Five shades of flipping: using learning analytics in mixed models of learning

This is a featured post from ALT Sponsoring Member Talis.

Earlier this academic year I found myself teaching in ‘flipped classroom’ mode for the first time in my career.  Although my lectures are often lively and have included just about everything from Christmas crackers to car steering wheels, they are usually, beneath the glitz, variants of chalk and talk.  So for me this was a new adventure.

I came to this for a variety of reasons.

  • First I had many hours of video from an online human–computer interaction (HCI) ‘mini-MOOC’ I had delivered in early 2013 [1]. One of my aims when I prepared that course was to understand the production of re-usable video materials, so this was a chance to try first hand.
  • As well as being an academic at the University of Birmingham, I also work for the education technology company Talis, so this was an opportunity to pilot Talis’ new universal media player (Talis player), part of a project codenamed Lighthouse.
  • Finally, I have been writing about the use of learning analytics [2], but in my past teaching I only had access to at best crude click-thru, page visit or download statistics. Using extensive online resources with Talis player would allow me hands-on experience of richer analytics.

Most of the flipped classes were to a mixed group of undergraduate and MSc students, some of whom would have studied HCI before, but others not.  The use of online materials was particularly useful for this group.  Often you have to start at a very basic level in order to include those with less previous background, but in so doing risk boring other students.   However, online resources allow the delivery of both ‘catch up’ or remedial materials and also extension material allowing a  level of individual learning choice.

The basic pattern was similar for each of the classes: video and other materials online plus face-to-face lecture slot.  However, each class had a very different nature – I’m not sure if there are fifty, but there are certainly many shades of flipping!  In one class the video material was largely ‘basics’ that some students might have studied before, and the lecture ‘stand-up-and-talk’, but focused on integrative issues.  Others were fully flipped learning with information delivered in videos and the face-to-face session used for discussion on objects brought in by students or small group work.  Others sat somewhere between these extremes.

All the online materials were uploaded into Talis player.  This provided a uniform interface to view a variety of media (video, PDF, slides, audio), which are normally viewed in different players and platforms.  As well as providing students with a more consistent user experience, the single player is able to collect fine-grain usage data and provide correspondingly detailed learning analytics, not only that resources have been viewed, but precisely how much and how long, what Buckingham-Shum terms ‘micro-level’ analytics [3].

Individual student analytics
Individual student analytics

Not being used to this level of data I at first used it in a crude way, simply noting what proportion of students had looked at resources and gently chide the class when I could see some (typically about 1/3) hadn’t done enough preparation.   However, I gradually began to use the more detailed analytics.  A particular example was for a journal article I’d suggested the students read.  I could see that a substantial number had read the beginning, but then given up.  In class I was able to tell them how the nature of the material changed in the last section of the paper so worth reading this also.

In general the analytics gave me a greater sense of control when I couldn’t see the students eyes in front of me!

Detailed analytics for a partially read document
Detailed analytics for a partially read document

The slides below are taken from a presentation at an internal University of Birmingham eLeaning Practioners Forum, but I will be presenting this material again at Insight, Talis free HE teaching and learning conference in April.

[1]   Now available as OER at Interaction Design Foundation

[2] A. Dix and J.Leavesley (2015). Learning Analytics for the Academic: An Action Perspective. In Journal of Universal Computer Science (JUCS), Jan. 2015 (in press).

[3] S. Buckingham Shum (2012).  Learning Analytics. UNESCO Policy Brief.

Alan Dix, University of Birmingham and Talis

If you enjoyed reading this article we invite you to join the Association for Learning Technology (ALT) as an individual member, and to encourage your own organisation to join ALT as an organisational or sponsoring member

Leave a Reply

Your email address will not be published. Required fields are marked *