Background: Standardized patients can be trained to assess the communication and interpersonal skills of medical students and graduates.
Purpose: The purpose of this study is to present data to support the psychometric adequacy of the communication ratings provided by standardized patients.
Methods: Using the data from testing of 3, 450 examinees over a 1-year period, a number of psychometric analyses were undertaken. These included a variance component analysis, the calculation of various validity coefficients, the comparison of communication ratings for select examinee cohorts and case characteristics, and the investigation of some potential sources of score invalidity.
Results: Communication skills scores are moderately correlated to other competencies (knowledge, skills) and may be influenced by candidate characteristics such as gender and English language proficiency. They are not dependant on the age of the examinees, the clinical case content, or the gender of the standardized patients.
Conclusions: For a multistation assessment, a reasonably precise and valid estimate of a candidate's communication ability can be obtained from trained standardized patients.