Keeping analytics do-able

By Mark Nichols, The Open University (UK)

When it comes to successful interventions and solutions in education, it seems they’re either incredibly simple or diabolically expensive. Of course, sometimes the incredibly simple only takes you so far before you really need something diabolically expensive. I suspect, though, that in online learning we sometimes go straight to the diabolically expensive because the incredibly simple seems less innovative!  

Analytics is likely a case in point. I think there are some incredibly simple analytics systems we could propose as online learning professionals that require little or no new financial investment whatever. I'm certainly advocating that we push decision-makers harder for ambitious analytics systems and innovations. Alongside this, though, most of us likely already have all of the data and support we need for making significant improvements to student learning through 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" (Wikipedia). This is a very useful and practical definition. Problem is, sometimes we can assume it requires significant IT infrastructure, complex data-mining, and detailed reporting. Actually, effective use of analytics can be quite simple and already within anyone's grasp. Four questions will get any analytics system underway:

  1. What is your specific objective?
  2. When do you want to intervene?
  3. What data do you already collect?
  4. How effective are your institution's existing support services?

It is too easy to over-spend and over-complicate analytics systems. A specific objective lends itself to accessible solutions. Start with the data you already have, and the support functions you already have; ensure support functions have usable access to data, and systematise data use.

By way of practical example, let’s say you want to improve student retention during a period of study. You have access to LMS logins, even though the course under study may not require online access; you have access to assessment performance; you have access to student records. The data you can already access, though at a rough level of detail, will be enough for you to know which students would benefit from a personal email or telephone call. Ormond Simpson, previously of the Open University, demonstrated how even a simple phone call could make a great deal of difference to a student. He had ways of identifying which students would benefit, and his 2003 book is still very useful in this area.

So, we don’t necessarily need access to highly detailed information or complex reporting systems to make a real difference to student outcomes using analytics. Incredibly simple.

One Comment

  1. Having all the data at your fingertips also does not ensure that you have an institution with a will to do anything with it. Data and its analysis is useless if it sits isolated in a data sink somewhere. I totally agree with you Mark, that we can do a lot for a very little spend if we're smart and know what we're trying to acheive, but I sometimes wonder whether we are collecting all the stuff without really knowing what we intend to do with it.

Comments are closed.