Using data to understand student learning is a long-standing practice in educational research. Recent improvements in our ability to assemble and analyze large sets of data before, during, and after a course has been offered, however, have enhanced our ability to understand how various learning behaviors, instructional practices, demographics, academic history, and instructional materials shape student learning, faculty teaching effectiveness, and student retention and success.

Instructors typically use LA data and tools to

• gain a sense of the likelihood of success of students prior to the start of a course,
• identify students who may be in danger of failing or performing poorly in a course while the course is being offered,
• identify learning behaviors that are correlated with student success (or the lack thereof) in a course,
• identify course materials and assignments that are correlated with student success, and
• encourage students to engage in effective learning behaviors and engage with relevant instructional materials (e.g., through messaging).

Sources of LA data include:

• learning management systems, which provide information about logins, completion of assignments and homework, course materials accessed over time, performance on quizzes and exams, access to files, and use of discussion forums, among other data (Daniel, 2015; Zhang et al., 2018)
• learning tools provided by vendors and publishers, such as adaptive learning tools and interactive exercises (Lewkow et al., 2015) as well as learning platforms (including McGraw-Hill’s Connect and Macmillan’s Achieve) that provide information about student behaviors and performance, typically with the goal of identifying students who might benefit from intervention by the instructor
• eReaders (such as Unizin’s Engage platform), video players, and other tools for accessing and interacting with course content (Junco & Clem, 2015; Shoufan, 2018)
• “multimodal” data sources, which can reveal student location and other activities in real time, such as posting to social media and accessing wireless networks, by drawing on data from the Internet of Things, cloud data storage, and wearable technologies (Di Mitri et al., 2018)
• written texts produced in formal and informal assignments, including journaling and posts on discussion forums (McNely et al., 2012; Shum et al., 2016; Wise, Zhao, & Hausknect, 2013; Yu et al., 2017)

These data are often analyzed in combination with academic history, such as scores on college entrance examinations and performance in high schools—for example, high school GPA, CSU GPA, performance on related courses—as well as demographic information drawn from a student information system, such as race, ethnicity, gender, first-generation status, and financial aid information. In some cases, LA data from a specific course will be analyzed in combination with data about student participation in institutionally-supported activities, such as attending tutoring and study group sessions and meeting with faculty and academic advisors.