The company, Coursera, plans to announce on Wednesday the start of a pilot project to check the identities of its students and offer “verified certificates” of completion, for a fee. A key part of that validation process will involve what Coursera officials call “keystroke biometrics”—analyzing each user’s pattern and rhythm of typing to serve as a kind of fingerprint.
The company has long said that it planned to bring in revenue by charging a fee to students who complete courses and want to prove that achievement. And Coursera has long recognized that its biggest challenge would be setting up a system to check identity. Other providers of free online courses, which are often called massive open online courses, or MOOCs, have decided to work with testing centers and to require students who want certificates to travel to a physical location, show an ID, and take tests while a proctor watches to prevent cheating.
The article goes on to explain that students will need to “hold up a picture ID in front of a Webcam, and then pose for a second picture of themselves, for an initial identity check,” and as a way of authenticating yourself after that, students will type a short phrase to register and then authenticate the pattern of typing.
Let’s just set aside the basic question of whether or not this would work. It is worth mentioning that the CHE article quotes a security expert who implies it might sort of work but if someone types differently because they are in a “bad mood” it might not work, and all you have to do is imagine typing at a different keyboard or with just one hand or something to come up with ways this could all be thrown out of whack. But let’s not go there.
If Coursera (and other MOOC startups) really thinks that the “biggest challenge” it faces is checking student identity, then they really aren’t aware (or forthcoming) about the real challenges they face. Checking identity seems pretty low on the list to me, though I am also a writing teacher who does not obsess over things like plagiarism either.
And let’s be clear: if verifying identity is a problem in online classes, then it is as least as large of a problem in face to face classes too. I’ve been teaching for almost 25 years now and I have never asked to see a student’s ID in a class. I don’t know anyone who has, frankly. The closest I’ve come to that is to get a student’s ID number when I’m advising them, and even then, I don’t check the picture on their ID. I assume students are who they say they are, but if the same person shows up throughout the term and tells me they are Joe Schmoe, I’m not at some point going to stop and ask for their papers or their typing to verify that.
No, in my view, the biggest problem Coursera et al face is finding a way for education and teaching to “scale” as well as learning or content (and as I have suggested in previous posts, I don’t think this will work) and to either convince “the world” at large (employers, for example) that certification in a MOOC course ought to count the same as a college course in the grand scheme of things, and/or to convince colleges and universities that they ought to accept Coursera certificates as transfer credits for certain classes. Stuff like identity or plagiarism or cheating is low-hanging fruit.
Another interesting aspect of this article is the explanation of much this certification would cost:
The company and colleges are still struggling to decide what to charge for the certificates, though in its latest announcement Coursera said the price would run $30 to $100.
“It’s a huge decision: You’re essentially setting a market,” said Daphne Koller, a co-founder of Coursera, in an interview this week with The Chronicle. “No one has ever priced this before.”
So why would someone pay for the verified certificates?
Peter Lange, provost at Duke University, which plans to offer one of the courses in the new pilot, said each free certificate would have a clear disclaimer on it: “It says something to the effect of, We cannot vouch that the person who got this document took the course or did the work.”
First off, Koller is flat-out wrong in claiming “no one has ever priced this before” because there are lots of different online certificates for training and, of course, Kaplan and Phoenix and other online proprietaries long ago figured out what to charge. This is a pattern with Koller, this idea that she and her partners at Coursera are in completely uncharted territory when it comes to teaching online.
Second, once again I have to wonder if Coursera’s academic partners (like Duke) are going to honor these certificates at their own institutions– that is, could I take a Duke run Coursera MOOC and have it actually “count” toward something if I were admitted to Duke? I am fairly confident that the answer is “no,” which raises another question for me: if the institution teaching the MOOC is not willing to itself honor the Coursera-issued certificate, then why does Coursera think that other institutions (like the EMUs of the world) will be willing to call that credit legitimate?