Purpose: To investigate the construct validity of the miniclinical evaluation exercise (miniCEX).
Method: Forty faculty participants from 16 internal medicine residency programs enrolled in a randomized, controlled trial of faculty development. Using a standard nine-point miniCEX rating form, participants watched and rated performances of standardized residents on nine scripted clinical videotapes depicting three levels of performance (unsatisfactory, marginal/satisfactory, and high satisfactory/superior). The nine-point rating scale was 1-3 = unsatisfactory, 4-6 = marginal/satisfactory, and 7-9 = superior. The performances were rated for three clinical skills, history taking, physical examination, and counseling.
Results: For each of the three clinical skills, the faculty participants were able to successfully discriminate among the three levels of performance using the miniCEX scale. Differences among ratings of the three performance levels were statistically significant; however, the range in ratings among the participants for each videotape was wide.
Conclusion: The authors believe this to be the first study to document the construct validity of the miniCEX. Although the miniCEX appears to have reliability and construct validity, further research is needed to improve individual faculty observation skills and reduce interrater variability.