When to use NVivo
NVivo is software for organising and analysing qualitative data such as interviews, focus groups, open-ended survey answers, documents and social media text. It helps you store the material in one place, code it, retrieve coded segments and see patterns.
NVivo does not analyse data for you. You still decide what the codes and themes are. Use it when you have enough material that manual coding would be slow or hard to track, and when your university gives you access.
Preparing and importing data
Transcribe interviews accurately and anonymise participant names. Give files clear names, such as "P01_Interview". Import documents, then create cases or classifications for participant details such as age or job role if you plan to compare groups.
Check that the transcripts are complete and that you have participants' consent. Keep an original backup of the data and save your NVivo project regularly in more than one location.
Coding and nodes
Read through the data first. Then create codes (called nodes in NVivo) and attach them to relevant passages. You can start with a few codes based on your research questions (deductive) or create them from what you see in the data (inductive).
Give each code a clear name and a short description so your coding stays consistent. Merge duplicates and refine as you go. Keep a memo or journal to record your decisions and reflections.
Building themes
A theme is a pattern of meaning across the data, not just a topic. After initial coding, group related codes into broader categories, review them against the coded extracts and then check whether they tell a coherent story about your research question.
Name each theme clearly, define it and select examples. Following a recognised approach, such as Braun and Clarke's six phases, strengthens your method section. Use NVivo's coding queries and visualisations to explore relationships, but base conclusions on the data itself.
Reporting qualitative findings
Report each theme with a short definition, a summary of what participants said and a few quotations that illustrate it. Label quotations with a participant code, such as (P03), to protect identities.
Explain what the theme means in relation to your research questions and the literature. A table of themes, sub-themes and counts of participants can help, but avoid treating frequency as the only sign of importance.
Common mistakes
Common mistakes include:
- Creating too many codes without organising them.
- Treating codes as themes.
- Using quotations without interpreting them.
- Coding only the data that fits your expectations.
- Not recording your coding decisions.
- Losing files because of poor backup.
- Claiming that results are statistically representative from a small qualitative sample.
Check your approach with your supervisor, and consider a second coder for part of the data to improve credibility.