Past research has consistently found that evening-types typically report poorer academic adjustment and higher levels of substance use compared to morning-types. An important development within the morningness-eveningness and psychosocial adjustment literature has been the hypothesis that social jetlag (i.e. the asynchrony between an individual's "biological" and "social" clocks) is one factor that may explain why evening-types are at a greater risk for negative psychosocial adjustment. Yet, only a handful of studies have assessed social jetlag. Furthermore, the few studies that have assessed social jetlag have done so only with concurrent data, and thus have not been able to determine the direction of effects among morningness-eveningness, social jetlag and psychosocial adjustment. To address this important gap in the literature, the present 3-year longitudinal study employed the use of a cross-lagged auto-regressive model to specifically examine the predictive role of perceived morningness-eveningness and social jetlag on two important indices of psychosocial adjustment among university students: academic adjustment and substance use. We also assessed whether there would be an indirect effect between perceived morningness-eveningness and psychosocial adjustment through social jetlag. Participants were 942 (71.5% female; M = 19 years, SD = 0.90) undergraduates at a mid-sized university in Southern Ontario, Canada, who completed a survey at three assessments, each one year apart, beginning in first-year university. Measures were demographics (age, gender and parental education), sleep problems, perceived morningness-eveningness, social jetlag, academic adjustment and substance use. As hypothesized, results of path analyses indicated that a greater perceived eveningness preference significantly predicted higher social jetlag, poorer academic adjustment and higher substance use over time. In contrast, we found no support for social jetlag as a predictor of academic adjustment and substance use, indicating that social jetlag did not explain the link between perceived morningness-eveningness and negative psychosocial adjustment. An important finding was the significant predictive effect of higher substance use on social jetlag over time. Results of the present study highlight the importance of employing a longitudinal framework within which to specifically determine the direction of effects among the study variables in order to validate proposed theoretical models that aim to guide our understanding of how perceived morningness-eveningness, social jetlag, academic adjustment and substance use relate to each other.