Thursday, January 20, 2011

The Marked Benefits of Learning Analytics

I want to bring together three things I've read or heard so far in the #LAK11 course to make a point about letting students see the data we hold on them. In a miracle of joined-up thinking this post picks up the point I was making in the last paragraph of my previous post in which an Amazon in a parallel universe couldn't see the point of letting its customers see the data on their previous purchases.

The first of the three: I liked the depiction of the 'knowledge continuum in Baker (2007):
Data -- Information -- Knowledge -- Wisdom.

I'd like to get as far as wisdom but, as my last post suggests, unless there's a sudden outbreak of common sense I'll struggle to get as far as data. I want data driven decisions for improved educational outcomes but...

The second of the three: "Academic culture favours analysis over action; institutions have placed a high degree of importance on their reputations rather than on improving the academic performance of their students." (Norris, 2008). Oh, how true that is! But in an ideal world...

The third of the three: John Fritz in his Elluminate session talked about using activity analysis as a predictor of success (as opposed to an indicator of success).

All of which brings me to the thought that if we have data, and we have predictors of success, are we not ethically obliged to share those data and patterns with our students?

Here's an analogy. We all know that there was a high correlation between wearing a red shirt on Star trek and coming to a sticky end on an alien planet. Even though it was just a correlation, and there was absolutely no suggestion of causality, did the guys in the transporter room not have an obligation to share this information with the young man in question and perhaps have a few spare mustard-coloured shirts on hand just in case?

I'd like to thank my friend Keith for the pun in the title.

Edited on Sunday 6 Feb: Added link to Analytics According to Captain Kirk

Learning analytics and how it differs from business intelligence

I'm taking an online course in Learning Analytics :

The course documentation defines learning analytics as: “the measurement, collection, analysis and reporting of data about learners and their contexts, for purposes of understanding and optimising learning and the environments in which it occurs.” I'm finding it challenging but fascinating and timely.

Everything I've read about Learning Analytics so far says that it has grown out of Business Intelligence and Web Analytics. It's a growing field and, according to Baker and Yacef, 2010, may bring "to educational research the mathematical and scientific rigor that similar methods have previously brought to cognitive psychology and biology." That sounds like something I need to know about.

But back to this analogy with business analytics. Consider a parallel universe in which Amazon is run like an ancient HE institution. Here's a guided tour from the Director of Sales: "We keep Karon's Name and address over here in this database and the records of the books she's bought this academic year over here in this computer. The two computers can't talk to each other but a secretary has access to both so that's OK. The marketing department can apply for access to the database of books sold this year if they wish but they'll have to ask the secretary to give them records from previous years because the data in the database is deleted every September and the secretary keeps copies of old records in a number of Excel spreadsheets on her C:drive. Of course, Karon can't see a list of what she's bought in the past. Why would she want to do that? She can see the data from some special offers we ran last year because it's been put in the WebCT gradebook for some reason. We must make sure that doesn't happen again."

Time to blog

This is an experiment. It works for housework, so perhaps it will work for blogging and I have so much in my head that needs to be put down in words I have to try something or I'll explode. So I'll repeatedly set my kitchen timer to 30 minutes and start writing. There will be a theme - the uses and abuses of data in education.

But then I took a photo and spent time looking for the power lead for my laptop and then I checked my email - look there's an article called the "unreasonable effectiveness of data" I must read... isn't that funny I've just reread Wigner's seminal article "The Unreasonable Effectiveness of Mathematics in the Natural Sciences" and I've only got two minutes left and I've got to upload the photo yet...