In 2011, the LW was introduced as an elective longitudinal course with in the context of Studium Fundamentale- a competence-oriented range of additional studies in humanities, arts, philosophy and other fields that are obligatory for students of all disciplines at Witten/Herdecke University [35, 36]. The development of the LW was a joint venture of staff and students. Students participated in the conceptual design, as well as in implementation, organization, and teaching . The longitudinal LW structure comprises one intensive day in class (8 h) and six subsequent meetings of 90 minutes each on a weekly basis; the schedule may vary in accordance with semester planning. The LW concludes without a specific examination, but proof of performance can be issued to gain ECTS (European Credit Transfer and Accumulation System). Attendance in >80% of the course was mandatory to gain credit points. More information about the concept and the content of the LW is available upon request from the first author (MT).
An explorative qualitative research design was chosen to reach out for the nature of a phenomenon without having formulated any hypothesis. The aim is to get direct insight into the experiences of the participating students and to understand how the phenomena manifests in different perceptions .
The authors followed the Standards for reporting qualitative research (SROR) in presenting and structuring the research question, data collection, sampling, data analysis and conclusions resulting therefrom . Earlier theories and models on learning are duly considered, but do not dominate, the research process, analysis, or conceptual design. Procedures are inductive and take the form of an iterative research process including regular validation, expansion, and cross-checking by various other researchers.
For the survey, the authors used the Qualitative Short Survey (QSS) method. This instrument of intra-method mixing permits a meaningful combination of qualitative and quantitative data (triangulation). In this manner, subjective findings can be evaluated with the help of quantitative survey data, participants can be assessed individually, groups can be compared, and elaborate statistics can be produced .
This type of survey has the following advantages for the present study and the available human and financial resources:
meaningful format of evaluation which ensures anonymity and reduces distortion in response behavior caused by social desirability, compared to personal in-depth interviews 
allows comparison between different subgroups 
permits time efficient on-site evaluation immediately following the seminar, with a higher response rate than online evaluation at a later time.
Assessments by LW participants underwent analysis, and the questionnaire used in this case was specifically designed for evaluation of the LW. It comprises 4 pages with sociodemographic questions, 13 semi-quantitative and three open-ended questions, as well as a self-assessment of one’s own abilities and an overall rating.
For the purposes of this study, three questionnaire elements (Additional file 1) are analyzed asking for a reflection on the development of the respondent’s learning behavior and reveal direct information about medical students’ learning life.
We used two open-ended questions with the option of free-text answers:
Have you noticed any changes in your learning life due to your participation in the LW? If yes, to what extent?
How do you evaluate this change in general? (Satisfaction)
And thirdly, a rating on a quantitative scale:
The overall grade I give to the LW is (1–6): ___
(German system of school grades: (1) = Very Good; (2) = Good; (3) = Satisfactory; (4) = Sufficient; (5) = Poor; (6) = Deficient)
Questionnaires were distributed to all participants at the end of the last session of the LW, immediately filled in by students and collected subsequently. Students were informed that participation in the evaluation was voluntary. Completion of the entire questionnaire took between 12 and 18 min approximately.
Ethics approval and consent to participate
Data collection was anonymous. Students were advised that completion and return of questionnaires were voluntary, and that non-participation in the evaluation had no negative consequences for attendance at the LW nor potential proof of performance. Participants were free to decide how much time to invest in completing the questionnaire, how profound, detailed and precise their answers should be, and how much personal information they were willing to give. Thus, informed consent was obtained from all subjects involved in the study. A vote of ethical approval was obtained from the Ethics Committee of Witten/Herdecke University (submission no. 205/December 2019). So, all methods were carried out in accordance with relevant guidelines and regulations.
Available evaluations from 2011 up to 2019 form the basis of this study. In this period of time, the LW was conducted 17 times. Due to our focus on medical education, only evaluations made by medical students were included in the study. Those by students of other disciplines as psychology, dental medicine, economics and cultural studies were excluded for the study at hand.
Fifty students filled in the evaluation of whom six did not enter their demographic data, but did complete their evaluation. Their answers were included in the qualitative analysis. The response rate was 30.67%. The average age was 22.5 years (between 19 and 30 years of age). Participants (28 f/16 m) had completed 2.7 semesters on average (between 1 semester min. and 9 semesters max.). Responses were heterogenous in terms of length, form, and depth. Some evaluations were short lists of key aspects. Other participants wrote long paragraphs using full sentences. All transcripts were translated and back translated from German to English for publication purpose.
Three student assistants and the author (MT) transcribed the answers given in the questionnaire. The basic technique used to analyze the data is the qualitatively oriented content-analytical method of structuring content analysis with subsequent frequency analysis in adherence to Mayring . Following this method, an iterative step-by-step procedure served to inductively develop a system of categories. The authors used the computer software MAXQDA2018 to analyze complex non-numerical data . The software SPSS Statistics 25 was used for the quantitative analyses for research question number three and the descriptive demographics statistics.
The content-analytical criterion of inter-coder reliability was applied to test the reliability of qualitative findings. Data was comprehensively analyzed by a researcher who is experienced with the method (MT) and two students (MB & RS) who assisted in the design and the conduct of the LW. They analyzed a randomized excerpt of the data to permit the inclusion of the student perspective on the findings in the process of analyzing and interpreting results. Two other researchers, familiar with the method, encoded up to 50% each of the entire data material (DT & FA). Assignment of material was randomized. As a consequence, the coding-team was a heterogenous group of various researchers with different personal and professional experiences and preconceptions resulting therefrom .
The reliability of the dataset derives from the time of the measurement (17 realizations within a period of eight years), the sample size (N = 50) and the setting of the workshop (longitudinal, attendance check to gain ECTS, students’ involvement).
Firstly, the reliability of the analysis derives from the transparency of the process (by depicting every step in detail, the analysis becomes comprehensible and assured, that the results are valid and reliable). Secondly reliability of qualitative research is high, if intersubjectivity (plausible description of the iterative rating process and presentation of the results with reflection) is achieved [38, 45]. The five encoding schemes were disclosed, presented, jointly considered, and discussed in an inter-coder-reliability-team meeting. The researchers specifically tried to identify divergencies in the interpretation of terminology and thus to obtain a clear picture of the respective degree of subjectivity in such interpretations . Excerpts from the material were studied together to see whether essential qualities of students’ statements had been appropriately and congruently summed up in encoding. The emerged encoding schemes coincided; incongruous notions were defined more distinctly. Discussion topics were conclusiveness, precision, and logical congruence in the definition of the encoding schemes and encoding guidelines. Congruent and contradictory elements of the five coding schemes were debated, and mutually agreed definitions were established. Based on this the definitions a final encoding scheme and encoding guideline were finalized by MT and used in the further elaboration of the model and the analysis .