), and Selective coding
(interpreting the interrelationships to produce an explicit theory).
• Ethnography: Analysis occurs simultaneously with data collection; as understanding
occurs, new questions emerge to shape the final narrative description.

Key Differences in Processing

• Classification: In quantitative work, this refers to grouping by numerical properties or

distinct categories (e.g., age ranges); in qualitative work, it involves grouping by attributes
and common characteristics found in narratives.
• Tabulation: Quantitative tabulation arranges numbers for calculation; qualitative
"tabulation" might involve designing a simple table to list how many participants mentioned
specific themes.
• Software Use: Researchers often use software like Excel, SPSS, or SAS for quantitative
statistical manipulation, while qualitative researchers may use tools like NVivo to help
classify data using codes

(source: Research Methods in Education, book)