Purdue’s Non-Answer on Course Signals

Michael Feldstein, Nov 12, 2013
Commentary by Stephen Downes

One of the claims being made by learning analytics is that it increases student retention. But this is the sort of claim that needs to be empirically verified and open to public scrutiny. One such program, Purdue's Signals, is being scrutinized and the critics are finding issues. In September, Mike Caulfield examined a Purdue study that claimed "taking *two* classes using Course Signals boosts graduation rates by 21%." However, "two years later, that retention effect has disappeared entirely," he writes. A month later, Alfred Essa corroborated Calfield's results using a simulation. "The simulation data shows us that the retention gain for students is not a real gain (i.e. causal) but an artifact of the simple fact that students who stay longer in college are more likely to to [take the CS course]." Purdue research scientist Matt Pistilli responds, "The analysis that we did was just a straightforward analysis of retention rates. There’s nothing else to it." But "Purdue’s credibility is on the line," says Michael Feldstein, and they're going to have to come up with a better answer than this. See also this 'issues explainer' from Feldstein.

Views: 0 today, 216 total (since January 1, 2017).[Direct Link]