Analysis
The group discussions were audiotaped and transcribed in Spanish by a bicultural, bilingual
research team member. The Spanish transcriptions were then translated into English.
Transcriptions in Spanish were verified by 2 bilingual research team members, and each English
translation was reviewed and verified by 2 bilingual research team members for accuracy. The
research team analyzed the text to identify barriers to diabetes management using thematic
content analysis. Line-by-line coding of interview text was conducted by members of the team
and mutually agreed upon. In content analysis, codes are mutually exclusive, meaning every
statement is coded with no more than 1 code. Codes were derived both theoretically, taking into
account the research questions, and also from the transcripts directly. Codes were then placed into broad categories and then collapsed into major themes that answered the research questions.
A similar process was used to analyze the final question that was specific to family members.
Credibility of the data was enhanced by using bilingual and bicultural interviewers.39 Member
checking was used to ensure the accuracy of the interpretation and theme extraction.