Martine Kruijtbosch

77 App l i cab i l i t y o f t he Au s t ra l i an Pro f es s i ona l E t h i c s i n Pha rmacy t es t RESULTS Three hundredninety respondents (81%early career pharmacists; 19%pharmacist supervisors) completed the PEP-NL test. Fourteen pharmacists (all early career pharmacists) ranked two or more meaningless statements, and their questionnaires were therefore discarded. The PCA was performed for the data of the remaining 376 respondents. Of these respondents, 63% were women and the median age was 27 years (IQR = 25 - 35 years). The PCA analysis confirmed the construct validity of the PEP-NL data. The KMO index was 0.74, and the Bartlett test was statistically significant ( p < .000). The scree plot showed small increments in explained variance beyond 5 components. Therefore, the PCA-varimax rotation was performed with 3, 4 and 5 components to extract. When the rotation was set at 4 components, the explained variance increased with 5% to 32% and when set at 5 components with another 4% to 36%. However, the statements that correlated in the fourth and/or fifth component, did not provide new moral reasoning schemas on top of the first three moral reasoning schemas. So did the component with statements that represented ‘rules and regulations’ split in two, but all statements were related to aspects of law or regulations related to the profession. The same applied to the statements that represented the ‘business orientation’ moral reasoning schema. Therefore, we set the number of components to extract to 3. The three components explained 27% of the variance in the data and had eigenvalues larger than 2. Table 2 provides the scenario statements’ correlation loadings for the three PCA components. Table 2 shows these loadings per scenario (moral dilemma scenarios 1, 2 and 3, Appendix). As illustrated in Table 3, the internal reliability of the three PCA components of the PEP-NL data showed Cronbach’s alpha values of 0.60 (first component), 0.63 (second component) and 0.54 (third component); for the test as a whole, this value was 0.63. Comparing eligible statements and schemas The comparison of eligible statements per component resulted in two moral reasoning schemas that were also found in the Australian PEP study - ‘rules and regulations’ (conventional schema) and ‘business orientation’ (pre-conventional schema) - and in one new moral reasoning schema, which we labelled as ‘professional ethics’ (perceived as a post- conventional schema). The statements that loaded as the ‘professional ethics’ moral reasoning schema deviated completely from the statements that loaded in the PEP study as the post- conventional schema (patients’ rights schema). Table 4 shows the three components and eligible statements.