QM UT2 Reviewer

Research Question

• A research question is a clear, focused, and specific inquiry that a researcher aims to answer, guiding the study and determining the methods used, while defining the problem or issue being addressed.
• It is used early in the research process to guide the study, determine methods (qualitative, quantitative, or mixed), and analyze data.
• A good research question is clear, focused, researchable, and feasible within the study's scope.
• Research question examples:
  • How does social media use affect the self-esteem of teenagers?
  • Does exercise reduce the symptoms of anxiety in individuals diagnosed with depression?
  • How does student engagement impact academic performance in online learning?

Hypothesis

• A hypothesis is a testable statement or prediction about the relationship between two or more variables, based on prior knowledge or theory.
• It is used before data collection to provide direction, test relationships, and validate or refute theories.
• Null Hypothesis (H₀): States there is no effect or relationship between variables.
  • Example: "The new drug has no effect on blood pressure."
• Alternative Hypothesis (Ha): States an effect or relationship exists.
  • Example: "The new drug decreases blood pressure."
  • Two-tailed: Predicts a difference but doesn’t specify direction.
  • One-tailed: Predicts the direction of the effect.
• Key difference between a research question and a hypothesis:
  • A research question asks what you want to investigate (open-ended).
  • A hypothesis predicts the outcome of the investigation (specific and testable).
• The question leads to the hypothesis.
• The hypothesis helps test the research question by guiding data collection and analysis.

Qualitative Research Questions

• These questions explore or interpret experiences, phenomena, or meanings in depth, focusing on the "how" or "why" rather than numerical data.
• Qualitative research questions are open-ended, exploratory, and context-specific, focusing on experience.
• Qualitative research typically does not use hypotheses, instead focusing on exploration without preconceived assumptions.
• Types of qualitative research questions:
  • Central Question: Broad, exploratory inquiry.
  • Subquestions: Narrow the focus of the study.
• Examples of Qualitative Research Questions:
  • Phenomenological: "What is the lived experience of individuals recovering from a heart attack?"
  • Ethnographic: "How do traditional healing practices influence health-seeking behavior in rural African communities?"
  • Grounded Theory: "What are the factors that influence women to leave abusive relationships?"
  • Narrative: "What stories do refugee children tell about their journey to safety?"
• Qualitative research provides insights into human behavior, experiences, and perceptions that quantitative methods can’t capture.

Quantitative Research Questions

• Quantitative research collects numerical data and measures variables using statistical methods.
• It aims to establish relationships, test hypotheses, or quantify differences between groups.
• These questions are close-ended, measurement-focused, test relationships/differences, and are objective and specific.
• Hypothesis-driven, formulated to test a specific hypothesis.
• Types of quantitative research questions:
  • Descriptive: "What is the average age of university students?"
  • Comparative: "Do male and female employees differ in job satisfaction levels?"
  • Relational: "Is there a correlation between hours of sleep and academic performance?"
• Quantitative research provides objective, measurable data for decision-making, theory testing, and trend analysis.

Descriptive and Inferential Research Questions

• Descriptive: Aims to describe characteristics, behaviors, or situations without exploring relationships, e.g., "What are the students' achievement levels in science classes?"
• Inferential: Investigates relationships between variables or explores their impact, e.g., "How does critical thinking ability relate to student achievement?"
• Descriptive questions focus on describing individual variables.
• Inferential questions explore relationships between variables and test hypotheses.

Mixed Methods Research Questions

• Combines both quantitative and qualitative approaches for a comprehensive understanding.
• Examples:
  • "How does technology use affect students’ grades, and what are their views on its effectiveness?"
  • "What is the link between teachers' experience and classroom management, and how do they describe their challenges?"
• Writing mixed methods questions:
  • Use separate qualitative and quantitative questions to narrow and focus the purpose statement.
  • Can emphasize both approaches or combine them into a single integrated question.
• Provides a fuller picture of complex research problems by combining statistical trends with personal experiences.
• Quantitative data shows what happens, while qualitative insights explain why it happens.

Surveys and Experiments

• Surveys: Collect data using structured questions to understand opinions or behaviors, such as customer feedback on a product.
• Experiments: Test cause-and-effect relationships by manipulating variables in controlled conditions, such as testing a new teaching method's impact on student performance.
• Surveys aim to collect data about opinions, behaviors, or demographics using structured questions (multiple-choice or scales) via online, phone, or in-person methods.
• Experiments aim to test cause-and-effect relationships, involving control and experimental groups and manipulating variables to observe outcomes in laboratory or field settings.
• Surveys describe characteristics or opinions of a population.
• Experiments determine causal relationships between variables.
• Surveys involve self-reported data via questionnaires/interviews.
• Experiments observe outcomes under manipulated conditions.
• Surveys: Used in market research, social sciences, healthcare.
• Experiments: Applied in psychology, education, and agriculture.

Components of a Survey Study

• A method plan outlines approach to collecting data using surveys to ensure a well-structured and valid process.
• Research Problem: Identify the key issue or question to explore.
• Objectives: Define what the survey aims to achieve (e.g., understanding customer satisfaction).
• Designing the Survey Instrument involves using open-ended or closed-ended questions like Likert scale, multiple choice, or yes/no and pre-testing on a small sample to refine.
• Selecting the Sample:
  • Target Population: Define the group from which data will be collected.
  • Sampling Method: Random, stratified, or convenience sampling.
  • Sample Size: Ensure reliability of results.

The Survey Design

• Clearly state the purpose of the survey.
• Identify respondents (age, location, occupation, etc.).
• Ensure relevance to research goals.
• Crafting effective survey questions:
  • Open-Ended: Allow detailed responses.
  • Closed-Ended: Provide specific options (yes/no, multiple choice).
  • Likert Scale: Measure agreement levels.
  • Ranking: Prioritize choices.
• Survey Format & Distribution:
  • Informed Consent: Participants must voluntarily agree with a full understanding.
  • Anonymity & Confidentiality: Protect respondent identity and responses.
  • Survey Validity: Ensure accurate and reliable measurements.
• Data Analysis and Reporting:
  • Data Cleaning: Remove errors, incomplete, or duplicate responses.
  • Statistical Analysis: Descriptive statistics, correlations, regressions.
  • Result Interpretation: Link findings to research objectives.
• Ensuring Survey Validity:
  • Content Validity: Questions cover all aspects.
  • Criterion Validity: Compare to existing measures.
  • Construct Validity: Measure intended concept.

Variables in a Study

• Variables are characteristics that change, such as age, income, or education.
• Dependent Variable: Measured outcome (e.g., exam scores).
• Independent Variable: Causes change (e.g., study time).
• Control Variable: Kept constant to avoid interference.
• Categorical (Qualitative) Variable: Groups or categories (e.g., gender, race).
• Continuous (Quantitative) Variable: Numeric values (e.g., height, weight).
• Levels of Measurement:
  • Ratio: Numeric with absolute zero (e.g., income).
  • Interval: Numeric but no true zero (e.g., temperature).
  • Ordinal: Ordered categories (e.g., satisfaction ratings).
  • Nominal: Categories without order (e.g., colors, names).

The Population and Sample

• Data Collection: Statistical studies need data.
• Understanding the distinction between population and sample is crucial for accurate analysis.
• Population: The entire group of individuals or items being studied, which can be finite or infinite.
  • Example: Students in a school or stars in the universe.
• Sample: A subset of the population, chosen to be practical and cost-effective.
  • Example: 500 students from different schools in a country.
• Sampling saves time, money, and effort.
• Probability Sampling: Equal chance for each member.
• Non-Probability Sampling: Not every member has an equal chance.
• Types of Probability Sampling:
  • Random: Equal chance for all.
  • Systematic: Every nth person.
  • Stratified: Divided by characteristics.
  • Cluster: Random groups selected.
• Types of Non-Probability Sampling:
  • Convenience: Based on availability.
  • Judgmental: Researcher's judgment.
  • Quota: Specific proportions.
  • Snowball: Uses referrals.
• Choosing Sample Size:
  • Larger size improves accuracy.
  • Too small leads to biased results.
  • Use formulas to determine the correct number.
• Avoiding Sampling Bias:
  • Random Methods: Avoid selection bias.
  • Diversity: Ensure representation.
  • Proper Selection: Follow statistical guidelines.

Quantitative Methods

• Focuses on numerical data and statistical analysis, using structured techniques (surveys, experiments, secondary data analysis).
• Aims to quantify variables and establish relationships, using descriptive and inferential statistics.
• Provides objective, replicable, and generalizable results.
• Methods of Data Collection:
  • Surveys: Structured questionnaires or interviews.
  • Experiments: Controlled studies manipulating variables.
  • Secondary Data: Pre-existing sources (government reports, databases).

