DOI: 10.36347/sjams.2016.v04i11.004

Pages: 3876-3882

The objective is to improve assessments by measuring the quality properties of the Objective structured clinical examination scores of 10, 3rd year Clinical Medicine students in the Obstetrics and Gynaecology department of a resource-limited medical school in Tanzania, using psychometric methods, in July 2015. This descriptive and cross-sectional study used literature review and structured observation as data collection methods. Students’ performances were assessed using checklists by 5 examiners in 4 OSCE stations. Stations 1-2 were manned and stations 3-4 were in written format. Permission to carry out the study and the ethical approval were obtained from Kampala International University and the Amana Hospital IRBs and management. Statistical analysis was carried out using Microsoft Excel and SPSS computer packages. The stations were too few and tasks in the stations barely covered 50% of the learning outcomes. The mean scores in the stations were between 34-91% and pass mark was set between 35-95%. The tasks were too easy in 3 of the stations, the variance was generally too high for a criterion referenced test and none of the means was centrally located. The scores were transformed to z-scores which eliminated all the extreme scores and prevented false awards of distinction and fail grades to students. Examiners’ error was very high at 89% and the internal consistency was weak (Alpha was 0.008). Item analysis was poor in stations 1-3 and the examiner who marked station 3 was identified to be a dove. The analysis recommended that stations 1 and 3 be discarded, station 2 be reviewed and station 4(whose properties were good) be banked for future use. Moreover, the department is adviced to generate between 8-15 OSCE stations in subsequent OSCEs for better reliability coefficients. The properties of the OSCE experienced in resource limited settings may not be similar to those practiced in the established medical schools of developed countries.