According to the Society for Learning Analytics Research (SOLAR), learning analytics is the measurement, collection, analysis and reporting of data about learners and their contexts, for purposes of understanding and optimizing learning and the environments in which it occurs. Although this definition was created in 2011, it still holds effective today as this process from measurement to reporting has become more digitized.
Just like social media analytics are gathered to examine and influence a user’s interaction with a platform like TikTok or Instagram, and like sports analytics are used to assess and improve a soccer player’s performance on the field, learning analytics are used to describe, diagnose, and prescribe changes to a student’s learning experience in a classroom, program, and/or university setting. And just like there’s a science to consumer behavior and a science to kicking a soccer ball, there’s a science to the way humans learn. Learning analytics is meant to leverage our knowledge of learning science, combined with what we can learn about what takes place within and across classrooms to give students the resources they need to be successful.
Today, student data points are gathered at all levels of education, from elementary school to masters and PhD programs. Of the hundreds of quantitative and qualitative data points that are gathered about a student learning inside the classroom, a few common examples include:
- Programmatic/enrollment data: credit completion rate, credit accumulation rate, course completion rate
- Individualized learning data: average course completion time, average assignment completion time, number of engagements (discussion posts, assignment submissions, reading annotations made) within a given period (week, month, semester)
- Qualitative learning data: Course pre-assessment (what skills do students already have coming into a course) questions asked by students about one topic in the course content compared to others, student written feedback about the course, written assignments (coded and analyzed), transcriptions from in-class discussions
- Cross – course data: identifying “bottleneck” courses (courses in the program that prevent a considerable number of students from continuing in that major/discipline), course withdrawal rates, assignment completion by course
- Disaggregated data: data that uncovers achievement gaps, professional development gaps, and inequitable experiences along the lines of race, ethnicity, gender, sexual orientation, class/year, and other personal identifiers
Notice that traditional assessment data points like GPAs and standardized test are not included in the list above. One of the goals of learning analytics is to look beyond standardized metrics like GPAs and test scores, as these often produce an inaccurate picture of how and what students learn, while also exacerbating equity gaps among marginalized students. The learning analytics equation should include students, faculty, instructional designers, data scientists, graphic artists, and computer programmers, all of whom play different roles in ensuring the success of the student by using data points like the ones above.
To learn more about learning analytics and its research, visit solaresearch.org.
