Developing a codebook to guide content analysis of expressive writing transcripts

This article describes a team-based approach to the development of a comprehensive codebook for multiple researchers to use during content analysis of the transcripts of the expressive writings of women (in this study, N = 89) with metastatic breast cancer. The codebook structure was developed iteratively by reaching a consensus on the analysis of shared transcripts to create an all-encompassing set of codes, with definitions, inclusion and exclusion criteria, and exemplar text from the transcripts. The Qualitative Solutions and Research International NVivo software program was used to maintain an electronic database of the consensus analysis of transcripts, information about each code, and a detailed log about the process of developing the codebook. The team ultimately created a comprehensive codebook that contained 27 codes with definitions, inclusion and exclusion criteria, and example text. The codes were verified by each team member through reanalysis of a set of shared transcripts that had been previously coded using an earlier version of the codebook. The team met to discuss individual coding and reached a consensus on the final version of the codebook. No new code was identified during the reanalysis, and there was fairly uniform agreement on the coding. The final version of the codebook will be used to guide each team member's individual analysis of the remaining (74) transcripts, which will be divided among the team. Periodic meetings are planned to discuss the individual analysis and to resolve any issue associated with using the codebook. As new codes are identified and agreed upon by the team, they will be added to the codebook. A team-based approach can facilitate the development of a practical and accurate codebook to guide the analysis of a large amount of qualitative data.