Thursday, February 07, 2013

A Prediction And A Wish

In January the day after the Illini Men's Basketball team got pulverized by Minnesota, I wrote in the tail end of an email to some friends:

Still taking enough pain meds to believe the Illini can bounce back and make the Final Four.

There is an art to prediction.  If you're going to be wrong then do so spectacularly.   Now, with a 2-7 record in the Big Ten, I don't have enough confidence in the team to feel like watching a game will be enjoyable, the outcome predetermined, another loss.  The fortunes of this team have been more Dr. Jekyll and Mr. Hyde than most (though remarkably similar to what happened last year) but there is a lesson in it that the recent past doesn't predict the future.  First, you may have the underlying model wrong and thus make an incorrect inference based on what is a small sample.  In the case of the basketball team, the model that was floated in December was that Coach Groce had really energized the team and the new offense was much better for these players than the one that Coach Weber used.  That doesn't look right now.  Second, even if you've got the model right, most people fail to understand regression to the mean.  I made both errors in writing that line above.

Economists are trained in efficient market thinking, and for stock market prices in particular, I think it a reasonably good first pass at not using history to predict future price movements because history is already fully accounted for in the current price.  There are occasional anomalies, sure, so those aware of them can take advantage.  But the more serious problem is naive investors buying near the peak and selling near the trough, because they are forecasting poorly.  The Efficient Market Hypothesis is a humbling theory that emphasizes our ignorance.  It is helpful to have it in mind.

Nevertheless, after Norma sent me this link to Online Class on How To Teach Online Classes Goes Laughably Awry, she had been enrolled in the course and was saddened by what happened, I felt a need to say something, especially since I wrote my post The Learning Technologist Becomes A Luddite not too long ago. So here is my prediction.

Neither Thomas Friedman nor David Brooks nor any other well known columnist who has written recently in a bullish way about MOOCs and the future of Higher Education will write anything at all about the experience in that Coursera instructional design class.  These columnists have a track record of bullishness on our economic issues (of which education is a substantial component) and that will not change.  Further, they'd like to see that bullishness spread, so even if they are aware of the attendant risks with any new approach, they are loathe to talk about these because it cuts against the diffusion goal. 

For myself, I started to think about energy policy rather than higher ed.  For years I've believed in a $.50 or $1.00 tax increase on a gallon of gasoline as the right thing to do in encouraging conservation and paving the way for alternative energy sources.  Tom Friedman has advocated for this too and it is quite possible by my being for this policy idea comes from him.  While I still think this is a good idea, I have to admit that the alternatives are each risky, so especially if we expect quick results from such a tax we might be terribly disappointed.

Getting back to Higher Ed, MOOCs, and that particular Coursera offering, I will on the one hand take the efficient market approach and not try to predict what will come next in this dimension, but on the other hand hope that the experience with this recent course gets us to seriously consider alternative approaches.  I believe, quite strongly, that the online components of on campus courses should be open for the most part.  The grade book needs to stay private but much of the rest of the course can be publicly available.  (Copyright concerns can be managed in a way where the restricted material is behind an authentication wall but the rest of the course is not and in the rest of the course reference is made to that material where outsiders are invited to find it on their own if they can.)  This issue has been on the back burner, at best.   Mainly, it's not being talked about at all.   I've done that in my own teaching (most recently here) to good effect for those students taking the class.  It certainly would be a benefit to students who are considering taking the course in the future.  And it might very well be a benefit to outsiders who can peruse course materials for their own use and possibly contribute to class discussion.  Further, it would ignite some talk about OERs and bring that alternative more to the forefront. So my hope is that other people take up this notion and we move the discussion about online learning to where it is part of what on campus instruction is rather than separate from it.

No comments: