Validity of self-reports in three populations of alcoholics.

This study examined whether population type (voluntary outpatient, voluntary inpatient, coerced outpatient) and question type (alcohol, nonalcohol, demographic) differentially affected the validity of alcoholics' self-reports. Three distinctly different populations of alcoholics independently completed life history questionnaires. The veridicality of subjects' answers was assessed using official records and documents. Generally, alcoholics in this study gave highly valid self-reports, a result that parallels the findings of earlier studies. Question type differentially affected the validity of subjects' interview answers, as significantly fewer invalid answers were given to demographic questions than to alcohol and nonalcohol questions. Population type, however, did not significantly affect the validity of self-reported life history information. Invalid interview answers were more often overreported than underreported when compared with official records.