Abstract
Daniels and Hewitt (1978) reported “an extremely strong relationship” between test-anxiety scores and college course examination scores, whereas earlier investigators found smaller relationships between these variables (and one study reported no significant relationship when intelligence was partialed out). Daniels and Hewitt administed the test-anxiety measure immediately after a course examination, whereas the other investigators administered the test-anxiety measures at a different time than the course examinations. In the present study the Sarason Test Anxiety Scale was administered to 40 students immediately after a course examination and to 50 other students after a regular class period. A significant relationship between test-anxiety and examination scores was obtained only for the former group, and no significant relationship was found for either group when intelligence was partialed out. Thus, the high relationship obtained by Daniels and Hewitt may have resulted at least in part from their students' test-anxiety levels being altered due to their having just completed a course examination and from their response to demand characteristics inherent in the experimental procedure.