Education Business Competition Looks at Data Use
The third annual national Education Business Plan Competition, held here last week by the University of Pennsylvania Graduate School of Education and the Milken Family Foundation, offered some interesting insights into how the next generation of education entrepreneurs are thinking about research and data in education. There were a flurry of mostly Web-based tools and programs for educators and students, but in the end, the most interesting presentation was for an education intervention that never got off the ground.
Alexandre Scialom, 32, of San Francisco, swept the awards with theCourseBook (not to be confused with Facebook), an online city-by-city compendium of local and online classes, rated by students just as sites like Yelp! review restaurants or services. Mr. Scialom's site design, now in beta testing for the San Francisco area, won both the $25,000 grand prize and an additional $25,000 prize from the Hewlett Foundation's Startl Prize for the best open-source education business.
As he explains it, if you're spending the night in a new city and want to find a good spot for dinner, it's relatively easy to hop online and use one of the many local review sites like Yelp! to identify the best local hamburger joint. If you're a high school student trying to find a GED program or a principal trying to add an online French class, "63 percent have trouble."
Several of the nine nonprofit and for-profit finalists used extensive data collection and sharing to approach education problems, from taping teacher classes to provide online mentoring to allowing school staff to automatically upload attendance, grade and other data via a bar code scanner. For example, Mr. Scialom said he would provide feedback reports to schools on what worked for students in a online photography class, or which teachers had the best instruction ratings.
"There was a lot of stuff around technology ... around trying to figure out ways to capture information and analyze it, and use it. There was a lot of stuff about data this year that I found interesting," said Douglas Lynch, the vice dean at the University of Pennsylvania Graduate School of Education and co-creator of the competition.
Yet at the end of the competition the presentation that struck me most was fellow San Franciscan Matt Pasternack's story of GoalPost, a seemingly popular reading intervention that never got off the ground. Mr. Pasternack, a former teacher in Harlem and developer in New York City's School of One, developed GoalPost, a social networking platform to allow students to log their reading lists online, reviewing and recommending books to friends while also allowing teachers to track how quickly students moved from book to book. The program would also analyze the data in the same way sites like Amazon.com do, to recommend available books that students might also enjoy.
Very quickly, Mr. Pasternack received a merger offer from a children's book publisher and started to see interest from school districts. "When we took the tool out to schools, administrators absolutely loved it and we had dozens of principals who wanted this in their schools as quickly as possible," he recalled.
Yet when he started to test the tool in classrooms, he found teachers liked the program—but they didn't use it. "It wasn't embedded into the daily structure of the classroom the way we though that it would be."
Why? According to Mr. Pasternack, while all of the classes he tested the reading program in had access to computers, either in the classroom or in a computer lab, they simply weren't ubiquitous enough to integrate computer-based programs into instruction. For example, while students previously had logged their books and opinions about them in paper journals, after the computer program was implemented, a student had to "check out a computer mouse from the teacher, go to the computer, log into the computer, navigate to our Web site, log into our Web site, navigate through the system to get to where they needed to do journaling, and they probably got about 50 percent as much work done on our system, our 'optimized' system, as just using pencil and paper."
Teachers either had to send students to log books a few at a time, or schedule class time to the school's computer lab every time students needed to log books. Moreover, Mr. Pasternack found that even in the wealthiest schools in Silicon Valley, there was always at least one child who did not have a computer at home and had difficulty handling daily homework assignments online.
The field test was a wake-up call, he said. "The platform of the classroom is paper. The daily transactions between teachers and students are all paper," he said, and while classes might be less efficient when compared to technology-rich environments, "adding small incremental increases to technology actually decreases productivity. Until we make that big jump up to better design, we need to be very, very careful about how we introduce these little bits of technology into classrooms."
He has since put the GoalPost project on indefinite hold and is now working on a separate project on analyzing student data.
At a competition in which many judges and contestants alike voiced concern about the time and money required to research implementation and effectiveness of interventions before putting them to market, Mr. Pasternack's experience proved a cautionary tale of the difference between an idea on paper and in the classroom.
"He could have sold it to every school district in the country, because
it says a lot about his moral fortitude that he said I'm not going to do this just to make a buck if it's not going to add value," Mr. Lynch told me.