Open AccessTHE CONTRIBUTION OF MEASURES OF APTITUDE (SAT) AND ACHIEVEMENT (CEEB ACHIEVEMENT AVERAGE), RESPECTIVELY, IN FORECASTING COLLEGE GRADES IN SEVERAL LIBERAL ARTS COLLEGES
Author(s) -
Wilson Kenneth M.
Publication year - 1974
Publication title -
ets research bulletin series
Language(s) - English
Resource type - Journals
eISSN - 2333-8504
pISSN - 0424-6144
DOI - 10.1002/j.2333-8504.1974.tb00664.x
Subject(s) - aptitude , academic achievement , test (biology) , achievement test , liberal arts education , predictive validity , psychology , scrutiny , mathematics education , standardized test , developmental psychology , higher education , paleontology , political science , law , biology
The validity of the CEEB Scholastic Aptitude Test‐Verbal and SAT‐Mathematical scores, separately and in combination with the high school record (rank in class or high school average grade) for predicting college grade averages, has been widely studied and reported, but the value of CEEB Achievement Tests for such prediction has not been subjected to comparable scrutiny. Consequently it is not widely recognized that a composite (sum or average) of scores, based on whatever combination of Achievement Tests a candidate takes, can be useful for assessing the general academic‐performance potential of candidates. This paper presents evidence bearing on the predictive validity of the CEEB Achievement Average—defined as the arithmetic mean of all Achievement Test scores included on a candidate's CEEB Admissions Testing Program score‐report and now routinely reported to colleges—for women students in several liberal arts colleges. The findings reviewed are consistent with those reported earlier by Schrader that the CEEB Achievement Tests contributed modestly to improvement of predictive effectiveness over that provided by the SATs and the school record. In the studies reviewed, there was a tendency for the SAT scores (particularly the SAT‐M score) to operate, in effect, as suppressor variables when included in a battery with the CEEB Achievement Average. It would appear not only that the CEEB Achievement Average tends to be a more valid predictor of college grades than the SAT in the settings studied, but also that this composite contains all the SAT‐type information that is useful for such prediction—i.e., that the SAT scores become redundant after taking into account information provided by the CEEB Achievement Average, with its substantial SAT‐type components. Questions are raised regarding the extent to which candidate freedom of choice in test selection is a factor which contributes to the predictive effectiveness of the CEEB Achievement Average.