[dropcap]W[/dropcap]hy is it so hard to mimic success in education? After years of trying to improve schools, good schools are not yet the norm in any large district. Why does one school do well, while another one in a nearby neighborhood doesn’t?

If education reform is to have any lasting impact, it can’t just happen in a few schools. Change cannot depend on an innovative teacher, because that teacher might leave the system. It cannot depend on a charismatic principal, because she may burn out. And it cannot depend on a progressive community, because politics change.

Models that work need to be able to be replicated broadly. But that’s been the sticking point for far too many education reforms. Once we know something works, how do we make sure it can take hold everywhere?

Pittsburgh might have an answer. And it starts with partnerships.

Educators at the Elizabeth Forward School District near Pittsburgh—tucked away in a former industrial stretch in the Monongahela River valley —have teamed up with Carnegie Mellon University’s Entertainment Technology Center and Zulama, an educational gaming firm, to develop the Entertainment Technology Academy, where students study the history and application of gaming, and then design and program their own educational apps and video games.

Researchers at the Learning Research and Development Center at the University of Pittsburgh, in collaboration with cognitive science teams at Carnegie Mellon University, have developed research projects like the Activation Lab, which creates pathways between formal and informal learning so kids can extend their interests in science or the arts beyond the classroom.  The Creative Tech Network, created by the Pittsburgh Technology Council, acts as a trade association for regional technology businesses, many of which are education-related.

There are many other examples in Pittsburgh of researchers, university labs, cultural institutions, entrepreneurs and schools forging dynamic partnerships to bring learning opportunities to kids in as many places as possible. And it is in this collaborative approach where we can see a new model of “scale.”

“When I look at Pittsburgh I get really excited,” said Amy Morton, executive deputy secretary at the Pennsylvania Department of Education. “The potential for scalability is really evident, not in the sense of scalability meaning growing something bigger and bigger, but rather replicating something community by community based on each community’s own unique potential for coalition.”

By working in tandem, Elizabeth Forward School District, Carnegie Mellon, and many other institutions big and small are learning from one another, testing and tailoring what works in education, and providing many opportunities for kids to connect with learning both in and out of school.

[dropcap]Y[/dropcap]et there’s another element that is equally important to the success of these partnerships, and this new model of scale. That element is the “glue” that holds the disparate partners together, ensures that their goals fit in with a larger plan, and helps secure resources to support new programming. In Pittsburgh, that organization is the Kids+Creativity Network, a coalition of schools, cultural institutions, university research labs, and community organizations working together toward a shared vision for the region.

While cities everywhere offer a robotics program or an art class, Pittsburgh takes it a step further by deliberately coordinating them to create a seamless playing field for kids— and for educators— to find a connection or a spark, whether in the classroom or outside of it. The city was also chosen by the U.S. Department of Education to be an Education Innovation Cluster, where schools, research teams, and entrepreneurs collaborate to create an R&D pipeline in education.

Photo courtesy CREATE Lab

According to Carina Wong, deputy director at the Bill & Melinda Gates Foundation, scalability is no longer about starting small and then ramping up with direct replications. Rather, it’s about tapping into the right networks, then adjusting until you get it right.

“You could do a pilot and then try to prove it works and scale it up,” Wong said. “But I haven’t found that to work. When I’m making investments, I look for founding partners who can help us solve and iterate on a problem. I keep expanding the networks around a problem to see what holds and what doesn’t hold.”

“We believe there is value in regional connections,” said Richard Culatta, deputy director of the Office of Educational Technology at the U.S. Department of Education. “We want to see research partners, schools, and developers of software and hardware sitting around the same table. That creates an awesome opportunity to leverage the strength of everyone.”

Culatta was initially inspired by the work of Michael Porter at the Harvard Business School, who studies “innovation clusters” in the business world like in Silicon Valley or the medical technology hubs near Cleveland. Porter was struck by how much physical geography still impacts successful innovation—even in the age of the Internet, when location isn’t supposed to matter anymore.

According to Culatta, researchers at universities and teachers in elementary schools know how to teach math better than anyone else, but they don’t know how to get a math-teaching tool out into the world. That’s where developers come in.

“You need people involved from the beginning who know how to scale,” Culatta said. This doesn’t mean turning over the reins to business; instead, it’s about partnerships.

Carina Wong of the Gates Foundation said that following this new model of scalability through partnerships with business “can make for strange bedfellows.” But, she said, it’s essential. “By bringing in so many networks of people, I knew I could get to scale because everyone I brought in had so many of their own different connections and networks.”

According to Wong, the education reform community also has a lot to learn from designers, including strategies like rapid iterative cycles. Mark Surman, executive director of the Mozilla Foundation, agrees. He said education reformers need to start thinking not only like designers, but also like product developers.

“One of the big problems is you don’t see people thinking about learning innovations as products,” Surman said. “That may sound harsh, but you see a lot of people coming at this with very fuzzy experiments that aren’t focused on the learner and don’t have scalability built into them. So you get lots of cool new ideas, but you don’t have, ‘How do we package this so it can be replicated?’ That’s a part of product design.”

Or, put another way, how do you package a successful curriculum and method of teaching so the school down the street can be as successful as the A-list school?

Photo: Joey Kennedy

[dropcap]B[/dropcap]ut not everyone sees this approach as addressing the “scale” issue. Education researcher and economist Greg Duncan worries that this approach lacks one key ingredient.

This approach, he wrote in an email, “really makes a case that program development would profit from the involvement of many actors, and might play out over quite a period of time as new circles of actors try it out and add to or modify it in important ways.”

But going to scale is more than just a group of local actors taking a core idea, modifying it to their particular context and in light of their own ideas. Somewhere in the process it has to be tested to see if the elements actually work. “We have no idea if program modifications make the program more or less effective, or perhaps ineffective, until we evaluate that version against a counterfactual,” he said.

“The best sequence, it seems to me, is to go through a highly iterative cycle of development, arrive at a version of a program that is most likely to succeed across contexts, and then test it in a number of those contexts. It may be the case that the program impacts are context specific, in which case the development and testing of a program is a tedious process indeed. But it could also be the case, as is hoped with the Common Core Curriculum, or Success for All, that one size does fit all and that fidelity to the program is the key to success.”

Maybe it’s simply a matter of terminology. Gregg Behr, executive director of the Grable Foundation, said he doesn’t even like to use the word “scale.”

“That implies we’re trying to replicate franchises or follow a recipe perfectly,” he said. “I’m more interested in the idea of ‘spread.’”

“Scale shouldn’t mean ‘every single McDonald’s in every single city needs to be exactly the same,’” Behr said. “It makes more sense to say, ‘Here’s an idea that was implemented based on evidence and has had some promising outcomes; then you replicate the idea in communities across the country in ways that make sense for your area.’”

Teaching is one part pedagogy and two parts alchemy. We’ve all been in a classroom where that magic happened, and we’ve been in classrooms where it didn’t.  Is it even possible to bottle that alchemy, and will it hold its magic for every child? Perhaps it’s an impossible goal. Perhaps it’s better instead to create a wide variety of opportunities for learning in different settings, just as they’re doing in Pittsburgh.

The editorial team at HiredPen contributed to this story.