Objective: The construct validity of checklist and global process scores for an objective structured clinical examination (OSCE) in psychiatry was assessed. Multiple regression analysis was used to predict psychiatry OSCE scores from the clinical skills examination, an obstetrics/gynecology (OB/GYN) OSCE, and the National Board of Medical Examiners (NBME) psychiatry subject examination.
Methods: Archival data from two successive classes of third-year medical students (1999-2000, N=142; 2000-2001, N=144) were aggregated and analyzed.
Results: The pattern and magnitude of convergence and discrimination were indicative of adequate construct validity for both the psychiatry checklist scores and the global process score. Clinical skills examination scores for history taking, interpersonal skills, and physical examination were related to psychiatry OSCE scores that reflected the same skill set. Construct validity was fairly higher for the global process rating.
Conclusion: Evidence of construct validity of a psychiatry OSCE was obtained from multiple measures of performance, including the clinical skills examination. Findings lend support to the continued use of checklist and global process evaluations as part of OSCEs in psychiatry.