Data analysis

The only documentation of the FGDs was the audio-recorded discussions, and the analytical process concentrated exclusively on manifest content that was visible and obvious in the transcriptions of the FGDs. The 10 discussions were imported into Nvivo 11 Pro software [40]. The qualitative content analysis was performed in the following steps [41]: 1) every discussion was considered a unit of analysis, 2) data were reduced to meaning units emerging from the text and in conjunction with themes like confidence, teamwork, and motivational factors reflected in the interview guide using a coding framework of Nvivo nodes, 3) each meaning unit was condensed in order to preserve the core of the content while shortening it, 4) categories were created based on commonalities among the codes, as the most essential part of qualitative content analysis, and 5) the nodes were accumulated into three global categories.

The authors were involved in different steps of the data analysis. Initially, the first author (SE) read the transcribed text several times to gain an understanding of its overall content, while the interviewer (DM) reread the transcripts. The first author coded the transcribed text and meaning units were identified. After accumulation of meaning units, the remaining part of every unit of analysis was checked by the first author for meaning units that could have been overlooked. To ensure optimal categorization and to strengthen the credibility of the analyses, the interviewer checked the codes for correctness to avoid personal bias. Four of the authors (SE, DM, BK, LEB) evaluated the nodes and categories after all relevant meaning units were extracted from the units of analysis. All authors (SE, DM, BK, LEB, HK) discussed the findings and reached consensus. The categories emerging from the content analysis were systematized according to the research questions to explore the perceived learning features and learning outcomes from simulation training.