Putting Learning Analytics Data in the Hands of Students


Data and analytics are at the center of today’s student success movement. Increased interest and access to institutional data are helping colleges and universities identify systematic and structural barriers to retention and graduation. Predictive analytics is making proactive advising possible at scale.

Largely absent from mainstream conversations about educational data and student success is a consideration of the role of the student as a learner. The perspective of the institution is incredibly important, but how can we bridge the gap between this macro view concerned with retention and graduation, and the micro perspective concerned with teaching and learning?

Student-facing dashboards are among the most promising responses to the micro-macro integration problem facing the use of data in support of student success today. The path to graduation begins with successful course completion. If students do not complete the courses they need to graduate, they can’t progress. Grades of D, F, and W mean wasted credits, increased time to degree, increased educational costs, and decreased chances of graduation. But our interest as educators goes beyond graduation.

Recent research from Blackboard’s data science team and our partners is showing strong evidence that providing learners with relevant analytics can increase their performance by fostering self-regulated learning, particularly among otherwise low-performing students. Watch John Whitmer present some of this recent research in this video:

Behaviors of High Performing Students

Using a sample of 70,000 courses from 927 institutions, with 3,374,462 learners, researchers at Blackboard found that the most consistent predictor of student achievement was how frequently a student looked at their grades. Of course, this does not mean that grade-checking causes high performance. All this tells us is that grade-checking is a common behavior among high-performing students. But the result still has a couple of important implications.

RECOMMENDED READING >> How successful students use LMS tools – confirming our hunches

First, this research underscores the importance of feedback. Students want feedback on their activity and performance. In most classes – whether traditional, online, or blended – the most common and commonly accessible form of feedback is grades, so that’s what students check. What if we were to provide students with feedback in a way that was both more regular, and more actionable?

Second, the findings of our data science team resonate with our experience of performance feedback in general. We are naturally more inclined to seek feedback when we are high-performing or seeing improvement than if we are underperforming or in periods of decline. I am far more likely to step on the scale when I am feeling skinny than I am in the week(s) immediately following Thanksgiving. This insight suggests a promising opportunity. We know that performance feedback has a positive impact on behavior, and that it is less likely to be actively sought by those who could use it the most. Imagine what would happen if we surfaced relevant learning analytics to students at times when they are most likely to benefit and least likely to seek it out themselves.

Self-Regulated Learning

Research conducted by John Fritz at University of Maryland Baltimore County has shown that giving students information about their Blackboard Learn activity compared to others significantly increases their course performance. In 2008, UMBC launched a feedback tool called Check My Activity (CMA), which allows students to compare their level of activity in Blackboard Learn with an anonymous summary of the activity of others. In Spring 2012, students using Check My Activity were nearly twice as likely to earn a grade of C or higher than students who did not. On the basis of this promising research, we built student-facing activity reports into our flagship learning analytics solution, Analytics for Learn and include a student-facing dashboard as an important feature of our newly launched Blackboard Predict, as well.

RECOMMENDED READING >> Using Analytics at UMBC: Encouraging student responsibility & identifying effective course designs

The potential of student-facing activity dashboards to decrease the rates at which students earn grades of D, F, or W is impressive. But what is more interesting, perhaps, is the reason why they are so effective. By putting data in the hands of students, universities actively foster self-regulated learning. A concern that some have about the use of data in higher education is that the interventions it supports might prevent students from taking responsibility for their education. By giving student data to the students themselves, and encouraging active reflection on the relationship between behavior and outcomes, colleges and universities can encourage students to take active responsibility for their education in a way that not only affects their chances of academic success, but also cultivates the kind of mindset that will increase their chances of success in life and career after graduation.

Reaching the Right Students at the Right Time

How do we know that student-facing dashboards are useful for students at risk? How do we know that these kinds of feedback tools aren’t just serving as another resource to those who would be successful anyway? To the extent that low-performing students are affected by access to activity data, how do we know that they are affected positively?

RECOMMENDED READING >> Surprising lessons from research on student feedback about data dashboards

To answer these questions, Stephanie Teasely from the University of Michigan and Blackboard’s John Whitmer collaborated on a qualitative research project to understand the differential impact of student-facing dashboards on high and low-performing students. What they found was really exciting.

As a result of user testing analytics displays in Blackboard Learn, complemented by automated performance feedback in the form of notifications, Teasely and Whitmer found that low GPA students were more likely than higher performers to take immediate action on the basis of alerts. They also found that the same students were significantly more likely to check dashboards and turn on notifications. All students found automated nudges about low performance (grades and/or activity) helpful, regardless of GPA.

The results of this research support our excitement about the potential impact of student-facing analytics to decrease D, F, and W grades, increase progression, and affect graduation rates at scale. The dashboards and notifications that we tested in collaboration with the University of Michigan are now generally available as a standard feature of Blackboard Learn with the Ultra experience.