Dreambox Or Data Mining Nightmare?

December 31, 2018

Finally, more than a year after it was implemented, my kids’ school is hosting an “all about Dreambox” night. It will be on an upcoming Monday, from 6-7 p.m. Childcare will not be provided.

In other words, parents, please don’t actually come.

Dreambox is just one of several new educational technology (edtech, for short) programs invading classrooms across the Minneapolis Public Schools. It is a math-based computer program that kids can work on individually, at school and at home. It is all about algorithims and adaptive learning, but more about that later.

My youngest child is a 4th grader at a K-8 MPS site, and her class spends approximately 15 minutes per day, per kid, on Dreambox. Other parents have told me that kids as young as kindergarten are now using Dreambox at school for thirty minutes or so per week, and that some older students are being assigned Dreambox as homework.

I pushed back during the last school year when I first became aware that my child’s school was suddenly using Dreambox. I asked that she not use it in class until I had a chance to learn more about it. That’s because I research and write about edtech all the time for a writing gig I have, and so, I know to be skeptical of any new educational software products.

Often, the claims that precede them are little more than false fronts. Blow on them just a little and they fall right over.

Example. The Dreambox Night notice that was sent to parents recently, advertising the chance to learn more about Dreambox from MPS’s Dreambox rep, included this info paragraph:

DreamBox is unlike other learning products, DreamBox goes beyond tracking only right and wrong answers. Using virtual manipulatives, which build from the ground up to formatively assess learning and adapt in-the-moment of learning, DreamBox continuously analyzes and understands student thinking during the learning process, understanding what strategies students are using, where their gaps are, and when they are ready for increased challenges. With this deep understanding, DreamBox identifies trajectory, pace, and path to build competency and confidence, responding to students and providing the right level of scaffolding at the right time.

I know an ed tech marketing pitch when I see one, and so I popped this paragraph into Google and, of course, it comes straight from Dreambox’s own website. Parents would never know that, though, since no citation was provided along with the flowery language. One might easily assume, then, that the school staff planning Dreambox Night are so in love with the product that they wrote this themselves.

Therein lies the danger. Dreambox is a privately held, for-profit company. It is in the business of selling products to make money. It is part of the rapidly expanding global educational technology sector worth billions of dollars, thanks to investments from venture capitalists. Go on, do some research on the topic. It takes time to find even one source that critically analyzes the promise and hype of products like Dreambox. (Here’s a good one, though.)

That’s because we are in the middle of a gold rush here, and teachers and kids will likely be the tumbleweeds left behind when the investors pack up and leave town.

Let’s go back to Dreambox’s claims about its products for a minute, if you can get through the buzzwords (formative assessments for all!) without losing your mind. Dreambox promises that it can know what students are thinking and why they get a problem right or wrong. Using that information, the system then “identifies trajectories, pace and path to build competency and confidence.”

This is called adaptive learning. Computer-based programs like Dreambox decide what students should know, using algorithims (not a surprise, since Netflix founder and algorithim champ Reed Hastings has pumped millions into Dreambox). Here’s an overview of the concept from foremost education writer and personalized learning critic, Audrey Watters:

In other words, adaptive software focuses on the domain – the information that’s supposed to be taught; on the student’s responses – often using “bayesian algorithms” to determine the likelihood of a student getting the next question right or wrong; and on the instruction – what sorts of hints or feedback a student might need to move forward.

That’s it, right there. Programs like Dreambox, it can be argued, are less about revolutionizing education and more about repackaging traditional, top-down delivery models into market-ready, investor pleasing bundles. Yes, Dreambox is fun, as my 9 year old has informed me, but is it really about teaching and learning? Or is it more about data mining and pigeon holing kids?

The Dreambox website openly gushes about how the product can “capture fine-grained data” and “adapt in real-time to student performance.” Here’s more:

DreamBox is the next generation of individualized learning. DreamBox learning’s detailed data mining and analysis not only continuously adapts to the learner within a given lesson, but adapts the sequence of lessons as well.

We should all pause at this and ask some thoughtful questions. Is “individualized learning” an appropriate goal for young children? Is data mining of kids all just a-ok, especially as Dreambox reps claim it harvests “50,000 data points per hour”? Or should we be wary, as parents, teachers and school administrators? Data mining, as a function of the ed tech gold rush, can be dangerous territory, although it is often what makes ed tech such an attractive investment.

