Background <p>Technology-enhanced classrooms now integrate a range of educational apps designed to improve student outcomes. The effectiveness of these applications is influenced by multiple factors related to the courses and the applications themselves. A critical factor is student engagement, which involves interacting with the course content (academic engagement) and the educational applications (app engagement) to achieve potential academic benefits. This study examined the following research questions to explore this relationship: (1) To what degree are students’ academic and application engagement associated with their performance across different courses? (2) How does the correlation between students’ engagement (academic and application) and academic performance vary in each course? (3) What unique contributions do academic engagement and app engagement make to student performance?</p> Results <p>This study employs a multimethod approach to examine the relationship between academic engagement and app engagement with students’ academic performance among 252 students: 110 in a first-year engineering (FYE) course and 142 in a physics (PHYS) course. In these courses, students used the CourseMIRROR mobile application, where they created and submitted a reflection after each lecture throughout the semester, and reflections were summarized using natural language processing (NLP) algorithms. This study collected app engagement as the number of times students viewed those summaries. For academic engagement, the data were collected using a validated self-reported survey comprising 22 items. For academic performance, we used the final course score. We analyzed the data using Pearson product-moment correlation, multiple regression, and stepwise hierarchical regression methods. The findings showed app engagement was significantly correlated with academic performance in FYE (<i>r</i> = 0.235, <i>p</i> &lt; 0.01), while in PHYS, academic performance was strongly associated with three out of four constructs of academic engagement (<i>r</i> = 0.267 to 0.289<i>, p</i> &lt; 0.01). Notably, app engagement explained most of the variance in the engineering course, while social-academic engagement was significantly related to students’ academic performance in the physics course. In both courses, app and academic behavioral engagement positively and substantially contributed to academic performance.</p> Conclusion <p>This study provides insightful results on the multifaceted and multidimensional engagement construct using two conceptualizations and the contexts of two courses, i.e., the FYE and PHYS courses, while accounting for gender and race/ethnicity. It demonstrates that the effectiveness of engagement is context-dependent, influenced by its intended purpose and implementation method.</p>