The Job Demands-Resources Framework (JD-R) was originally developed to explain employee burnout and job performance in occupational settings. More recently, the framework has been effectively applied to academic environments to understand student well-being, including student mental health and academic performance. In addition, new developments in longitudinal analysis have identified weaknesses in cross-lagged panel models (CLPM), a traditional approach for analyzing longitudinal relationship. Instead, a new approach—random intercept cross-lagged panel models—has been proposed to address these concerns. This study explores the direct and feedback effects of a critical student resource—supportive teaching— on students’ exhaustion levels. By doing so, the study also illustrates the advantages of RI-CLPM, including how the approach can uncover hidden relationships. The study also explores the idea that resources not only affect exhaustion, but also that exhaustion can affect students’ perceptions of available resources.