620 The Professional Counselor | Volume 10, Issue 4 (2015) recommendations, interview questions were open-ended, “why” questions were avoided, questions were as neutral as possible, and questions were asked one at a time. The interview protocol was piloted prior to the study commencing. Several questions were divided into two questions to ensure that only one question was asked at a time. A conventional semi-structured interview was used with follow-up questions allowed to ensure understanding of the participant responses. Each participant reviewed and signed the informed consent agreement approved by the last author’s IRB prior to the interviews. Data Analysis The team coded, categorized, and analyzed data from the 15 interview transcripts. Transcripts were coded using an open verbatim coding process, followed by the development of axial codes using the constant comparative method to create themes that emerged from the data (Corbin & Strauss, 2015). The research team used a shared Google Sheet to document codes, descriptions for the codes, and, later in the coding process, broader categories. All members of the research team had access to each transcript in a Microsoft Word document through Dropbox. Each line of the transcript was read and discussed by teammembers, and then a verbatim portion(s) of the line that answered the research question was copied to the shared Google Sheet. Once the code was selected, the group determined a description for the code, using the context in the transcript as a guide. The first three of the 15 transcripts were coded collaboratively as a team during online coding sessions. This was necessary to establish consistency among the researchers and to increase trustworthiness in the data (Corbin & Strauss, 2015; Merriam & Tisdell, 2016). Transcripts 4 through 11 were coded in pairs and the final four were coded individually with a second researcher available if there were questions. The first and second authors established a code key upon the completion of the first four transcripts. Following initial coding, the first two authors met three additional times to come to a consensus regarding the collapsing of codes. For example, the code “playing politics” was initially identified. After all transcripts were coded, this code only had two direct quotes for support. Therefore, during a subsequent coding team meeting, the first two authors determined that this code had limited evidence from the data and it was collapsed into the code “selecting and working with committee members.” Creating a code book made it possible for the team to track categories and to ultimately identify the themes that emerged from the data. Verbatim quotes of participants were noted and organized into themes. Themes were agreed upon by the first two authors and reviewed and supported by the third and fourth authors. The research team continued coding until completion, maintaining agreement on new categories and descriptions added to the code key. Any discrepancies were resolved by all teammembers, reaching consensus on the final coding. The following steps were adhered to by the team: 1) asking clarifying questions of each other to be sure every perspective was considered; 2) rereading previous lines of the transcript and reading ahead a few lines to better understand context; 3) allowing space to reflect on what each person was thinking and feeling about a code; 4) considering new codes when participant statements seemed to indicate different data points within the same line of the transcript; and 5) referring back to the research question when considering if a statement fit the purpose of the study. Trustworthiness To ensure a reliable process, the researchers adhered to a 4-step process proposed by Moustakas (1994). First, the researchers bracketed personal experiences and assumptions regarding what was instrumental in completing the dissertation process. For example, the research team discussed the nature of their own dissertation experiences as people of color at predominantly White universities;
RkJQdWJsaXNoZXIy NDU5MTM1