EFFECTS OF APPLYING DIFFERENT TIME LIMITS TO A PROPOSED GRE WRITING TEST

Abstract

To determine the role of time limits on both test performance and test validity, we asked approximately 300 prospective graduate students to each write two essays — one in a 40-minute time period and the other in 60 minutes. Analyses revealed that, on average, test performance was significantly better when examinees were given 60 minutes instead of 40. However, there was no interaction between test-taking style (fast vs. slow) and time limits. That is, examinees who described themselves as slow writers/test takers did not benefit any more (or any less) from generous time limits than did their quicker counterparts. In addition, there was no detectable effect of different time limits on the meaning of essay scores, as suggested by their relationship to several nontest indicators of writing ability.