Data Sources and Reliability

• Primary Sources: Directly collected data (e.g., surveys, experiments).
  • More reliable but time-consuming and expensive.
• Secondary Sources: Pre-collected data (e.g., census, journal articles).
  • Cost-effective but may be outdated or biased.
• Ensuring Data Reliability:
  • Use standardized collection methods.
  • Verify sources for credibility.
  • Reduce bias (random sampling, careful questionnaire design).
  • Cross-check data with multiple sources.

Data Preparation

• Data Cleaning: Remove duplicates, correct errors.
• Data Transformation: Convert data for analysis.

Handling Missing Data & Outliers

• Identifying Missing Data:
  • Check blank fields, placeholders (e.g., "N/A").
  • Use summary statistics to detect missing patterns.
• Handling Missing Data:
  • Deletion: Remove incomplete records (if minimal).
  • Imputation: Replace with mean, median, or predicted values.
  • Interpolation: Estimate based on trends.
• Detecting & Handling Outliers:
  • Detection: Box plots, histograms, Z-scores, IQR method.
  • Handling:
    • Correction: Fix entry errors.
    • Transformation: Normalize data.
    • Exclusion: Remove extreme values.

Descriptive Statistics

• Measures of Central Tendency:
  • Mean: Sum of values ÷ number of observations (sensitive to outliers).
  • Median: Middle value (not affected by outliers).
  • Mode: Most frequent value (useful for categorical data).
• Measures of Dispersion:
  • Range: Difference between max and min values.
  • Variance: Average squared deviation from mean.
  • Standard Deviation: Square root of variance (measures spread).

Data Visualization

• Bar Charts: Compare categorical data.
• Line Graphs: Show trends over time.
• Histograms: Display frequency distribution.
• Box Plots: Visualize distribution, quartiles, and outliers.

Inferential Statistics

• Draws conclusions about a population from a sample.
• Key Concepts:
  • Population vs. Sample: Sample represents the population.
  • Sampling Distribution: Used for estimating parameters.
• Common Methods:
  • Hypothesis Testing: Tests assumptions (Null H₀ vs. Alternative H₁).
  • Confidence Intervals: Range where true parameter likely falls.
  • T-tests: Compare means of two groups.
  • ANOVA: Compare means of three or more groups.
  • Chi-Square Test: Examines relationships in categorical variables.
  • Correlation & Regression: Measures relationships and predicts outcomes.

Interpreting Results & Discussion Section

• Interpreting Results:
  • Summarize Key Findings: Highlight significant results.
  • Compare with Hypotheses: Confirm/reject assumptions.
  • Relate to Research: Compare with past studies.
• Writing a Discussion Section:
  • Explain Implications: Real-world relevance of findings.
  • Acknowledge Limitations: Mention biases, sample size constraints.
  • Propose Future Research: Recommend further studies.
  • Provide a Clear Conclusion: Summarize key takeaways.

Methodology Overview

• A systematic research approach ensuring validity/reliability.
• Outlines research design, data collection, and analysis.
• Ensures study replication and verification.
• Purpose of Methodology:
  • Conduct research systematically and scientifically.
  • Minimize biases/errors.
  • Ensure credible, generalizable results.

Types of Research Methods

• Qualitative: Understanding human behavior (Interviews, Case Studies).
• Quantitative: Numerical data, measurements (Surveys, Experiments).
• Mixed-Methods: Combines qualitative and quantitative for deeper insights.
• Data Collection Techniques:
  • Surveys: Questionnaires for large populations.
  • Experiments: Tests cause-and-effect.
  • Observations: Records behaviors in natural settings.
  • Interviews: In-depth information gathering.

Experimental Studies Overview

• Investigate cause-and-effect relationships.
• Key Components:
  • Hypothesis
  • Variables
  • Experimental/Control Groups
  • Randomization
• Hypothesis & Variables:
  • Null Hypothesis (H₀): No effect/relationship.
  • Alternative Hypothesis (H₁): Assumes an effect.
  • Independent Variable (IV): Manipulated variable.
  • Dependent Variable (DV): Outcome measured.
• Experimental vs. Control Groups:
  • Experimental: Receives treatment.
  • Control: No treatment, for comparison.
  • Purpose: Isolate IV's effect.
• Randomization & Replication:
  • Randomization: Eliminates bias in group assignment.
  • Replication: Ensures consistent, reliable results.

