S.TRUEMAN PhD THESIS 2016

105

qualitative component of a mixed methods design. A number of limitations remained however, namely that verbatim transcripts of interviews were not analysed; there was limited explanation of the process of thematic analysis used; limited discussion of the qualitative data findings; and no differentiation between professions in the findings. Chur-Hansen et al. (2004) used a pre and post-evaluation questionnaire tool with no discussion concerning validation. The results of all questions are not reported although mean scores are provided for correct answers using z scores and p values and are significantly positive. The post-course satisfaction questionnaire is reported using mean scores and again, positive improvements are rated highly, but not according to profession, which limits the paper’s usefulness for nursing. No limitations or bias were identified. The qualitative phase does not outline the methods of data analysis; verbatim transcripts were not utilised; and themes arising from the interviews were not clearly identified. The overall results have limited generalisability (Chur-Hansen et al., 2004) as they relate to one training course in one state. In Aoun and Johnson (2002) there was also no pre-course testing, resulting in baseline at completion of course and then follow-up at four months. Questionnaires containing both Likert scales and open-ended response questions, related solely to assessing the training, were utilised at both stages of data collection. The follow-up questionnaire contained Likert scales to assess levels of knowledge and practice and some opened questions rating impact of training on clinical practice and support from management. The first questionnaire response rate was good at 87%; there was a good spread of participants from across the state but the results were not reported using a breakdown of the professions of attendees. Results were descriptive in reporting the