Luppo Kuillman

Chapter 3 64 suggested by Bentler and Bonett, we used multiple criteria to judge the model fit (Bentler & Bonett 1980). These criteria were: (a) non-significant chi-square, indicating that a non-significant proportion of variance in the data remains unexplained (Kline 1998); (b) root mean square error of approximation (RMSEA) less than 0.06, indicating a good fit to the data (Browne & Cudeck 1993); (c) comparative fit index (CFI) > 0.97, indicating good fit (Schermelleh-Engel et al. 2003); and (d) Tucker- Lewis Index (TLI, also known as the non-normed fit index). Both CFI and RMSEA were used because it has been argued that they provide more stable and accurate estimates than several of the other fit indices (Hu & Bentler 1999). Bivariate and statistical analyses were performed using SPSS 25 for Windows. All multivariate statistical analyses were executed using STATA 14.0. Measurements Sociodemographic characteristics For comparability testing of the NP and PA samples, background characteristics were collected, namely: gender, age, religious beliefs and political affiliation. Additionally, respondents were asked to typify their working environment as (a) ‘working in a hospital’; (b) ‘in general practice’; (c) ‘in mental health care’; (d) ‘in mentally disabled care’; or (e) ‘other’. Validity, reliability and rigour of the instruments Three instruments were used as indicators for the constructs of interest: (a) the Defining Issues Test, measuring the level of moral reasoning; (b) the Big Five Inventory, for measuring personality traits and construing the meta-traits; and (c) the Moral Disengagement Scale, measuring the propensity to morally disengage. These three instruments were translated into Dutch according to the procedure proposed by Guillemin and colleagues (Guillemin et al. 1993). First, the original English versions of the questionnaires were translated into Dutch by two certified translators working independently. Second, two more certified translators each translated these Dutch translations back into English. The resulting English versions were compared with the originals and all discrepancies were discussed by three researchers who spoke both Dutch and English. The remaining discrepancies were discussed with a native English speaker from the University of Groningen Language Centre. For a detailed description of the instruments, see Appendix 1.