Step 10: Analysing Qualitative Data
- Coding means labelling segments of text with short descriptive tags; themes are broader patterns built from related codes.
- Thematic analysis runs in six phases: familiarise, code, generate themes, review, define and name, write up.
- Codes can be deductive (from theory, decided in advance) or inductive (emerging from the data); most projects use both.
- A codebook lists each code with a definition and an example, so coding is consistent.
- Quality in qualitative work is judged by trustworthiness: credibility, transferability, dependability, confirmability. A second coder and Cohen’s kappa can check consistency.
From transcripts to findings
Ten 40-minute interviews can produce over 60,000 words of transcript. Qualitative analysis is the systematic process of turning that mass of text into a small number of well-evidenced findings that answer your question. It is not picking a few quotes you like.
Thematic analysis in six phases
| Phase | What you do |
|---|---|
| 1. Familiarise | Read and re-read the transcripts; jot initial ideas |
| 2. Generate codes | Label every relevant segment systematically, across the whole dataset |
| 3. Construct themes | Group related codes into candidate themes |
| 4. Review themes | Check themes against the coded extracts and the whole dataset; merge, split or drop |
| 5. Define and name | Write a clear definition of each theme and give it an informative name |
| 6. Write up | Tell the story of each theme with analysis and well-chosen quotations |
This six-phase approach is widely associated with Braun and Clarke’s work on thematic analysis, which you should cite if you follow it.
Chapter 10 of Research Methods Made Simple covers this in full, coding ten real transcripts from first codes to a finished thematic analysis.Get the textbook · $15
Coding in practice
Code this extract from an interview with “Aisha” (pseudonym), a second-year student who works 18 hours a week in a supermarket.
“Honestly, the shifts are what keep me on track. If I know I’m working Thursday night, I do my reading Wednesday. But in exam season it’s chaos, because the manager won’t give me time off and I end up revising on my break.”
One short extract carries four codes, and two of them point in opposite directions. That tension is often where the most interesting findings are.
A codebook
| Code | Definition | Include when | Example |
|---|---|---|---|
| WORK GIVES STRUCTURE | Work schedule helps the student plan study time | The student links shifts to planning or routine | “shifts keep me on track” |
| EMPLOYER INFLEXIBILITY | Employer refuses or limits changes to hours | Any reported refusal of time off or shift swaps | “won’t give me time off” |
From codes to themes
After coding all ten transcripts, related codes are grouped. For example, EMPLOYER INFLEXIBILITY, SHIFT CLASHES and EXAM-PERIOD PRESSURE might form a theme named “When work and study collide: the exam-season squeeze”, while WORK GIVES STRUCTURE and TIME-MANAGEMENT SKILLS form “Work as scaffolding”. A good theme name says something, rather than just naming a topic (“Employers”).
Checking consistency: a second coder
If a second person codes a sample of segments with your codebook, you can measure agreement. Percentage agreement overstates consistency because some agreement happens by chance; Cohen’s kappa corrects for this.
Two coders judge 50 segments for whether the theme “work as scaffolding” is present. Coder A says present for 20, coder B for 22. Both say present for 17 and both say absent for 25.
Trustworthiness
| Criterion | Parallel in quantitative work | How to show it |
|---|---|---|
| Credibility | Internal validity | Member checking, quotes, deviant cases |
| Transferability | External validity | Thick description of the setting and participants |
| Dependability | Reliability | An audit trail: codebook, memos, decisions |
| Confirmability | Objectivity | Reflexive notes; findings grounded in data |
Practise until it sticks: Research Methods Made Simple Practice Questions has 1,004 questions, each with a worked answer.Get the practice book · $9
Practise: 6 questions
Try each one on paper first, then check the worked answer.
EasyQ1. What is the difference between a code and a theme? Show worked answer
EasyQ2. List the six phases of thematic analysis. Show worked answer
MediumQ3. Code: “My job pays my rent, so I can’t cut hours, even though I know my grades have dropped.” Show worked answer
MediumQ4. Two coders agree on 90% of segments, and chance agreement is 0.50. Calculate kappa. Show worked answer
MediumQ5. Why is “Employers” a weak theme name, and what would be better? Show worked answer
ExamQ6. Explain how a researcher can make a thematic analysis trustworthy. [6] Show worked answer
Check yourself
1. Codes decided in advance from theory are
2. Cohen’s kappa improves on percentage agreement by
3. An audit trail mainly supports
Your score: 0 out of 3
The textbook and the practice book together: the Research Methods Made Simple bundle for $19 (was $24).Both books · $19
The short version
- Code systematically, build themes from codes, then review, define and name them.
- Keep a codebook; check consistency with a second coder and kappa.
- Judge quality by credibility, transferability, dependability and confirmability.
- Analyse every quotation; never let quotes stand alone.
Studying more than one subject? The Complete Book Bundle includes every Economics, Statistics, Mathematics, Econometrics and Research Methods book.
Interview and Thematic Analysis Kit · $5Interview guide, coding frame and a worked thematic analysis. Free sample on the product page. Or get all 7 research kits for $19.