90 | Chapter 4 We distinguished three stakeholders that use the portfolio: trainees, supervisors and faculty. We specified one inclusion criterion: experience with the portfolio (6 months for trainees, 12 months for supervisors and faculty). Since there are hierarchical differences between the groups, with some depending on others for employment and/or assessment, we chose homogenous focus group compositions to ensure safety and facilitate the conditions for open communication.47 Furthermore, we wanted respondents to be able to participate at the medical centre they are affiliated with, to minimise the burden of commutes. Consequently, we planned to organise at least nine focus groups, to represent three stakeholders at three institutes. The delay that resulted from the COVID-19 pandemic, contributed to our decision to stop data collection after these nine focus groups. At this point a new version of the portfolio was about to be implemented and we did not want this to interfere with our analysis. While data/thematic saturation is often referred to in concern to the decision to stop data collection, Varpio et al. insightfully explain why the concept of saturation does not align with our contextualist world view. Likewise, this article substantiates why member checks were not performed.48 The initial approach of potential respondents was via an email, that encompassed information about the research project, practicalities about the focus groups and the informed consent form. When this email did not result in enough sign-ups, alternative routes were used to approach respondents (e.g. a presentation at a trainee congress). The respondents who participated received a gift card for their contribution. The focus groups were moderated by the first researcher and the other authors consecutively accompanied her to make observational memos. A semi-structured interview guide was used to guide the discussion. Throughout the study two versions of the guide were developed through discussion in the research team (see Appendix A). The first version did not contain any questions targeting learning or SRL. We were curious if SRL (related topics) would surface spontaneously in a discussion about portfolio use, as we wanted to know if and how respondents associated portfolio use with SRL. After five focus groups it was clear that there was limited discussion of SRL (related topics) when this was not encouraged. Therefore, the second version did contain more targeted questions focusing on learning. SRL was still not addressed directly, as we expected the range of discussion and understanding between respondents to be restricted by a concept as complex and multi-interpretable as SRL. All focus groups were audio recorded and verbatim transcripts were made of these recordings, personal specifiers were not transferred to these transcripts. Coding of the transcripts commenced after the first focus group, so relevant topics and themes identified during analysis of the first focus groups could be further explored in subsequent focus groups. Table 1 shows the characteristics of the focus groups.
RkJQdWJsaXNoZXIy MjY0ODMw