SELF-COMPETENCE AND MOTIVATION IN MIDDLE SCHOOL MATHEMATICS: PROFILES, MECHANISMS, AND INTERVENTIONS

Abstract Mathematical competence is a crucial skill for individuals’ academic, professional, and personal development, as it is recognized as a strong predictor of academic attainment, later occupational opportunities, and participation in STEM pathways. Despite its importance, international assessments report a decline in students’ math competence and in STEM (science, technology, engineering and mathematics) career engagement especially during middle school. A growing body of evidence shows that emotional and motivational factors, such as self-competence and value beliefs, theory of intelligence, and math anxiety, play an important role in shaping students’ engagement and achievement in math. While prior work has often examined motivational and emotional constructs in isolation, their interrelations and dynamic interplay remain largely underexplored, especially during middle school, as well as how these motivational constructs can be effectively supported and enhanced through targeted educational interventions. Therefore, this dissertation aimed to advance an integrated understanding of the emotional-motivational mechanisms underlying math learning and STEM vocational interest, drawing on Control-Value Theory of Achievement Emotions (CVT)and a Self-Regulated Learning (SRL) perspective. Specifically, the studies in this dissertation sought to: 1) identify motivational-emotional profiles defined by math control (e.g., math self-efficacy), intrinsic and extrinsic math value, and math anxiety, and test whether these profiles predict STEM vocational interest one year later; 2) examine whether theory of intelligence is associated with math anxiety indirectly through math self-efficacy, and explore gender differences in the model; 3) synthesize intervention evidence targeting math self-competence beliefs (self-efficacy and self-concept) in middle school; and 4) evaluate the effectiveness of an SRL intervention designed to enhance multiple motivational factors and math performance, using post-test and six-month follow-up assessments. Results showed that motivational-emotional profile membership predicted later STEM vocational interest beyond prior achievement and prior STEM interest, with the most adaptive profile characterized by high control and value and low anxiety. An incremental theory of intelligence was indirectly associated with lower math anxiety via higher self-efficacy; although no gender moderation emerged, girls reported higher anxiety levels. The review on interventions documented heterogeneous approaches, with collaborative and technology-based programs showing promising effects on self-competence beliefs. Finally, the SRL intervention yielded gains in math perseverance, math utility value, theory of intelligence, and math performance, with math perseverance, theory of intelligence, and math performance remaining higher at follow-up. In sum, this dissertation supports a theory-driven approach to math education that jointly addresses control, value, emotions, and self-regulation to promote achievement and sustained STEM engagement.