Predictive validity of uniform entrance test to select health professional : a psychometric study at a private university

To determine the predictive validity of the Uniform Entrance Test for academic performance in the first two years of three health science programs as demonstrated by scores on the first four semester examination. Methods: It is a correlation study using prediction research design. A retrospective analysis of admissions data and of academic performance of students admitted in undergraduate programs of medicine, dentistry and pharmacy during 2010-2012 was taken. The independent and dependent variables were entry test score and academic achievement measured as semester score respectively. Descriptive measure for all variables was determined. Spearman\u27s Correlation coefficient was computed to determine the association between entrance test score and semester result for three groups. (p value = 0.05). Simple linear regression was applied to see the effect of gender, prior education and entrance test score over student\u27s performance. Results: Majority of the students were from MBBS (61 %), majority of the study participants were females (65 %) and about half of the candidates (53 %) were from the Pakistani system of education. For MBBS the highest correlation coefficient was observed for semester one is = 0.334 and lowest in semester four rs = 0.208. Within the BDS program both highest and lowest correlations were in semester one. In the Pharm- D program significant correlation were seen in 2010 but not in the subsequent years. Entrance test score statistically significantly effect student\u27s performance, F (2, 283)=21.1, P-value \u3c0.05, R = 0.359, & R2 =0.129. The pre admission tools predicted 12.9 % of the variance in student performance. Conclusion Better performance in the entrance test is predictive of higher semester scores and more likelihood of achieving higher scores in the first year as compared to the subsequent years. The Uniform Entrance Test is more predictive of performance in MBBS and BDS than in Pharm-D