Online education is becoming increasingly integral to postsecondary education. However, student performance in online courses do not parallel performance in in-person courses. This dissertation presents a series of studies investigating three undergraduate online courses at a large public four-year university.

The first study examines how online instruction affects different personality types, as measured by student responses to items on the Big Five Inventory. Using historical course grades for all students enrolled in two online public health courses and employing student fixed effects in a linear regression models, I estimate student performance in online courses relative to their own performance in in-person courses. I include interaction effects with personality to examine whether online courses differentially impact academic performance for different personality types. Findings suggest that highly conscientious students and students with high openness to experience are differentially impacted by online instruction.

The second and third study examine a scheduling intervention implemented in three different courses. Using a randomized control design, students in the courses were assigned either to a treatment group or control group. Treatment students were asked to schedule their coursework on a weekly basis for the full duration of each course. One of the two studies incentivized the act of studying as scheduled while the other simply incentivized the act of scheduling without regard to whether or not the student followed his or her schedule. Using course clickstream data, gradebook data, institutional data, and survey data, a series of linear regression models are built to estimate the intent-to-treat effect of the scheduling intervention on students’ study times and course grades. Mixed results are found across the two studies, though they suggest neutral to positive results for the scheduling intervention.

The last study examines students’ subsequent course performance and choice of major following the scheduling intervention in the second of the two studies. Using subsequent course grades and course performance averages for both the given and past terms, regression models are built to estimate the effect of the scheduling intervention. Findings suggest long-term positive impacts of the intervention on academic performance.

As educational institutions increasingly adopt digital tools for daily operations, unprecedented amounts of data are generated at different levels of the education system. The granularity of these big data makes it possible to understand and support educational processes in a data-informed, easy-to-scale manner, and educational data science (EDS) has emerged as a nascent field to realize this potential. This dissertation specifically focuses on the promise of EDS to address issues related to educational equity, a central theme of education research. To begin with, a two-dimensional taxonomy is presented to characterize equity-oriented EDS research -- whether the work is focused on explanation or prediction, and whether the problem of interest takes a micro- or macro-level perspective of education research. The interaction of these two dimensions partitions the research space into four quadrants, and one empirical study in higher education contexts is presented to illustrate each quadrant. The first two explanatory studies leverage novel data sources (i.e., digital behavioral traces) to understand systematic sociodemographic gaps in 1) peer interaction experience in virtual learning environments (micro-level), and 2) academic engagement over time at the institutional level (macro-level). The latter two prediction-oriented studies investigate algorithmic fairness from the perspectives of 1) choice of data sources in online learning performance prediction (micro-level), and 2) use of sensitive attributes in early warning systems (macro-level). These studies highlight how EDS research can advance the understanding of existing educational inequalities and guide preventive action against potential inequities. Finally, recommendations for future research on equity-oriented EDS are discussed.

This dissertation examines college-age Spanish learners' experiences participating in the Spanish language version of the massively multi-player online game World of Warcraft (WoW). A two-study design is used, first examining play from volunteer participants, and then incorporating WoW into a 1-credit Spanish course. The experiences of 5 key representative participants are reported here. Using data culled from participant observation, interviews, logs of in-game chat, student journal entries, and course activities, I describe how MMOGs like WoW create affordances for language learning. Specifically, address how the virtual environment (game features) of MMOGs like WoW mediate language learning, how social interaction with other players in WoW affects language learning, and how L2 Spanish proficiency and MMOG gaming proficiency shape the language learning experience.