—Dr Endicott questions whether our report on the National Institute of Mental Health DIS in the April 1981 issue of the Archives was a study of reliability or validity. She seems to think that there is a clear answer and comes down squarely on the side of reliability. We think that there is no clear answer and that it is not always easy to conceptually differentiate reliability and validity; we addressed this difficulty in the "Comment" section of the article itself.
Dr Endicott contends that a "proper" test of lay interviewers' validity would be to compare their diagnoses with those of clinicians using criteria "as they generally are used." Of course, at the time this study was being conducted, one of the criteria sets (DSM-III) was in draft form and was not in "general" use at all. However, apart from that, it is not clear what an appropriate standard of