Psychometric properties of the state Probability and Consequences Questionnaire for social anxiety disorder

Abstract

Cognitive models of social anxiety propose that overestimation of the probability and cost of negative evaluation plays a central role in maintaining the disorder. However, there are currently no self-report state-based measures of probability and cost appraisals. The current paper examines the psychometric properties of the Probability and Consequences Questionnaire for social anxiety (PCQ-SA), which measures probability and consequence appraisals both in anticipation of, and in response to, an impromptu speech task. A total of 532 participants were recruited for the present study, consisting of 409 participants with a principal diagnosis of Social Anxiety Disorder (SAD), and 123 non-clinical controls. Results of exploratory and confirmatory factor analyses supported a two-factor solution for the PCQ-SA. The PCQ-SA demonstrated excellent internal consistency, excellent test-retest reliability, good convergent validity at both time points (i.e., pre and post speech task), and sensitivity to treatment. Finally, using Receiver Operating Characteristic Curve Analysis, clinical cut-off scores were calculated for probability and consequences at both time points, with the PCQ-SA scales showing good sensitivity, specificity, and positive and negative predictive values. Overall, the results provide evidence that the PCQ-SA possesses excellent psychometric properties. The PCQ-SA is suitable for use in clinical and research settings to assess key cognitive maintaining factors for SAD.

Keywords: Factor analysis; Probability and consequences questionnaire; Psychometric properties; Self-report; Social anxiety.