The major objective of the proposed research is to determine the extent to which client posttreatment self-reports can provide data that are useful in evaluating treatment outcomes and how the reliability and validity of self-reports can be improved. The proposed research uses a controlled field experiment designed to 1. assess the reliability of posttreatment self-reports using internal consistency assessments, concurrent self-administered reports, and reinterviews in a test-retest design. 2. assess the validity of posttreatment self-reports using multiple criteria including biochemical tests, collateral reports and record data, 3. examine the effects of the reliability and validity of posttreatment self-reports of the . method of interviewing (phone versus face-to-face versus self-administered questionnaires) . interviewer characteristics (professional interviewers versus experienced alcohol treatment counselors), . respondent characteristics (sociodemographic characteristics, treatment history, denial, cognitive impairment), and . interactions of these factors (method by interviewer, method by respondent, interviewer by respondent), 4. evaluate the quality and utility of client self-reports gathered under different methodologies by comparing analyses using data collected with face-to-face interviews, phone interviews, and self-administered questionnaires. The results of the study will help to develop methods to obtain more reliable and valid self-report data. Measurement error models will be used to better assess and compare the utility of results of alcohol treatment effectiveness studies based on different self-report methodologies.
