interview and focus group should be summarised to record ideas and comments that were made. Full transcription allows more detailed analysis, and also establishes a permanent written record of the discussion that can be shared with other interested parties (if confidentiality arrangements allow). However, transcription also takes time and may add cost, and may not be necessary if detailed notes or an interview summary is prepared.

Whoever collected the data should also be involved in the analysis. Even a full transcript does not reflect the entire character of the interview, and non-verbal communication, gestures, and behavioural responses are not reflected in a transcript. The tone with which words are used may be an important source of information and can alter the interpretation of a statement.

An initial group debrief can be very useful in summarising themes and planning analysis.

Analysis and interpretation of interview and focus group data require judgement and care, to identify key themes and ‘decode’ meaning in what people have said. Analysis of qualitative data should identify the main themes (in your words) and illustrate these with quotes (in the words of respondents). The analysis should also identify ‘outlier’ positions – perspectives that are equally valid and important, but less common.

Coding

Coding is a useful strategy when working with a large volume of qualitative data:

• The first step is to create some categories. Start by reading the first few pages of comments and identifying consistent themes. For example, in response to the question ‘What do you like about working at this school?, themes might emerge like ‘Friendly/supportive environment’, ‘Passionate/skilful colleagues’, ‘Diverse student population’ and so on.
• When coding open-enders in a spreadsheet, start with each comment in its own row. Each theme then gets its own column. Ensure there is an ‘other’ column.
• Now for each comment, put a ‘1’ in each of the columns where that theme is identified.
• If there is a need to clarify, expand or split a theme during coding, pause at that point to check comments that had previously been tagged in that theme and reapply the new code-frame.
• At the end, review the comments coded ‘other’ and ensure that they are genuine outliers.

Once qualitative data have been coded, the data should behave like a multiple response question, allowing additional analysis to be undertaken.

For further guidance, see the general notes on working with qualitative data in the School Excellence Framework evidence guide.