Predictive validity of the modified essay question

This paper examines the predictive validity of the modified essay question (MEQ) when applied to undergraduate physical therapy students. A 25-item MEQ was developed and pretested on a group of clinical experts. Following the pretest, the MEQ was administered to a group of 25 physical therapy students immediately prior to their second-level orthopaedic clinical placement. A reliability coefficient of 0.63 (intraclass correlation coefficient) was calculated for the test. An estimate of the predictive validity of the MEQ was obtained by correlating the student's MEQ scores with their clinical placement scores. A correlation coefficient of 0.65 was obtained between the MEQ score and the sum of the clinical reasoning and treatment planning scores. The results suggest that the MEQ may be a useful tool when predicting students' clinical reasoning and treatment planning skills. This paper also discusses several factors pertinent to establishing a reliable MEQ, such as the number and homogeneity of test items.