The University

from coursera to d2l: who’s gonna pwn you first?

This is from CUCFA President Meister’s Open Letter to Coursera Founder Daphne Koller, concerning Coursera pwning student user-assessment data.

Eventually, all students in my Coursera class will learn that data that they now provide to the company for free -perhaps so that it can grade them -,will be the private property of Coursera, which can then sell it back to them in the form of “services,” which could include their own performance record but also different “views” comparing it with that of students at better universities, those with higher test scores and with advanced degrees. The possibilities for renting this information back to its students are endless, not to mention the added possibility of developing other markets for the user-assessment information that Coursera will “own.”

But it’s not just Coursera who collects student data to sell back. Here’s The Register reporting on the British coalition government selling product

At the end of 2012, Education Secretary Michael Gove told Parliament that he wanted “to share extracts of data held in the National Pupil Database for a wider range of purposes than possible in order to maximise the value of this rich dataset”.

Ultimately, the government wants the private sector to tout “tools and services which present anonymised versions” of records on Blighty’s kids.

But getting pwnd is closer to home for MnSCU schools through D2L. D2L recently announced that they are creating products to sell student data back to the students “to improve student performance”

Featuring Student Success System™, an analytics engine that delivers fact-based and accurate insights on learning progress, the new Desire2Learn Learning Suite will improve learner engagement and instructor’s insight into each individual’s learning path.

“Harnessing big data and predictive analytics has transformed many industries, yet to date, the analytics to support next generation learning has been missing from education,” said John Baker, President and CEO, Desire2learn. “With today’s release, Desire2Learn will be delivering predictive analytics to millions of learners who will benefit from more successful outcomes. With this innovation, we can now provide valuable insights that will improve completion rates, lead to higher outcomes, and allow for the development of more impactful personal learning experiences.”

Keep in mind where the big data for those predictive analytics are coming from: faculty and students who have been using D2L for the past few years. Not just one university’s data but an entire system’s worth of data.

So D2L is finding its success not in the software platform it manages, which is an atrocious design error on stilts, but in using the data that its customers (schools, university systems) collect on their stakeholders (students and faculty). Years ago (206? 7?), when BSU moved to D2L, I was involved in the discussion locally. Some of us were concerned that D2L would collect and use data, but we were assured by The System that this wasn’t an issue. In the end, we never had a look at the contract with D2L, either. D2L’s disingenuousness is not new nor surprizing, and no one likes to hear that their vendor is parasitic. But they are.

D2L collects and aggregates data on classes to sell to vendors and students and, likely, back to the university.

So tell me why these scenarios aren’t likely, and perhaps even occurring:

– Say that D2L aggregates data on how often students pass reading quizzes in the Pro Ed programs across the MnSCU universities. Presumably, if students do well on the first pass through the quiz, the teacher is effective. If students need to take the quiz multiple times, the teacher is less effective. This wouldn’t be difficult to control for student variables. Evaluate which teachers are more “efficient” by those scores, then sell that information to students and to administrations. Students take the more “efficient’ – or is it the easiest? – teacher, and admins add the measure of “inefficiency” to the faculty member’s tenure and promotion evaluation. The admin, not the faculty, has the data to demonstrate it. D2L gets to claim they are improving the educational experience for students.

– A teacher has students use a Cengage textbook quiz bundle. D2L aggregates and sells scoring frequency data to Cengage. This lets Cengage revise their quizzes and textbook. Students and the state, however, do not receive remuneration on the data. Instead, Cengage releases a revised text, making the old text and quiz useless and requiring that both teacher and students buy new stuff, at a higher price. Students are creating their own increase in textbook prices. Cengage and D2L get to claim that they are improving the educational experience.

Three observable problems:

  • To be useful to D2 – that is sale-able –  the aggregated data must be decontextualized and relabeled as “best practices.” However to be useful to the teachers and students the data has to remain in context.
  • The state pays D2L for the software, a cost we openly pass on to students (We charge a fee for online courses). D2L then sells student performance numbers to back to to the and to others without remuneration.
  • Neither students nor teachers have any control of how the data is used, yet they both have vested interest in both their individual and collective performance.

In order to use the data that would help teachers become better – a better narrowly defined as what can be collected and analyzed –  we have to buy it back from the vendor who charged us for it in the first place. I like a situational irony as much as anyone else, but this one is too expensive for the humor.

I could be wrong about this – I don’t have access to the D2L contract. If I am wrong, if D2L isn’t using the data it collects to create products to sell back to those who generated the data in the first place, I’d appreciate a correction. But until then, I’m steering students I work with clear of D2L. It won’t make a difference, but I get to be smug.