Classification of Motivational Dispositions: A Psychological Systems Perspective of Academic Behavior

Academic motivation is a complex psychological construct. It is multifaceted, hierarchical, dynamic, and developmental in nature. Students exhibit substantial heterogeneity in the ways in which academic motivation develops and manifests in short and long-term behavior. Much of the prior literature has considered motivated behavior in an academic subject like math to be the result of math motivation alone. This dissertation draws from the Eccles’ expectancy-value theory of achievement motivation, which posits that academic behavior is the result of subjective psychological perceptions of competence and task-value, in order to investigate the hierarchical and dynamic nature of motivational dispositions and their associations with academic behavior. In Chapter 2, I use variable and person-centered approaches to investigate the development of math and English self-concept of ability throughout adolescence and their dual role in the process of selecting a college major. I demonstrate that the relationship between math and English self-concept changes over time, from being positively correlated to uncorrelated to negatively correlated between sixth and twelfth grade. Cluster analyses uncover heterogeneity in the patterns of math and English self-concept that students hold, and these clusters are predictive of the college major that students eventually select. The cluster also reveal gender differences in self-concept hierarchies that ultimately relate to college major choice. In Chapter 3, I use longitudinal structural equation modeling to study the development of academic task-values in high school. I focus on the role of dimensional comparisons, which refer to cross-domain influences, across the domains of math, English, biology, and physical science. Results indicate that achievement relates to cross domain subjective task values (STV), and STV in 10th grade relates to cross domain STV in 12th grade. STVs in 12th grade relate to college major choice. In Chapter 4, I use cluster analysis and hierarchical logistic regression to investigate daily academic behavior in an undergraduate online course. I synthesize theories of motivated behavior and demonstrate that unique profiles of subjective task values and emotions relate to both expectations of task attainment and actual task attainment. In Chapter 5, I conclude with some critiques regarding the scientific study of motivation, such as problems with modeling latent constructs using linear models. I then present a dynamic systems approach from the physics to modeling human motivation