Background. The development and validation of the Social Phobia Diagnostic Questionnaire (SPDQ), a new self-report diagnostic instrument for social phobia is described in three separate studies.

Study 1. The participants were 125 undergraduates seeking help for an anxiety disorder of whom 60 had social phobia. Receiver operating characteristics (ROC) analysis was conducted comparing SPDQ diagnoses and clinician-based Anxiety Disorder Interview Schedule-IV (ADIS-IV) diagnoses of social phobia. Diagnoses made by the SPDQ showed an 85% specificity, an 82% sensitivity and kappa agreement with the ADIS-IV of 0·66.

Study 2. The participants were 462 undergraduates who completed the SPDQ and a battery of additional questionnaires. The SPDQ had good internal consistency (α=0·95), good split-half reliability (r=0·90) and strong convergent and discriminant validity.

Study 3. The participants were 145 undergraduates who completed the SPDQ at two time points separated by 2 weeks as well as several additional questionnaires. Scores on the SAD, FNE and SISST of SPDQ categorized undergraduates were also compared to scores on these measures from 35 clinical community participants to determine the clinical validity of the SPDQ. The SPDQ had strong 2-week test–retest reliability and good convergent and discriminant validity. Undergraduates diagnosed with social phobia by the SPDQ were not significantly different on the SAD, FNE and SISST from the socially phobic community sample, but both groups had significantly higher scores than undergraduates identified by the SPDQ as not meeting criteria for social phobia, demonstrating clinical validity of the SPDQ.

Conclusions. These three studies provide preliminary evidence of the strong psychometric properties of the SPDQ as a measure to identify socially phobic participants.