Further Research Relative to Trice's Study of Juniors' and Seniors' Self-Reports of Scholastic Aptitude Scores

Abstract

Trice (1990), extending the work of Goldman, Flake, and Matheson (1990), by studying the Scholastic Aptitude Test (SAT) self-reports of juniors and seniors at a women's liberal arts college reported that the total SAT mean for students not reporting their scores was significantly lower (M = 802, SD = 164) than those who reported scores (M = 881, SD = 198; t = 2.04, p < .05). Using their earlier data from university freshmen and sophomores (Goldman, et a/ . , 19901, the present researchers found results similar to Trice's in that the 83 university students who did not report their SATs had a total mean significantly lower (M = 876, SD = 119) than those 200 who reported scores (M = 977, SD = 151; t = 6.02, p< ,001). Also, the present researchers compared high school and university grade point average means for those who reported and those who did not report them. Although the differences were small and not significant, agxn, those who did not report GPAs had lower means than those who reported them (high school GPAs not reported: n = 88, M = 2.80, SD = .46 vs high school GPAs reported: n = 183, M = 2.81, SD = .51; and university GPAs not reported: n = 173, M=2.37, SD = .79 vs university GPAs reported: n = 126, M = 2.49, SD = ,701. Trice (1990) further demonstrated the difficulty of relying on self-report data by comparing the correlation of self-report SAT-Verbal scores and actual college GPAs (r = .37) with the correlation of actual SAT-Verbal scores and actual GPAs ( r = .44). The present researchers found similar results for their analysis although their correlations were smaller, i.e., self-report SATVerbal scores and actual university GPA correlated .20 while actual SAT-Verbal scores and actual university GPA correlated .35. In addition, they made this comparison for SAT-Mathematics scores and again found similar results, i.e., self-report SAT-Mathematics scores and actual university GPA correlated .27 while actual SAT-Mathematics and actual university GPA correlated .44. These findings further point out the distortions in college students' self-reported data as they tend to overestimate their actual achievement and in fact lower achieving students may not even report their academic accomplishments. Gaps still remain in this research area because Trice, being at a woman's college, was unable to study junior and senior men. Also, with respect to the present researchers' work it would be interesting to see a breakdown of the data by class (freshmen and sophomores) and by sex.