Concerning the Suitability of Recordkeeping for Validating and Generalizing about Reports of Health-Related Information

Abstract

Although the primary concern of validation studies is to assess the accuracy with which individuals report about behaviors or experiences, an important secondary concern of such studies is generalizability—the extent to which findings about the response validity of studied individuals are informative about the response validity of individuals who were not studied. Two recommendations are presented for assessing generalizability when participants contribute records to which subsequent reports are referred: (a) Reports should also be collected from a group of participants who do not keep such records and (b) control variables should be included in the design to evaluate the sensitivity of the studied measures to appropriate manipulations. These recommendations are described with examples of studies that have examined validity of reports about physical activity, menstrual distress, pain-related behaviors, and dietary intake.