Method Plan

• Method Plan: Steps for data collection, analysis, and interpretation.
• Study Design:
  • Experimental: Manipulates variables.
  • Non-experimental: Observes variables.
  • Longitudinal: Tracks participants over time.
  • Cross-sectional: Collects data at one point.
  • Quasi-experimental: Manipulates variables but no randomization.
• Participants & Sampling:
  • Population: Defined study group.
  • Sampling Techniques:
    • Random: Equal chance selection.
    • Stratified: Divides population into subgroups.
    • Convenience: Uses accessible participants.
  • Sample Size: Justified by power analysis.
• Data Collection & Timeline:
  • Defines methods (surveys, experiments).
  • Specifies schedule for data collection.
• Ethical Considerations:
  • Informed Consent: Participants aware and volunteer.
  • Confidentiality: Protects participant data.
  • Debriefing: Explains study after participation.
  • Ethical Approval: Must be approved before research.
• Statistical Tools/Software:
  • Examples: SPSS, R, Python.
  • Methods: Regression, ANOVA, hypothesis testing.

Interpreting Results & Discussion

• Key Findings: Summarize main results.
• Comparison with Research: Align with previous studies.
• Implications: Relevance to theory/practice.
• Limitations: Identify biases, constraints.
• Future Research: Suggest new study directions.
• Conclusion: Reinforce significance of findings.

Characteristics of Qualitative Research

• Seeks to understand human experiences, behaviors, and interactions in their natural settings, focusing on meaning rather than numerical data.
• Subjectivity: Researcher’s interpretation plays a key role in data analysis.
• Contextual Understanding: Focuses on participants' experiences within their environment.
• Inductive Approach: Observations lead to patterns, themes, and theories.
• Descriptive & Narrative Data: Emphasis on words, emotions, and interpretations rather than statistics.
• Holistic Perspective: Captures the complexity of social interactions and human experiences.

Qualitative Research Designs

• Qualitative research designs help explore various phenomena based on context and research goals.
• Case Study: In-depth examination of a single case (person, organization, event).
• Phenomenology: Focuses on understanding individuals' lived experiences.
• Ethnography: Studies cultural or social groups through immersion.
• Grounded Theory: Develops theories based on observed data.
• Narrative Research: Uses storytelling to explore individuals’ experiences.

The Researcher’s Role and Reflexivity

• Researcher is an active participant in qualitative studies, making reflexivity crucial.
• Researcher’s Influence: Acknowledging biases and perspectives that may affect data collection and analysis.
• Self-Reflection: Continuous evaluation of one’s role in the study.
• Transparency: Documenting how personal biases may shape interpretations.

Data Collection Procedures

• Qualitative research involves various flexible and adaptive methods.
• Interviews: Structured, semi-structured, or unstructured discussions with participants.
• Focus Groups: Group discussions to gather diverse perspectives.
• Observations: Watching behaviors in real-world settings.
• Document Analysis: Reviewing existing records, letters, or media.

Data Recording Procedures

• Accurate data recording ensures credibility and completeness of information.
• Audio/Video Recordings: Captures verbal and non-verbal communication.
• Field Notes: Observations and reflections recorded by researchers.
• Transcriptions: Converting audio into written text.

Data Analysis Procedures

• Analyzing qualitative data involves iterative processes to identify themes and insights.
• Thematic Analysis: Identifies recurring themes.
• Grounded Theory: Develops theories based on emerging patterns.
• Content Analysis: Examines meanings in text or media.

Interpretation

• Interpretation involves making sense of qualitative data by linking findings to research questions and theories.

Validity and Reliability in Qualitative Research

• Instead of traditional validity and reliability, qualitative research emphasizes trustworthiness.
• Credibility: Confidence in findings (e.g., participant verification).
• Transferability: How findings apply to other contexts.
• Dependability: Consistency in the research process.
• Confirmability: Findings should reflect participants’ experiences rather than researcher bias.

Writing the Qualitative Report

• A qualitative report presents the research process, findings, and analysis in a structured and narrative-driven format.
• Key Sections:
  • Introduction: Research background and purpose.
  • Methodology: Data collection and analysis procedures.
  • Findings: Themes, patterns, and participant quotes.
  • Discussion: Interpretation and implications.

The Qualitative Research Paradigm

• Qualitative research follows interpretivist and constructivist paradigms, viewing reality as socially constructed.