Predictors of alcohol abusers' inconsistent self-reports of their drinking and life events.

Although considerable research supports the veridicality of alcohol abusers' self-reports, all studies find that some proportion of self-reports are inaccurate. Recently, a few studies have examined variables predictive of inaccurate self-reports and found considerable intersubject variability. The present study examined predictors of alcohol abusers' inconsistent reports of life events and drinking using test-retest reliability data from two questionnaires. Results indicated that inconsistent self-reports were associated with the type (i.e., objective versus subjective) and amount (i.e., more drinking involvement at the first interview was associated with greater discrepant reports at the second interview) of information to be recalled. It appears that the nature of the questions asked may be as much or more of a contributing factor to inaccurate self-reports as subject or setting factors, especially for individuals who report high levels of alcohol use, for whom special efforts may be necessary to gather valid self-report data.