Data collection and analysis

Data collection was conducted using 18 semi-structured in-depth individual interviews by the researcher (SH) and 2 focus groups sessions in which the researcher (NY) was also present. SH was responsible for data collection and coding. SH has been working as a director of staff development unit for several years in the hospital, so she is familiar with clinical settings and clinical teachers and has experience in conducting qualitative research.

The focus groups were used observing that vis-à-vis debates between faculty clinicians and medical education and e-learning professors as specialists would result in more valuable findings. So, it was decided to use this potential.

First, they were asked open-ended questions about their experiences with electronic FDPs and the strengths and weaknesses of these courses, and then they were asked to express their own preferences. Other questions were asked using the participant’s comments. Each interview lasted for 30–80 mins, and the focus groups lasted for 120 mins.

To better use the comments and views of the participants, the researcher recorded the interviews with their consent and wrote important and significant points. All of the participants agreed with digital voice recording. All interviews (individual and focus group) were analyzed with the conventional qualitative content analysis method in three phases: preparation, organization, and reporting. These steps are applicable to both inductive and deductive analysis processes, but they are different in method of work.16

The preparation phase included selecting the unit of analysis and obtaining a sense of whole. In this study, a complete interview, which could be regarded as a meaningful unit, was chosen as the most suitable analysis unit. First, the researcher (S H) transcribed the interviews and focus group discussions verbatim; then, she listened to the recorded voice files again and compared them with the text. Data analysis started with several readings of the text to achieve immersion and arrive at a general understanding of the content.

The organization phase included open coding, creating categories, and abstraction.16 So, it was started carefully by reading the text word-for-word. Then, the meaning of each unit was determined and labeled with a code. The codes were compared based on similarities and dissimilarities, and the similar codes were integrated as one category. Abstraction involved emerging and grouping of subcategories and categories with similar meaning to achieve the main goal of research.

Finally, the researcher used the constant comparison to clarify the meaning of each element and once again decide on the categorization of information in the earlier stages and continued this process until satisfactory results were obtained. All of the steps were reviewed at meetings with the rest of researchers; in the case of ambiguity, the interviews were revisited and the bugs were corrected. The validity and accuracy of the analysis process were assured continually.

The reporting phase, as its name suggests, involved reporting the analysis process and the results. After conducting several interviews, the researcher began to group and encode them according to the main findings. By conducting any interview and analysis, the classifications were reviewed and, if necessary, merged or increased. Subcategories were formed according to similar features, and the categories’ name showed their contents. A total of 309 initial codes were extracted by considering the overlapping themes and merging them. They were classified into five categories and several subcategories.