The Chromozone / IAL Psychology / Topic A: Social psychology
Directed study: thematic analysis of qualitative data
Teach these first
This page builds on this lesson. Teach it before students start here:
Teacher note: a sensitive context
Context: this self-study uses a current and sensitive world conflict that many students follow on social media. The three responses take no side: they are about the pressure to post, sharing without checking, and stepping back.
Ground rules: students analyse the participants' reasons (normative and informational influence, echo chambers), not the politics; nobody has to share their own views; calm, respectful language; no searching for, sharing or showing posts or images about the conflict.
Opt-out: any student, or the whole class, can use the second data set on the page, three responses about a school phone ban. The tasks, marks and mark points work the same way.
Key words: thematic analysis
- Qualitative data
- Non-numerical information, for example from interviews, open questions in a questionnaire, or posts on social media.
- Thematic analysis
- A way of reducing a large amount of qualitative text to a few general patterns, trends and themes, without losing its meaning.
- Code
- A short label for a feature or idea in the data.
- Researcher bias
- When the researcher's own views or expectations affect how the data are interpreted.
- Independent coder
- A second researcher who analyses the same data separately, so the two sets of themes can be compared (inter-rater reliability).
1. Core concepts
When psychologists study obedience, conformity or public opinion during world events, they often use qualitative data. Because qualitative data sets are large and unstructured, they use thematic analysis to reduce them to manageable themes.
- Inductive approach: the researcher reads and rereads the data openly and lets the themes come from the data, with no expectations fixed in advance.
- Deductive approach: the researcher applies categories or a theory chosen in advance (for example compliance or conformity models) to code the text.
- Limitation: thematic analysis is open to researcher bias, because the themes depend on human interpretation. An independent coder can check inter-rater reliability.
2. Practical scenario and data set
A research team collected open-ended survey responses from university students about their online activity, their feelings and the social pressures they face when posting about the Israel-Palestine conflict. The responses take no side in the conflict. Read the three responses. (The participants are made up for this task.)
Participant A (Tariq)
"Every time I open Instagram, my feed is full of posts about the conflict, and people argue strongly in the comments. At first I felt frozen. I didn't want to say the wrong thing and get criticised or dropped by my friends online. Then I felt that staying quiet was wrong for me. I started sharing links to verified humanitarian aid charities. My family warned me that posting could affect my chances of finding work later, but I felt I had a duty to speak."
Participant B (Sarah)
"My university held a debate, and students quickly split into two groups. My close friends wanted everyone to share protest posters and infographics straight away. I didn't check half of the infographics myself, but everyone else was reposting them. If you didn't post, people said you didn't care. It felt less like real activism and more like going along with the group to fit in."
Participant C (David)
"I chose to mute all political hashtags and stop reading comments about the conflict. Online, people argue so strongly that there is no room for nuance, and everyone feels pushed to pick a side straight away. Stepping back protects my mental health, but I sometimes feel guilty, because some people online see staying out of it as not caring."
Other option: a school phone ban
Use this data set instead if you or your teacher chose the other option. A research team asked school students how they felt about an online argument over a new phone ban. (The participants are made up for this task.)
Participant A (Nadia)
"When the school announced a phone ban, the year group chat filled with angry messages. I actually thought the ban might help me focus, but I didn't want to say so and get laughed at. So I said nothing in the chat. Later I wrote to the student council in private, because I felt someone should give the other view."
Participant B (Ben)
"Everyone in my friendship group signed an online petition against the ban, so I signed it too. I didn't read what it said. If you didn't sign, people said you were on the teachers' side. Looking back, I signed to fit in, not because I had a view."
Participant C (Kofi)
"I left the group chat for a week. The arguing never stopped and nobody listened to anyone. Leaving helped me feel calmer, but I felt a bit guilty, as if it meant I didn't care about my friends."
Task 1: coding and data immersion (3 marks) 8 min
Read the three responses. Identify three quotes, one from each participant, and turn each one into a short descriptive code that captures the psychological reason behind their behaviour online. (3)
Participant A: quote and codeParticipant B: quote and codeParticipant C: quote and code
Reveal the answer key: Task 1
Task 1: coding (3): one mark per quote and code
AO2 (3 marks)
Credit one mark for each accurate quote with a code that fits it, up to three marks.
- Tariq: "I didn't want to say the wrong thing and get criticised" → fear of being rejected online (1)
- Sarah: "If you didn't post, people said you didn't care" → conforming to group norms (peer pressure) (1)
- David: "Stepping back protects my mental health" → stepping back from social media to protect himself (1)
Credit any accurate quote with a code that fits it, for example Tariq: "I felt I had a duty to speak" → a sense of duty to speak.
Phone ban set: Nadia: "I didn't want to say so and get laughed at" → fear of rejection; Ben: "I signed to fit in" → conforming to group norms; Kofi: "Leaving helped me feel calmer" → stepping back to protect himself.
Look for other reasonable marking points.
Task 2: theme generation (4 marks) 8 min
Group your codes from Task 1 into two inductive themes that sum up how social media users deal with a divisive topic online. Give each theme a title and a one-sentence definition. (4)
Theme 1 titleTheme 1 definitionTheme 2 titleTheme 2 definition
Reveal the answer key: Task 2
Task 2: themes (4): two marks per theme
AO2 (4 marks)
Credit one mark for each theme title and one mark for each definition that fits the codes.
- Theme 1 title, for example "Social pressure and fear of online rejection" (1)
- Definition: users change what they post to avoid rejection or criticism (Participants A and B) (1)
- Theme 2 title, for example "Psychological cost and stepping back" (1)
- Definition: users step back to protect their well-being, despite guilt (Participant C) (1)
Credit other sensible themes with a clear title and a definition that fits the codes.
Look for other reasonable marking points.
Task 3: stretch (4 marks) 10 min
A researcher uses a deductive approach based on normative and informational social influence. Explain how they might code Participant B's response differently from a researcher using an inductive approach. (2)
Explain how researcher bias could distort the findings of this study. (2)
Reveal the answer key: Task 3
Task 3 (a) (2)
AO2 (2 marks)
Credit one mark for the deductive coding and one mark for the inductive coding.
- Deductive: looks only for set categories, so codes Participant B's posting as NSI (compliance) (1)
- Inductive: lets new codes emerge, for example sharing (or signing) without checking (1)
Generic answers score 0 marks.
Look for other reasonable marking points.
Task 3 (b) (2)
AO3 (2 marks)
Credit one mark for identifying how researcher bias could distort the findings and one mark for justification.
- A researcher with strong views on the topic may misread a response, for example Participant C's stepping back as not caring (1)
- An independent coder can check this (inter-rater reliability) (1)
Generic answers score 0 marks.
Look for other reasonable marking points.