Systematic Qualitative Analysis and Thematic Frameworks
Course Logistics and Student Context
Current Timeline: The academic term is currently in Week . A term break occurs between Week and Week . Assessment is due in Week .
Student Workload and Wellbeing: Students report high levels of stress and a dense volume of competing deadlines. By Friday of Week , some students are submitting up to assessments. In Week , students like Tanisha have assessments due. Sheree is managing Psych Research Methods and Sociology units simultaneously, with approximately assessments due in Week .
Campus Activities: Classes are held at the Burwood campus. Students are encouraged to use the mid-term break for rest and recovery, particularly those recovering from illnesses like losing their voice.
Qualitative vs. Quantitative Research Fundamentals
Core Distinctions: Qualitative research differs significantly from quantitative research in objectives and data types. Quantitative research focuses on numerical data and statistical analysis, while qualitative research explores depth, meaning, and perception.
Qualitative Data Sources:
- Interviews: Involving structured or semi-structured dialogue. Assessment specifically utilizes interviews with AI bots acting as interviewees.
- Focus Groups: These involve group interactions and offer different dynamics compared to one-on-one interviews.
- Social Media: Qualitative data can be gathered from social media posts, a method utilized in the third-year unit "Conspiracies, Misinformation, and Hate Crimes in the Age of AI."
Principles of Qualitative Data Interpretation
Variability in Analysis: If two researchers analyze the same interview, they may not reach identical findings. Several factors influence this:
- Research Questions: Different initial research questions guide the researcher toward different sections of the data.
- Follow-up Questions: Individualized prompts during an interview change the resulting data.
- Analytical Angle: Researchers may approach the same AI response from different perspectives, such as "freedom of speech."
- Personal Biases: Every researcher has unique biases that influence their interpretation.
AI Responses: AI bots exhibit variability in their perspectives. While they may remain firm in certain views, they can make concessions depending on the perspective of the prompt (e.g., a freedom of speech angle). Real human participants typically show significantly more variability than AI bots.
Accuracy and Quality Mitigation: To ensure findings reflect actual data rather than researcher expectations:
- Rigorous Frameworks: Using systematic models like thematic analysis.
- Substantiation: Using verbatim quotes as evidence for every finding.
- Contextualization: Providing the specific context for how interpretations were formed.
- Peer Review: Having colleagues review and analyze the data to see if themes align.
- Participant Consultation: In real-world research, researchers may go back to participants to clarify meanings and verify if the analyzed findings accurately reflect their perspectives.
The Braun and Clark Thematic Analysis Framework
Overview: Developed by Brown and Clark, this framework originates in psychology but is applied across various disciplines. It is widely used for its systematic and rigorous approach to qualitative data analysis. While often introduced in fourth-year studies, it is accessible for second-year students to move beyond surface-level interpretations.
Step One: Familiarization: This involves reading the data multiple times to understand the content thoroughly. For Assessment , this means reading all four AI interview transcripts and highlighting recurring ideas regarding the legislation.
Step Two: Generating Initial Codes: Researchers attach labels or "codes" to meaningful pieces of text. This is a foundational step that captures specific observations.
Step Three: Generating Initial Themes: Related codes are grouped to identify patterns of shared meaning across the entire dataset. Themes are broader than codes and must answer the research question.
Step Four: Reviewing Themes: Themes must be checked against the original data to ensure they are accurate representations.
Step Five: Defining and Naming Themes: Researchers must clearly define what each theme encompasses and give it a descriptive name.
Step Six: The Write-Up: This involves synthesizing the themes and explaining them using quotes as evidence.
Distinguishing Codes and Themes
Codes:
- Codes are the "building blocks" of analysis and serve as a research tool.
- They capture a single idea or observation and stay very close to the participant's original words.
- Examples include "Freedom of speech," "Community safety," "Religious freedom," "Unjust," or "Inclusivity."
- Codes help navigate large datasets, such as to pages of transcript, by allowing the researcher to scan notes rather than reread full transcripts repeatedly.
Themes:
- Themes are the analytical outcomes or results of the research.
- They are groups of related codes that form a broader story or pattern.
- Every theme must address the research question directly.
- An example theme name might be "Balancing individual freedoms with protection from harm."
Assessment Guidelines and Best Practices
Analytical Objective: The goal is not to determine who is "right" or "wrong" among the interviewees, but to identify patterns across the four interviews to explain how different communities perceive the legislation.
Evidence and Quotes: Findings must be substantiated with quotes. In the write-up, the researcher should introduce the themes (e.g., "This research yielded three prominent themes") and then describe them while integrating participant quotes as evidence.
Appendix Usage: It is mandatory to include an appendix. Researchers are encouraged to include transcripts with their generated codes in the appendix to document their systematic process.
Referencing Requirements:
- The recommended referencing style is APA .
- While first-year assessments require a minimum of references and second-year assessments often require , students should aim for a range between and references.
- Quantity vs. Quality: Researchers should avoid excessive referencing (e.g., references) to ensure the focus remains on substantial engagement with the primary data.
Questions and Discussion
Dialogue on Coding Strategy: Jarman hỏi whether interesting points that are not directly related to the research question should still be coded. The response indicates that during the initial coding phase, researchers should not limit themselves. Any meaningful or insightful data—even if it providing context rather than a direct answer—should be coded. These points can be discussed in the final write-up if they are insightful.
Dialogue on Synthesis and Philosophy: Jakob observes that the process of thematic analysis builds on the principle of synthesis. By taking a vast amount of information and binding it together, the researcher builds toward a representation of truth. This is noted as having philosophical roots, specifically in the work of Hegel regarding synthesis, though existentialist philosophers sometimes opposed such synthesis. This framework is described as a way to find the "core" of research and avoid being "eternally doomed" by unorganized data.
Dialogue on Transcript Length: Grace expresses concern over managing pages of transcript. The systematic six-step framework is offered as the solution to make this volume of text manageable and logically structured.