Brief statement on learning analytics

I finally finished my brief statement on learning analytics for the panel I’m participating on at LAK12. Since I had to write it, I’m publishing it here. Perhaps it will put you to sleep; perhaps it will inspire you; perhaps you’ll experience an unexplainable pining for the fjords. Either way, here it is.

As part of his 2 sigma work, Bloom (1984) challenged educational researchers to devise practical methods – “methods that the average teacher or school faculty can learn in a brief period of time and use with little more cost or time than conventional instruction” – that would help learners reach their academic potential. My personal interest in learning analytics lies in its potential to answer extremely practical and socially responsive questions such as, “What is the most effective thing a teacher could do with her next 30 minutes?” and “What is the most effective experience a learner could choose next?” In my view, learning analytics as a term simply describes the extremely interdisciplinary endeavor of providing this pragmatic support for learning.

On the “back end” learning analytics combines knowledge and techniques from data mining and psychometrics to leverage both behavioral data and data about academic performance. From this perspective learning analytics is a synthesis of techniques like Naïve Bayes, Rasch modeling, collaborative filtering, and item response theory. Both data mining and psychometrics possess a rich set of tools that are applicable to the problems we want to solve using learning analytics.

On the “front end” learning analytics combines knowledge and techniques from data visualization and UI/UX to empower ordinary teachers or learners with little or no training to bring the full power of data to bear on their learning-related decisions. Data-related tools still look too much like the “Your Product” in the famous StuffThatHappens comic. We typically fail to acknowledge that the work involved in achieving Google or Apple-like simplicity in the front end design of learning analytics tools will require at least as much effort and attention as will solving “back end” problems.

Learning analytics, then, is a consumer of the knowledge created by the educational data mining community and depends on this and the work of numerous other fields in order to bring the full promise of technology (in this case, the data-enabled promises) to ordinary learners and teachers everywhere.

Bio: Dr. David Wiley is Associate Professor of Instructional Psychology and Technology and Associate Director of the Center for the Improvement of Teacher Education and Schooling at Brigham Young University, where he directs the Open Education Group. David is currently Senior Fellow for Open Education at the National Center for Research in Advanced Information and Digital Technologies (Digital Promise) and a Peery Social Entrepreneurship Research Fellow in BYU’s Marriott School of Business. Previously, David was a recipient of the National Science Foundation’s CAREER grant.

1 thought on “Brief statement on learning analytics”

  1. Since 1993 when I was a legislator working on our ed reform bill in MA, I have been semi-obsessed with the application of data toward improving student learning and outcomes.  With the new energy behind education technology and investing, I am hoping that well-thought out approaches such as you have described here will finally become a reality.  It has taken much too long.

Comments are closed.