Diverse screening and assessment techniques are used to make decisions about the interventions and sanctions that a detected impaired driver will receive. It is possible to discern several distinct underlying hypotheses that are the basis for most screening/assessment procedures, and these hypotheses have distinctly different implications for choice of validation strategies. Two primary hypotheses are risk assessment and treatment matching. Risk assessment hypotheses are shown to posit differences in base rates of risk, whereas treatment matching hypotheses are shown to posit specific interactions that must be tested to establish the validity of the screening/assessment procedure. Most underlying screening rationales can be shown to be versions of one of these hypotheses. Failure to identify the underlying rationale and hypotheses for use of screening/assessment can lead to inappropriate choice of validation procedures. Each hypothesis is defined, and the necessary conditions for establishing validity under each hypothesis and common combinations of these hypotheses are developed.
Abstract