A triangulated approach to the modelling of learning outcomes in first year economics

Abstract

Initial insights derived from modelling aspects of learning outcomes in the first year study of economics are presented. Of particular interest are three forms of entering students' prior knowledge: (a) subject-specific prior knowledge; (b) conceptions of learning, and (c) learning history. Part 1 compares the analyses of variance of inventory responses of the 1998 first year cohorts of the Universities of Adelaide and South Australia. Results across the two universities are consistent in confirming the significant effects of having studied economics at school, having English as a second language, and holding economic misconceptions, on (end of semester one) learning outcomes. Part 2 models learning outcomes using three approaches: (a) a single dimension of variation (selected from a more complex common factor model) which is then used to inform; (b) a k-means cluster analysis; and (c) a non-linear regression analysis. This "triangulation" of approaches produces different but complementary insights into the underlying relationships being modelled and improves our understanding of the learning engagement processes of entering first year economics students