Fluid Reasoning and Crystallized Knowledge as New Admissions Measures in German Higher Education

• In Event: Roundtable Session 10
  In Roundtable Session: Factors Predicting Success in College: An International Perspective

Abstract

Objectives and framework: Research in the field of college admission unequivocally demonstrates that achievement and ability tests are adequate predictors of subsequent graduate and undergraduate grade point average and other criteria reflecting learning success in college. Available evidence suggests that achievement and ability tests are incrementally valid over and above high school grade point average. Surprisingly, the well-established distinction between fluid and crystallized ability tests is not explicitly considered in well-known admission tests. German universities rarely use standardized ability and achievement tests for admission purposes. Based on these prior results and guided by German legislature we developed and evaluated a modified admission procedure for undergraduate degrees in psychology. A first objective of this project was to show that the general validity principles outlined above also apply for admission problems into psychology programs at German universities. A second objective was to investigate the joint validity of fluid and crystalized measures for the prediction of undergraduate degrees.

Methods: In a series of studies we piloted new measures with high school students and psychology students. In study 1 we tested newly admitted psychology students from several colleges and used their undergraduate degrees as criterion. In study 2 the newly developed admission procedure was operational for admission into a highly selective university. The newly developed battery included three reasoning measures as marker variables for fluid intelligence. Additionally, four measures from domains deemed highly relevant and critical for psychology programs by a college admission commission of the German Psychological Society and two requirements analyses with domain experts were developed: A biology knowledge test tailored to capture psychology relevant competencies, a mathematics competencies test with a focus on college entrance competencies in statistics and probability, an English as a foreign language reading comprehension measure, and finally a psychology comprehension measures that presented introductory level psychology textbook passages.

Data: In study 1 387 psychology freshman from five universities were included in a prospective study. All participants completed an early version of the new test battery. Undergraduate grades were retrieved two to three years later. In study 2 1,187 applicants completed a revised version of the test battery in an operational context. In this highly selective context criterion information for 80 participants were retrieved from archives 3 years after the testing.

Results: With both regression analysis and structural equation modelling studies 1 and 2 both show that as expected: a) ability measures are incrementally predictive over school grades in predicting exam grades in psychology; and b) that domain specific crystallized tests were incrementally predictive over fluid intelligence.

Significance: Both components of the new test battery – fluid and domain specific crystallized abilities substantially contribute to the prediction of undergraduate grade point average. Both components are incrementally valid over high school grade point average. Ceteris paribus test batteries reflecting the two distinguished components are arguably more valid that batteries focusing on either one of these aspects.

Author

• Oliver Wilhelm, University of Duisburg-Essen