Ethnography was the guiding methodology for this study. Specifically, we assumed the role of observer-participant in that we were part of the setting, yet had minimal interactions with participants (Lofland & Lofland, 1995). When requested we did participate in program activities and shared our opinions during large and small group discussions. However, our involvement was limited and thus our presence caused minimal reactivity bias.
We recorded our observations in the form of field notes. The field notes included our descriptions of the setting and participants, reports of events, behaviors and interactions, and transcriptions of participant conversations. We also included our impressions of our observations, including our experiences and reactions to the instructors’ reentry discourses. Written field notes were expanded and typed within 2 days of each observation, and these notes were shared between the researchers to confirm accuracy, identify similarities, and to resolve any differences given our different racial and gender lenses (the first author as an African American man, and the other field researcher as a white woman). Program documents (e.g., handouts, reports, and the program workbook) were also collected and included in the overall analysis.
The analytic process consisted of coding and memoing (Miles & Huberman, 1984). As coauthors (Lea and Abrams), we inductively coded each field note to identify and compare instances in the data where instructors’ and volunteers’ job search strategies and messages concerning opportunities and barriers to employment were discussed and negotiated. Following initial coding, we developed analytic memos to identify, classify, and categorize emerging and final thematic categories. Once we identified the major themes, we displayed excerpts from the field note data in a thematic matrix (Miles & Huberman, 1984) and discussed the initial and final categories to ensure we grounded our analysis and interpretation in the data. Findings from the analysis were also shared and discussed with a research group to ensure the trustworthiness of the findings (Lincoln & Guba, 1985).