Open AccessA STUDY OF THE DIFFERENTIAL IMPACT OF CURRICULUM ON APTITUDE TEST SCORES
Author(s) -
Angoff William H.,
Johnson Eugene G.
Publication year - 1988
Publication title -
ets research report series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.235
H-Index - 5
ISSN - 2330-8516
DOI - 10.1002/j.2330-8516.1988.tb00302.x
Subject(s) - psychology , aptitude , test (biology) , curriculum , sample (material) , developmental psychology , mathematics education , pedagogy , paleontology , chemistry , chromatography , biology
A sample of 22,923 students who had taken the GRE General Test in the academic years 1983‐84 and 1984–85 and who had also taken the SAT four or five years earlier were found, and classified by undergraduate field of study (four major categories of curriculum) and sex. Several analyses were undertaken to determine the degree of differential impact that sex and field of study might have on GRE‐verbal, GRE‐quantitative, and GRE‐analytical scores, after controlling on SAT‐verbal and SAT‐mathematical scores. It was found, first, that the correlations of SAT‐verbal with GRE‐verbal and SAT‐mathematical with GRE‐quantitative were extremely high, both for the entire sample, and within it, for the eight subgroups defined by field of study and sex. The correlations were .86 in the total sample and ranged from the low to middle .80s in the eight subgroups. The impact of curriculum and sex was found to be low on GRE‐verbal scores, but relatively high for GRE‐quantitative, with students in heavily quantitative fields enjoying an advantage over their peers in less quantitative fields of study. The impact was moderate for GRE‐analytical. Further studies designed to “purify” the fields of study and include only clearly verbal fields and clearly mathematical fields–omitting entirely students in social and biological science–showed small additional impact. An additional study indicated that there was a generally slight effect of the institution attended on GRE‐quantitative scores, after controlling for major field of study and initial ability, although the importance of institution attended was somewhat greater for higher ability students. Although these studies helped a bit to clarify the results, the basic conclusions remained unchanged. In a separate phase of the study an attempt was made by means of Mantel‐Haenszel analyses to identify the kinds of items that were relatively resistant to curricular and sex effects. Although the items differed from one another with respect to impact, they did not fall into identifiable categories that would make it possible to predict which items would be likely to show such impact and which would not.