Analytics for Learn: Using Data Science to Drive Innovation in Higher Education
Blackboard’s team of data scientists is driven by a desire to conduct research that answers important educational questions. Our goal is to create generalizable knowledge that can improve learning and teaching in higher education, both in the US and abroad.
As partners in education, we believe that we are engaged in a fundamentally ethical enterprise. How we teach, and the tools that we provide for education, can change lives. It is therefore with a sense of urgency that we work to better understand the nature of teaching and learning in online spaces, providing evidence in support of common assumptions while also calling others into question, so that everyone can benefit.
Research is important. At Blackboard, we are in a unique position to be able to conduct learning analytics research at tremendous scale; using cutting-edge techniques on very large data sets that span a broad variety of institutional types. As an educational technology company, we are also able to put our findings into practice through market-leading product innovation.
Today, we are thrilled to announce the addition of “course archetypes” to our flagship learning analytics product, Analytics for Learn. This announcement is the most recent example of how research and product development can come together to create something really special.
Course Archetypes & Instructional Design
In October 2016, Blackboard’s Dr. John Whitmer, Nicolas Nuñez, and Diego Forteze conducted an analysis looking at an anonymized sample that included data from 3,374,462 unique learners in 70,000 courses at 927 institutions. The research found that faculty design Blackboard Learn courses so that the time that students spend tends to follow one of several typical patterns, which they called “archetypes”:
- Supplemental – high in content but with very little student interaction
- Complementary – used primarily for one-way teacher-student communication
- Social – high peer-to peer interaction through discussion boards
- Evaluative – heavy use of assessments to facilitate content mastery
- Holistic – high LMS activity with a balanced use of assessments, content, and discussion
Beyond the categories, among the most fascinating findings of this research was the fact that, in aggregate, there was not a clear relationship between course archetype and student performance.
This is a really big deal for instructors and instructional designers.
It means that using the LMS as a static content repository is not bad, and that heavy use of tools like discussion boards and assessments is not good. What matters from a teaching and learning perspective is not how much the LMS is used (a pretty blunt metric), but rather how well online course environments are designed in accord with the instructional goals and specific needs of students in a particular university, school department, and course.
Building Course Archetypes Into Analytics for Learn
With today’s release of Analytics for Learn 4.3.4, we are now including course archetypes in our core data model. The ability to automatically classify courses by archetype opens up a universe of ad hoc reporting and learning analytics possibilities, and we have included a new set of visualizations to get our customers started.
By building our learning analytics and instructional design research into Analytics for Learn, we are excited to provide our customers with the information they need to more easily enter into conversations with faculty that goes beyond ‘hits and clicks.’ Now, universities, colleges, and departments can identify those course patterns that are seeing the greatest impact for different populations of students and scale high impact practices. Instructional designers and instructors can work together to fit course patterns to learning goals, engaging in evidence-based conversations about teaching methods in a way that, until now, was all but impossible.
View our data sheet on Course Archetypes in Analytics for Learn to learn more.