116 Chapter 5 facilitate an environment for participants to be able to feel safe and to learn to recognize the moral components of the dilemmas and stimulate moral reasoning (Stolper et al., 2016). The facilitator encourages participants to develop a dialogical attitude of mutual respect, ask open questions and postpone judgements (Weidema, 2014). Studies on MCD in health care show overall positive evaluations (Hem et al., 2015; Janssens et al., 2015; Molewijk, Verkerk, et al., 2008). Originating in health care, MCD has now found its way to other professions, such as the Armed Forces and Contra terrorism (Kowalski, 2020; Van Baarle, 2018). Context and aim of the study In 2017, a training programme was developed by the Dutch Custodial Institutions Agency (DCIA) of the Ministry of Justice and Security for all Dutch prison staff which, among other goals, aimed to support the prison staff in dealing with ethically challenging situations in their practice, and thereby increase their moral craftsmanship (MCS). Within the specific prison context, by the DCIA the concept of MCS is regarded as the awareness and recognition of ethical challenges, and the competence and willingness to be open to opinions of others in a constructive way (DCIA, 2016). Part of this training programme consisted of the implementation of a series of MCD sessions in which prison staff can bring up their moral challenges and jointly reflect on them through a critical yet constructive dialogue. To find out if and in which way a series of MCD sessions contributed to the MCS of prison staff, we developed a questionnaire that covers various elements of MCS. In this article we use the first version of the self-developed Moral Craftsmanship Questionnaire (MCSQ) to investigate the possible impact of MCD on MCS. METHODS Research design Five Dutch prisons took part in this research (Leeuwarden, Vught, Almere, Nieuwegein, and Zwaag). The institutions were classified in a control group and an intervention group. MCD was implemented at the intervention locations, and no MCD sessions or similar training took place at control locations during this study. Initially, two prisons were assigned to the intervention group (Nieuwegein and Zwaag), and two comparable prisons to the control group (Almere and Leeuwarden). Due to the mid-2018 closures of the prisons of Zwaag and Almere – during the research period – two additional locations were selected (Leeuwarden as an intervention location and Vught as a control location).