Two Shortened Versions of the Cornell Medical Index—A New Test of Their Validity

Abstract
Two shortened versions of the Cornell Medical Index (CMI) were tested for concurrent validity using two reference groups. Sensitivity, specificity and relative liability were computed and found to be of relatively low value. The usefulness of the questionnaires for screening the emotionally disturbed was found to be affected by respondents soclodemographic characteristics such as age and sex and, in particular, educational level and ethnicity.