Thematic analysis of interview data: 6 ways NVivo can help
10 October 2016 - BY Kath McNiff
On any given day, close to 80% of NVivo users are busy analyzing interviews.
Like you, they're looking for the best way to make sense of their data so they can deliver robust results.
If you’re new to NVivo or just getting started with analyzing interviews (thematic analysis) it can be hard knowing where to start. These tips can help guide you on your way.
#1. Transcribe the interview recordings
Let's face it, transcribing can be tricky - particularly if you're working with a lot of interviews or have limited time (and budget).
Luckily, NVivo gives you a number of options. For example, you can import the interview recordings into NVivo and:
- Upload them to TranscribeMe - a competitively priced transcription service that readily integrates with NVivo. The media files are transcribed and automatically downloaded into your NVivo project, ready for analysis. For more information about how it all works, refer to Purchase audio or video transcripts:
- Transcribe them yourself using NVivo's transcription tools - this can bring you closer to your data as well as giving you a head start on analysis:
#2. Group the responses to each question
If your interviews have already been transcribed, you can import the documents directly into NVivo and get started on analysis.
If you’re working with semi-structured interviews (where participants are asked the same set of questions), you can use heading styles to automatically organize the responses.
For example, you can gather all the responses to Question 1 in one place for easier analysis.
Refer to Automatic coding in document sources for step-by-step instructions.
If your interviews are more free ranging and conversational, you can use other tools to organize the content by theme.
#3. Find and catalogue themes to make sense of the data
Thematic analysis involves making sense of what your interview participants are saying:
- What main points are they making?
- What surprising perspectives do they have?
- How do their ideas differ? What are the points of commonality?
NVivo gives you ways to get a broad feel for what themes are in the data and it also lets you drill down into the material for deeper analysis.
For example, you can run a quick Word Frequency query to see which words your participants are using most often. The resulting word cloud can give you early insight into emerging themes – and it’s a fun way to ease yourself into analysis:
Taking a more thorough approach to thematic analysis, you can read through each interview and ‘code’ the emerging themes. This involves selecting interesting comments and putting them into containers called ‘nodes’. At any time, you can open a node to see all the references you’ve gathered there.
NVivo offers plenty of ways to speed-up the coding process – and if you’re working with NVivo Plus you can use the Automated Insights feature to find themes automatically.
Refer to About Automatic Coding Techniques for the details.
#4. See the connections between themes and move toward analytical insight
As you code your material by theme, you’ll start to develop a list of nodes. At regular intervals, you can groom this list – checking whether related themes could be grouped together in a hierarchy.
This is not just ‘good housekeeping’ – it’s a vital step in the analysis process and helps you to see the connections between emerging themes:
You can open any node (by double-clicking on it) to view all the content gathered there but you can also run queries to retrieve your data in revealing ways.
For example, you could see where participants talked about ‘water quality’ in terms of ‘development’ – or where ‘policy’ came up in discussions about ‘water quality’.
NVivo lets you query and visualize your data in all sorts of ways – refer to Move forward with queries and visualizations to find out more.
#5. Make comparisons between participants
If you want to compare what your interview participants say based on attributes like age, gender or location – then you can create a ‘case’ for each person and assign the demographic attributes.
This video gives you a quick overview of how cases work in NVivo:
Creating ‘cases’ for interview participants, paves the way for powerful queries and visualizations. For example, you could create a matrix to see how men and women respond to a selection of themes:
#6. Stay organized and focused on your research design
In the thick of data analysis, it can be easy to lose sight of your research question.
Gathering your material into theme nodes and organizing these nodes in a ‘sensible’ hierarchy helps you to stay organized and focused.
Mind maps in NVivo are another great way to consolidate your thinking. For example, you could visualize your conceptual framework in a mind map and update it as your thinking evolves:
You should also consider creating a project journal in NVivo.
Keeping an audit trail of your challenges, assumptions, decisions and epiphanies will come in very handy when your supervisor (or client) asks a difficult question.
The journal tells the story of your project, makes your decisions transparent and helps you avoid that terrifying blank page when it comes to writing up.
Find out more
These 6 approaches to thematic analysis are just the tip of the iceberg and we’ll expand on them in upcoming posts.
To find out more – check the online help, watch video tutorials or join the NVivo Users Group on LinkedIn.
How about you? What are your tips for analyzing interviews? Tweet us at @QSRInt using the #NVivotips hashtag.