Brief article Reliability of college student self-reported drinking behavior Sara Walker, (M.A.), Merith Cosden, (Ph.D.) 4 Department of Counseling, Clinical and School Psychology, University of California Santa Barbara, Santa Barbara, CA 93106, USA Received 17 November 2006; received in revised form 21 January 2007; accepted 18 February 2007 Abstract College students represent a unique population among substance users, with developmental needs and motivational factors different from those of other adults with substance abuse problems. As most college-based treatment programs focus on harm reduction, the reliability of self-reported substance use is critical to assessing treatment needs and outcomes. This study examined the reliability of students’ self-reported alcohol consumption for the month prior to entering a treatment program at two points in time. It was hypothesized that: (a) students required to attend a drug and alcohol educational program would underreport their history of use when asked about it at intake (Time 1), relative to their self-report of use for the same historical timeframe after attending the program (Time 2); and (b) students would describe these discrepancies as a function of feeling more or less concerned about the impact of their disclosure. Findings supported both hypotheses. Implications for an evaluation of student drinking behaviors are discussed. D 2007 Elsevier Inc. All rights reserved. Keywords: Alcohol; Harm reduction; College; Reliability; Self-report 1. Introduction Studies of substance abuse treatment are limited by the reliability and validity of measures used to determine the amount of drug and alcohol use. Self-report measures of substance use have been critiqued in the literature and have been found to be a valid method for determining the level of drug and alcohol use, depending on the context of the assessment (Babor, Steinberg, Anton, & Del Boca, 2000; Del Boca & Noll, 2000). Of particular concern is the tendency for respondents with drug and alcohol problems to underreport their use when asked to publicly disclose it; this not only creates unreliable data but also hinders the purpose for which this information is typically collected—to provide treatment. In their review, Langen- bucher and Merrill (2001) identify a number of factors that appear to affect the underreporting of drug and alcohol use, including the clarity of the measures them- selves, respondents’ inability to accurately remember events, and concerns about the confidentiality and reper- cussions of responding. Although some errors in responding may be corrected by clarifying assessment instructions, in many instances, poor reliability in drug and alcohol self-report data is a result of intentional underreporting. Distortions tend to occur when respondents are concerned about the embar- rassment—or other negative consequences—of accurately portraying the extent of their use (Del Boca & Darkes, 2003; Del Boca & Noll, 2000). In a series of studies, Fendrich and Kim (2001) and Fendrich and Vaughn (1994) documented the substance use history of their program participants and then asked them about their historic use for a period of several years. They found that respondents underreported their historic use over time. The authors described this underreporting as a retest artifact, as a result of respondents becoming more concerned about the social desirability of their responses in the course of the longitudinal study as they perceived their responses as becoming less anonymous. In other studies, however (e.g., Anglin, Hser, & Chou, 1993; Stinchfield, 1997), respond- ents have admitted to greater use for the same timeframe 0740-5472/07/$ – see front matter D 2007 Elsevier Inc. All rights reserved. doi:10.1016/j.jsat.2007.02.001 4 Corresponding author. Department of Counseling, Clinical and School Psychology, University of California Santa Barbara, Santa Barbara, CA 93106, USA. Tel.: +1 805 893 2370; fax: +1 805 893 3375. E-mail address: cosden@education.ucsb.edu (M. Cosden). Journal of Substance Abuse Treatment 33 (2007) 405 – 409