Recent studies have evaluated the use of social media as learning aids in tertiary education. Emerging research in this area has focused primarily on non-quantitative approaches to student social media engagement. However, quantitative engagement outcomes may be extracted from student posts, comments, likes, and views. The goal of the present review was to provide a research-informed taxonomy of quantitative and behavior-based metrics of student social media engagement. We selected 75 empirical studies comprising a pooled sample of 11,605 tertiary education students. Included studies used social media for educational purposes and reported student social media engagement outcomes (source databases: PsycInfo and ERIC). We used independent raters and stringent interrater agreement and data extraction processes to mitigate bias during the screening of references. Over half of the studies (52%, n = 39) utilized ad hoc interviews and surveys to estimate student social media engagement, whereas thirty-three studies (44%) used some form of quantitative analysis of engagement. Based on this literature, we present a selection of count-based, time-based, and text-analysis metrics. The proposed taxonomy of engagement metrics resulting provides the methodological basis for the analysis of social media behavior in educational settings, particularly, for human operant and behavioral education studies. Implications for future research are discussed.