Bill & Melinda Gates Foundation

Data and the Negative Nudge: Reflections on Making Data Work for Teachers and Students

June 15, 2015

Imagine walking through a cafeteria line and choosing lunch options. Recent research shows that the layout of the food will strongly affect your choices. This is what the behavioral economists, Richard Thaler and Cass Sunstein, describe as a “nudge.”  Whether intentional or not, the world is often structured to “nudge” us toward one action over another.

I thought of nudges when I read Teachers Know Best: Making Data Work for Teachers and Students. The study compiled interviews from 4,650 teachers who answered questions about how they use data to drive instruction in their classrooms.

There was good news and bad news. The good news is that more than 8 in 10 teachers are constantly looking for ways to engage students based on who they are, and nearly as many believe that data help validate where their students are and where they can go. The belief in data-driven instruction is there.

The bad news is that all too often teachers have to work hard to make the data actionable. More than two-thirds of teachers said they are not fully satisfied with the effectiveness of the data or the tools for working with data. One teacher in the study explained how much time it took for her to input data. “I just want to teach,” she said. Instead of making her planning more productive or fruitful, analyzing data takes away from her planning.

When it comes to data-driven instruction, we’ve got a negative nudge problem. It’s as if we laid out the cafeteria food to nudge you toward sweets and then said, “Make healthy choices.” You can do it but it takes a lot more effort than it should.

I experienced the negative nudge first hand when I was a principal at E.L. Haynes Public Charter School. During my five years at E.L. Haynes, our students achieved significant gains on their end-of-year assessments. A lot of that success was because of data-driven instruction. Every seven weeks or so we gave interim assessments and then followed up with a full day of professional development. We organized the data, analyzed the data, and created “Re-Teach Action Plans” based on the data. The result was a culture that was laser focused on student learning. Everyone was focused on how he or she could contribute.

But the process had serious shortcomings.

Data analysis is only as valuable as the action that comes out of it. Too often, in spite of all the time we devoted to data, our actions were not strong enough. Our teachers were negatively nudged to treat the re-teach plan as a literal re-teach; as in, teach it the same way you taught it last time.

Could we change the nudge?  What if we made it simple to find and use targeted lessons that got at conceptual understanding?  Three of the school’s top teachers started to capture their lessons on screencasts, and I put the videos on a website where they could be accessed and used by other teachers anytime, anywhere. Thanks to a Next Generation Learning Challenge Grant, the project grew. We moved to capture the expertise of top teachers from across the country and put it at the tip of all teachers’ fingertips.

LearnZillion now has more than 10,000 lessons and resources—all crafted by top teachers in the country (last year close to 4,000 teachers applied for our 140 “Dream Team” of lesson authors)—and they’re free. This summer we’re working on building a full, OER (Open Educational Resource) Curriculum for K-8 math.

Building a full, digital curriculum will allow us to address another negative nudge: time. Right now it takes way too long to turn data into action. The seven-week cycle we used at E.L. Haynes was an eternity when it comes to student learning.

In a digital world, we shouldn’t have to wait this long. The OER curriculum will integrate formative assessment questions on a daily basis, allowing teachers to get immediate feedback. Teachers will also get anticipated misconceptions and an intervention playlist that’s ready to go if they see the misconception crop up in class.

Not only will digital curriculum help address the time problem, it will also address the over-testing issue. Teachers don’t need to drop everything and give paper tests when they are assessing student learning daily as part of an authentic learning process.

My hope is that three years from now a similar report will find that the nudge has gone positive. Instead of working hard to use data, a teacher will have to work hard not to use data to address the needs of his or her students. It will all be right there—as easy as choosing a slice of pie from the cafeteria line. 

 
blog comments powered by Disqus