On Learning Analytics & Assessment

Yesterday and the day before, the Educause Learning Initiative (ELI) was hosting a spring focus session on learning analytics.  I have to admit that drew me to this talk (in addition to being interested in analytics of course!) was the talk that George Siemens presented at this ELI to kick things off.  The first day was quite productive, but the second day I had too many other commitments to attend to and could not attend for most of it.

One of the back-and-forths that I had on twitter my colleagues was about learning analytics and assessment.  A fellow colleague seemed to be very certain that learning analytics could be used for assessment, and I disagreed.  It's hard to carry on a meaningful debate in 140 characters, so I thought I would write a quick blog post about it. And who knows, perhaps I mis-interpreted what my colleague was saying!

First I think it's good to start off with a few definitions so that we are all on the same page:

Learning Analytics: 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. (source)

Assessments: Assessments are activities that learners undertake to demonstrate their competence on a (or a set of) course or module learning objectives.  There are various types of assessments, but two most common ones are:

• Formative: formative assessments are done throughout the duration of the course
• Summative: summative assessments are done at the conclusion of the course (think: final exam)

The key thing here is that what is assessed is the learner's ability/knowledge after the course. You are NOT assessing the effectiveness of the course.

Assignments: Assignments are activities, that are ungraded, that are meant to help students learn what they need to learn and/or give students adequate opportunities to practice so that they can acquire and perfect those skills.  Assignments might be readings, ungraded tests, and discussion forums.


So, where there was a bone of contention (or at least it seemed like it) on our twitter interactions was around Assessment.  It seems to me that people wanted Learning Analytics to function as an assessment tool. Even though I talked about grading as being fundamental to Assessment, I got back a response that this type of assessment (graded) was Summative, but they were thinking along the lines of formative.

This confused me a little, because it seemed like they were mixing assignment with assessment, then taking this understanding of assessment = assignment, and applying learning analytics of those assignments to see how students were doing in the course. At this point we weren't talking about two terms, but three (and in 140 characters that's not easy)  Furthermore, a couple of days removed from the discussion, it seems to me that this person was probably wanting to assess the effectiveness of the course by using learning analytics, but that (if we go by the definition above) is not the goal of Learning Analytics.

Taking a step back from this, it seems to me that before we (as individuals, institutions, programs) commit to learning analytics, we need to figure out what we want to do. What sort of information do we need?  How can we get this information?  Is this information Learning Analytics (as defined above) or something else?

At the end of the day, I still feel confident that Learning Analytics ≠ Assessments ≠ Assignments