The Process of Finding Themes and
Unique Perspectives
Analyzing qualitative data from interviews, focus groups, and open-ended survey questions involves a systematic, iterative process of identifying and making meaning from common themes, as well as from unique or dissenting perspectives.
Qualitative analysis can surface specific information and feedback that help tell the story of participants’ experiences in the program and how or why they did or not achieve expected outcomes. Qualitative data can be used to contextualize and enrich quantitative data to tell a more holistic and accessible story than numbers can alone.
- Be wary of the temptation to quantify your qualitative data. It is better to call out certain respondents’ answers or talk more generally about how many interviewees expressed a particular idea. For example:
- Good practice: “Many participants noted liking the enthusiasm of the trainer. In particular, one training participant said, ‘She was so excited about the material it was contagious!’”
- Generally avoid: “A majority of participants (74%) noted liking the enthusiasm of the trainer and 20% of those participants linked the enthusiasm of the trainer with increased learning.”
- Weave the results of your quantitative analysis together with your qualitative data into an integrated story. Use qualitative data to illuminate the quantitative data and draw the reader (the user of your report) into the story.
- Good practice: “Many participants noted liking the enthusiasm of the trainer. In particular, one training participant said, ‘She was so excited about the material it was contagious!’”
Preparing qualitative data for analysis
Ensure transcripts from your interviews or focus groups are clear and readable:
- Clarify meaning where needed by adding additional words in brackets to increase readability of sentences. Where appropriate, make fragments into full sentences.
- Write out abbreviations and acronyms as full words.
- Move text to its appropriate place (e.g., if Q5 was answered in response to Q1 for some reason, move the response text to Q5).
- Flag good quotes. Then remove names or identifying information, unless you receive permission to attribute the quotefrom the participant.
- Write out abbreviations and acronyms as full words.
Often, it’s helpful to include interviewer notes and insights at the end of an interview or focus group, for example:
- “This was the morning of a tragic event in the news and it was hard to get the group to focus.”
- “This interviewee had great insight and seems really connected to the questions we were asking.”
It can be helpful to track emerging themes, concerns, and notes in brief memos written shortly after interviews or a focus group. Memos may be especially useful if data collection happens over a long period of time (i.e., several weeks or months) or if there’s a gap in time between data collection and analysis.
Before analysis, you may find it helpful to combine data. Options include:
- Creating one Word document per question by pasting in all answers and color-coding by respondent.
- Creating one Excel file with one question per tab and one response per row. To indicate respondents you can color-code, or create a “Respondent ID” column and assign numerical values.
- Uploading the data to a qualitative analysis software package such as Dedoose, AtlasTi, or NVivo. Qualitative software is particularly useful for large datasets.
- Creating one Excel file with one question per tab and one response per row. To indicate respondents you can color-code, or create a “Respondent ID” column and assign numerical values.
Example of Qualitative Analysis Using Excel
You can use Excel to code and track themes in your qualitative data. You can create new themes and sub-themes in additional columns and clearly see what themes each participant’s responses relate to.
- Create a new Excel worksheet tab for each question so your data stays organized and manageable.
- Make sure each participant has a respondent ID so you can track who is saying what across questions. It’s important to ensure you aren’t giving a few participants more voice in your interpretation and reporting – this can ensure you use quotes from different participants and weight themes and unique ideas appropriately.
- Sum each theme like the example above for a quick pulse on how prevalent a theme is compared to other themes. However, resist the temptation to present the summed numbers in your reporting.
- Make sure each participant has a respondent ID so you can track who is saying what across questions. It’s important to ensure you aren’t giving a few participants more voice in your interpretation and reporting – this can ensure you use quotes from different participants and weight themes and unique ideas appropriately.