Data Preparation and Analysis. While in country all interviews were transcribed verbatim by a research assistant in Spanish. This was done to preserve the implications and richness of the data collected. All audio files recorded from interviews and focus groups were re-labeled by numerical audio codes to de- identify the participant. In addition, all transcripts were further de-identified by the researcher to protect the privacy of each participant and to maintain confidentiality. Upon her return, the researcher cleaned and de-identified all data prior to utilizing MAXQDA version 10 to code and analyze the data further (VERBI, 2011). The researchers involved in the larger project and their advisor worked closely to develop a codebook encompassing important themes that were present in the data. Researchers coded several interviews and compared results in order to edit codes and code definitions until they were agreed upon. The researchers completed several rounds of coding and editing code definitions before the codebook was finalized. Researchers then coded their own data individually by applying each of the 18 developed codes to all interviews and group discussions. The full codebook is presented in Appendix 5. After coding the interviews with fathers, the researcher systematically retrieved all coded segments by each individual code to begin analysis. The researcher analyzed each code individually by reading and summarizing all of the coded segments in each code and further sorting data into dimensions under each theme. Direct quotes from participants as well as summarized themes from each code were then compiled to develop results. The researcher translated all direct quotes verbatim from Spanish to English, and developed pseudonyms for each participant to protect his identity in the results.
Data Preparation and Analysis. All interviews were recorded using a recording application on a password-locked device. After reviewing the recordings, the interviews were transcribed verbatim using Microsoft Word and MAXQDA Pro Analytics. An initial codebook was developed in Microsoft Word using a deductive approach —common themes that were likely to appear within the data were based off literature review findings. After this initial codebook was developed, the codebook was tested on a single transcript from each of the “exerciser” and “non-exerciser” groups by using MAXQDA Pro Analytics’ coding mechanisms. Inductive codes were then added based on this trial coding and subsequent transcripts. Excessive codes were discarded as needed to streamline the process and develop a final codebook (see appendices). In total, 24 interviews were examined and coded. This data was stratified into two groups: 16 “exercisers” and 8 “non-exercisers”. After all transcripts were coded, coding was reviewed and reconciled codes to ensure accuracy. Two MAXQDA visualization features aided in performing thematic analysis: Coded Matrix Browser and MAXMaps. The Coded Matrix Browser allowed the researcher to view multiple documents and their coding frequency patterns. MAXMaps was used to view the relationship between codes and subcodes. These visualizations aided in the analysis, allowing the following themes to be characterized: perception about hair; perception about exercise; hair as a barrier to exercise; and methods to overcome hair as barrier to exercise. Relevant quotes were identified to emphasize these themes. Ethical Considerations The researcher acquired Collaborative Institutional Training Initiative (CITI) certification before designing and conducting this study. This study was reviewed and approved by Emory’s IRB for human subjects research. Any participant who desired to withdraw from the study could do so at any time. Participants were free to refuse any procedures such as having their BMI measured or answering any questions. Participants also had the right to prevent their research information from being used in the final project. To avoid issues of coercion, participants were told that they could withdraw from the study or refuse any parts of the study without forfeiting their $5 cash incentive. RESULTS