Think of the kids using products like Dreambox as lab rats. Their data–age, gender, location, math scores–can be packaged and sold, so that more ed tech products that offer even more tailored, individualized learning experiences can be created and sold. Of course, this is part of the bargain we all seem to make by having cell phones, using Facebook and otherwise living in a time of virtually unregulated data mining, but is this all okay for students? And are parents even aware of what’s going on?

There is no doubt that educational technology can be beneficial, especially for students with special education needs. Programs that help students with dyslexia complete writing assignments without the fear of spelling errors, for example, or those that help students with autism spectrum disorders are undoubtedly appealing. And they are often cheaper to use (cheaper than lowering class sizes, of course) and reflective of the kind of gaming and online lives many students already lead.

But can they ever take the place of real human interaction? At my kids’ school, Dreambox does not do that. It was brought in last year, as I understand it, in order to help teachers manage their class sizes, alongside the expectation that they–as single teachers alone in their classrooms, mostly–will at once differentiate instruction for close to thirty students while also prepping them for standardized tests and simultaneously nurturing the whole, social-emotional child.

Dreambox provides a break for teachers, or so the argument goes. It allows teachers, as I have observed, to work with one small group of kids while another group takes their turn at Dreambox.

In the week before winter break, I stopped into my kid’s school to help out with spelling groups. The kids in the classroom were still finishing their hour long math block (that’s an issue for another day), and I could see my daughter sitting in front of a Chromebook screen, happily chatting away with her neighbor. Once we were in the hall for spelling practice, I asked her if she had been using Dreambox.

“Well, no,” she told me with a sly smile. “I was on Santa Tracker!”

I wasn’t mad; I thought it was funny. She doesn’t really care too much about math equations or sticking with a program like Dreambox. She’d rather talk with her friends, have fun and do something creative or, perhaps, silly. She is nine, after all.

I will go to the Dreambox Night at her school. I will go with an open mind, but also with a bit of due diligence under my belt. Recently, while doing research for another edtech product, I came across this NPR piece about personalized learning. It provides an overview of this concept, baked into products like Dreambox, and offers critiques like this:

So there you have it. Personalized learning: a cost-effective, efficient way to improve direct instruction through pacing, while giving young people a little more autonomy. What’s not to love?

Jade Davis has thoughts about that. She’s an expert in emerging technologies in education, and the director of digital project management at Columbia University Libraries. When she thinks of personalized learning, “I think of kids with machines that have algorithms attached to them that move them through learning at the pace where the student is.”

Does that excite her?

“No, it doesn’t,” she answers. “Because learning is a collaborative process. When you take away the ability for people to make things together, I think you lose something.”

And, she adds, there’s another issue. Many recent critics have pointed out how biases, such as racial biases, can be baked into all kinds of algorithms, from search engines to credit ratings. Davis argues that educational software is no exception. “It’s going to sort students. It’s going to stereotype, put up roadblocks and make assumptions about how students should be thinking.” In other words, what’s sold as “personalization” can actually become dehumanizing.

Earlier in 2018, I asked MPS to provide any contracts between the district and Dreambox. At that point, Dreambox was being implemented on a school-by-school basis; I am not sure if that has changed yet. The dollar amount per school that flows to Dreambox isn’t high, but perhaps those data points are worth more?

2 thoughts on “Dreambox Or Data Mining Nightmare?

  1. Another problem with these packaged materials that rely on computers is that those who have computers in their home have access to whatever they need to learn the program or process or Dream box or whatever..while those without computer access on a regular basis outside of school will not be as familiar with these processes as those who have them available…
    Tired of our schools being given the “magic solutions” to low test scores in the form of programs and new gimmicks when what is so important is connecting kids to each other, to their teachers, and having low enough class sizes to make that feasible…

  2. Great job, again, Sarah, of shining a light on a topic crucial to education in Minneapolis and far beyond. There’s hours of talk and pages to be written before this gets sorted usefully. I will go back to what I said when I commented on The Christensen’s Institute’s Disruptive Class book back 2010 and to what you alluded to in your post – this is technology being used to substitute for a real teacher instead of to enhance the student teacher relationship. (Here’s my post from 2010 https://developingprofessionalstaff-mpls.blogspot.com/2010/08/response-to-m-horns-comments.html)
    There are multiple definitions of ‘personalized learning,’ ‘blended learning,’ ‘adaptive learning,’ ‘computer assisted learning,’ etc. The issue always comes back to – are teachers being adequately supported to provide as much as possible for their students? The answer is usually, NO?

Comments are closed.