Dimensional comparison effects on (gendered) educational choices.

Expectancy-value theory (EVT) proposes that students’ appraisals of success expectancy and task value are the main drivers of their study and career choices. Dimensional comparison theory proposes that these beliefs are themselves affected by students comparing their ability across different domains. However, only a few studies have aimed to integrate these approaches and clarify the role of dimensional comparisons within EVT. Using longitudinal data, we aimed to fill this gap by studying within- and cross-domain effects of achievement (grades and test scores), academic self-concept (as a surrogate for expectancy beliefs), and values on German adolescents’ (N = 519) high school course choices and their intentions to major in a STEM subject at university. We show that (a) self-concepts predicted course choices, whereas values predicted STEM study intentions; (b) dimensional comparison patterns (positive within-domain and negative across-domain relations) were present; (c) gender differences in course choices were mediated by differences in achievement, self-concept, and value; and (d) there was an incremental gender effect on STEM study intentions above and beyond achievement, self-concept, value, and previous course choices. Furthermore, overall, a model incorporating cross-domain paths representing dimensional comparisons fit the data better than a model without these paths. We conclude that direct and indirect dimensional comparison effects contribute to predicting choices of high school courses and university majors and to understanding gender differences in these choices. We recommend that studies in the EVT framework include cross-domain effects. (PsycInfo Database Record (c) 2021 APA, all rights reserved)