Among the many qualitative data analysis procedures, framework analysis is one that is easy to learn and welcomed by quantitative, qualitative and mixed methods researchers. The matrix/grid display and multi-purpose functions of thematic, case and thematic/case analyses all in one table earn its popularity among multidisciplinary research teams. Here is a summary of the steps, benefits and pitfalls of using this method. For detailed descriptions, please refer to the references.
Procedures
1) Transcription: Transcribing is a great opportunity to familiarize yourself with the interview (again) and to detect potential ideas for codes. Determine at this stage whether you want to do this yourself or outsource it to a professional service provider.
2) Coding: Start deductive coding and inductive coding with a few transcripts to come up with a preliminary coding framework.
3) Refining coding framework: Have team meetings with all or most members of the research team to revise the coding framework.
4) Applying the final/semi-final coding framework: Code the rest of the transcripts using the refined coding framework. This framework can change until the last transcript is coded.
5) Charting data into framework matrix: Create a framework matrix (see references for examples) and summarize the coding into the cells in the matrix.
6) Interpreting the data: Extract themes, devise relationships and/or discover explanations/causal relationships in the matrix table. These are your final results.
Benefits
1) All members of the research team can easily engage with the data in a table and offer their perspectives.
2) The columns allow for thematic analysis and the rows enable case analysis. Descriptions are “thick” and interpretations are more thorough.
3) An audit trail is found in the table, with each cell offering quotes to substantiate the themes.
Pitfalls
1) Quantitative researchers are tempted to turn the tabular results into numerical analysis, which is inaccurate because of the sampling procedure for most qualitative datasets.
2) It is time-consuming to summarize the data and fill in each cell in the table. It also takes experience and acuity to write meaningful, data-driven, usable entries.
3) Since this is a qualitative exercise, rigour/reflexivity/transparency still need to be present.
References
Gale NK, Heath G, Cameron E, Rashid S, Redwood S. (2013). Using the framework method for the analysis of qualitative data in multi-disciplinary health research. BMC Medical Research Methodology 13:117.
NatCen Learning. (2012). The framework approach to qualitative data analysis. UK: NatCen Learning.
Ritchie J. et al. (2013). Qualitative Research Practice, London: SAGE.
Procedures
1) Transcription: Transcribing is a great opportunity to familiarize yourself with the interview (again) and to detect potential ideas for codes. Determine at this stage whether you want to do this yourself or outsource it to a professional service provider.
2) Coding: Start deductive coding and inductive coding with a few transcripts to come up with a preliminary coding framework.
3) Refining coding framework: Have team meetings with all or most members of the research team to revise the coding framework.
4) Applying the final/semi-final coding framework: Code the rest of the transcripts using the refined coding framework. This framework can change until the last transcript is coded.
5) Charting data into framework matrix: Create a framework matrix (see references for examples) and summarize the coding into the cells in the matrix.
6) Interpreting the data: Extract themes, devise relationships and/or discover explanations/causal relationships in the matrix table. These are your final results.
Benefits
1) All members of the research team can easily engage with the data in a table and offer their perspectives.
2) The columns allow for thematic analysis and the rows enable case analysis. Descriptions are “thick” and interpretations are more thorough.
3) An audit trail is found in the table, with each cell offering quotes to substantiate the themes.
Pitfalls
1) Quantitative researchers are tempted to turn the tabular results into numerical analysis, which is inaccurate because of the sampling procedure for most qualitative datasets.
2) It is time-consuming to summarize the data and fill in each cell in the table. It also takes experience and acuity to write meaningful, data-driven, usable entries.
3) Since this is a qualitative exercise, rigour/reflexivity/transparency still need to be present.
References
Gale NK, Heath G, Cameron E, Rashid S, Redwood S. (2013). Using the framework method for the analysis of qualitative data in multi-disciplinary health research. BMC Medical Research Methodology 13:117.
NatCen Learning. (2012). The framework approach to qualitative data analysis. UK: NatCen Learning.
Ritchie J. et al. (2013). Qualitative Research Practice, London: SAGE.