Understanding the accuracy of tests with cutting scores: The sensitivity, specificity, and predictive value model

Abstract

While researchers usually are concerned about psychometric properties of psychological tests estimated using large samples, most clinical decisionmakers must evaluate the accuracy of test results for individuals. This is particularly true as regards tests that have cutting scores to determine, for example, whether to assign a particular diagnosis or accept an applicant into a training program. This paper reviews a conceptual model that may foster improved understanding of test outcomes for individuals. The terms “sensitivity,” “specificity,” and “predictive value” are defined, and the relations of positive and negative predictive values to population base rates are emphasized. Examples from the psychological literature are presente to illustrate the utility of these concepts in clinical decision‐making with psychological tests. Implications for test users, test developers, and instructors are discussed.

The Psychometric Properties of the Hypersexual Behavior Inventory Using a Large-Scale Nonclinical Sample

“…Sensitivity was defined as the proportion of true positives belonging to the most problematic group based on the LPA, while specificity was defined as the proportion of the true negatives (Altman & Bland, 1994a;Glaros & Kline, 1988). PPV was defined as the proportion of the individuals with positive test results that was correctly diagnosed as hypersexual individuals, while NPV was defined as the proportion of participants with negative test results that were correctly diagnosed as nonhypersexual individuals (Altman & Bland, 1994b;Glaros & Kline, 1988). Moreover, taxometric analysis was conducted to investigate the latent structure of hypersexuality (Ruscio, Ruscio, & Carney, 2011).…”

The Development of the Problematic Pornography Consumption Scale (PPCS)

“…Considering the membership in this group as a "gold standard", the sensitivity, specificity, positive predictive value, negative predictive value, and accuracy values for all PPCS cut-off points were calculated. Sensitivity was defined as the proportion of true positives belonging to the most problematic group based on the LPA, while specificity was defined as the proportion of the true negatives (Altman & Bland, 1994a;Glaros & Kline, 1988). Positive predictive value was defined as the proportion of the individuals with positive test results that was correctly diagnosed as problematic users, while negative predictive value was defined as the proportion of participants with negative test results that were correctly diagnosed as non-problematic users (Glaros & Kline, 1988;Altman & Bland, 1994b).…”

The predictive validity of the Depression Hopelessness Suicide screening form for self-harm among prisoners

“…Glaros & Kline, 1988). However, as shown in sensitivity analyses (see online Tables S1 through S4), all four measures may be biased when the prevalence differs due to misclassification.…”