Convergent validity in cognitive assessment of social anxiety: Endorsement versus production methods in deriving states of mind ratio

This study evaluated the convergent validity of 2 methods of assessing cognitive self-statements relative to social anxiety: an endorsement method (i.e., the Social Interaction Self-Statement Test, SISST; C. R. Glass, T. V. Merluzzi, J. L. Biever, & K. H. Larsen, 1982) and a production method (i.e., thought-listing; J. T. Cacioppo, C. R. Glass, & T. V. Merluzzi, 1979). Participants were randomly assigned to either method and exposed to a get-acquainted conversation after which they completed 1 of 2 self-statement measures. After converting self-statement scores to a states of mind (SOM) ratio, results showed the SISST yielded a significantly more adaptive SOM (i.e., .70) than the thought-listing SOM (i.e., .47). Also, the SISST was superior to thought-listing in accounting for predicted variance in 3 of 4 criterion measures. Reasons for lack of convergent validity and implications for future research are discussed.