Performance realism in test-anxious students

Abstract Twenty-four high test-anxious and 24 low test-anxious students completed two cognitive tasks under either high or low performance demand manipulations. Before each task subjects rated anxiety levels and performance expectations, with performance evaluations being conducted following completion of each task. Results revealed lower actual, anticipated and self-evaluated performance scores for high test-anxious students compared to their low test-anxious counterparts. Expected and self-evaluated performance ratings were then compared with actual performance. No difference was found between high and low test-anxious subjects in terms of the discrepancy between expected and actual performance. Both groups of students tended to be overoptimistic in their performance predictions. However, following the first cognitive task the test-anxious group demonstrated accurate self-evaluation, whereas the low test-anxious group were overoptimistic in their self-evaluation. This effect was not found for the second task, on which both groups were accurate in their performance-evaluation. The results are discussed in terms of cognitive theories of anxiety, self-focussed attention and realism amongst test-anxious students.