Demographic variables in medical school admission.

Admission committees report that demographic variables, although accurate, reliable, and easily obtained from applicants to medical schools, are only moderately important in their decision making. This may be because the committees are concerned about the validity and legality of using such data as admissions criteria. This essay discusses the research on the validity of demographic variables and the recommendations for their legal use in selecting students for medical school. The relationships of age, gender, size of hometown, parental education, parental occupation, parental income, and marital status to medical school outcomes of preclinical performances, clinical performance, attrition, specialty choice, and practice location are summarized. [Race or ethnic group is discussed in a separate essay in this issue.] Although the authors focus on the predictive value of demographic variables, these variables play a more important role as the moderator variables for other predictors of medical school outcomes. The full value of using demographic variables derives from data obtained in local validity studies. To comply with the equal-protection and due-process requirements, admission officers must assure that the use of demographic information in the admission process is explicit in bulletins given to applicants and is uniformly applied in the evaluation of all applicants.