Self-Efficacy, Intrinsic Motivation, and Self-Regulated Learning as Predictors of Thesis Quality and Process Efficiency Among Undergraduate Students: A PLS-SEM Study

The general objective of this study is to test an integrative PLS-SEM model that simultaneously explains thesis quality and process efficiency among undergraduate students through the affective-motivational mediators of research self-efficacy and project management. Students who stall in the final stage of their degree rarely do so because they lack technical skill. More often, confidence erodes under sustained uncertainty, motivation shifts from intrinsic engagement to anxious compliance, and the demands of organizing months of research exceed what willpower alone can sustain. This study examines those emotional and motivational dynamics directly, treating research self-efficacy and intrinsic motivation not as background variables but as the affective-motivational core of thesis performance. Using partial least squares structural equation modeling (PLS-SEM) grounded in self-determination theory and social cognitive theory, we tested an integrative model with data from 396 undergraduate students actively completing theses at public and private universities in the northern region of Peru. Four enabling factors—methodological competencies, intrinsic motivation, tutorial support, resources and conditions—were linked to thesis quality and process efficiency through two mediating mechanisms: research self-efficacy (the confidence to face methodological difficulty without retreating) and project management (the behavioral self-regulation that converts motivation into organized work). Resources and conditions showed the strongest associations in the model, with the largest effects on both project management (β = 0.533) and research self-efficacy (β = 0.418). Self-efficacy, in turn, showed the strongest association with thesis quality (β = 0.518), while project management and quality were jointly associated with process efficiency. The model explained 70.5% of variance in thesis quality and 81.4% in process efficiency. These patterns point to a concrete institutional lever: securing the material and temporal conditions that allow students to do the work, rather than attributing delays solely to failures of individual motivation. Because the design is cross-sectional and based on self-report, these relationships are interpreted as theory-consistent associations rather than causal effects.