Th J Ten Cate | Search PubMed

PubMed2005

Progress testing with short answer questions.

The concept of progress testing was developed in the 1970s. Significant features of progress tests are that the content is not linked to any specific course or unit, and that it reflects the final objectives of the curriculum as a whole. The questions are taken from a broad domain and cover a range of disciplines. Furthermore, the test is taken repeatedly over a period of time, to monitor students' progress. Known progress tests all use closed format questions. In 2002-2003 the University Medical Center Utrecht initiated a progress test with short answer questions. The test consists of 40 cases, each with a clinical and a basic science short-answer key feature question. This differs from other progress tests that use close format items, but also in the philosophy of mastery level testing and in the deliberate linking of basic science concepts to clinical case vignettes. The first four executions of the test show high internal consistencies (Cronbach's alpha 0.85 to 0.87) and satisfactory item parameters. The effort of marking answers is reasonable, the effort of writing case vignettes with short-answer items is less than writing MC-items if similar test reliabilities are to be achieved.

Education, Medical, Undergraduate↗

PubMed2004

Factors predictive of successful learning in postgraduate medical education.

PURPOSE: To establish which personal and contextual factors are predictive of successful outcomes in postgraduate medical education. METHOD: We performed a follow-up study of 118 doctors on a postgraduate occupational health training programme on the management of mental health problems. The following personal and contextual variables were measured as potential predictors of outcome: gender; age; years of experience as a doctor; university of graduation; learning style (Kolb); present employer (occupational health service), and educational format (problem-based or lecture-based). The main outcome measures were: scores on knowledge tests consisting of true/false and open answer questions, and performance in practice based on self-report and performance indicators. To determine the effect of potential predictive factors univariate analyses of variance and repeated measurement analysis of variance was applied. RESULTS: The mean scores of knowledge (P < 0.001) and performance (P = 0.001) of the participants increased after the educational programme. After multivariate analysis female gender was positively related to accruements in both knowledge and performance (both P < 0.05), independently of the influence of other factors. Accommodator learning style showed a relation with knowledge increase (P = 0.05), but had no influence on performance (P = 0.79). The problem-based educational format yielded a better performance outcome (P = 0.05), but had no influence on knowledge tests (P = 0.31). CONCLUSION: Gender and learning style were found to be related to an increase in knowledge. Gender was also found to be related to improvement in performance after a postgraduate medical education programme. We found no interactions with course design (i.e. problem-based or non problem-based learning formats), but further research could reveal other cues, suggesting practical consequences of student characteristics for course design in postgraduate training.

Adult↗