The inside-the-tent debate over virtual charter schools has highlighted an important issue about standardized testing and school accountability: There’s still plenty of room to improve the ways student performance gets measured.
First, a quick recap: Studies have found students enrolled in full-time online schools often struggle to make learning gains. This has alarmed many charter school advocates. Virtual charter school operators and their supporters have responded that their students often face unique issues that aren’t reflected in their test score gains.
Many of the people raising concerns about virtual charters acknowledge there could be hidden “x-factors” that affect students’ results, but say the information that’s available now is troubling even with that caveat. Virtual charter supporters, however, argue parents’ decisions about how best to educate their children should carry the day.
Matt Wicks, an vice president for data and policy at Connections Education who recently joined our podcast to discuss these issues, said lots of people in the online learning industry believe there’s a better way to gauge the academic progress of students with special circumstances, like those who are behind on credits when they enroll, or those who face non-academic issues like long-term medical treatments or travel outside the country.
In a portion of our interview that didn’t make the final edit, Wicks said that better way still needs to be found.
“We’ve done a lot of work on different ways to try to measure students success, and to be honest, I don’t think we have it figured out yet,” he said.
For that reason, he said, it’s worth looking at a new provision in the new federal school accountability law, which allows states to tinker with new approaches to testing.
Education Week recently explained the innovative assessment pilot program in the Every Student Succeeds act. It will allow a handful of states to try new assessments, starting in a handful of districts and eventually growing statewide. The new testing approaches will have to be accessible to all students, including those who grew up speaking a language other than English, or those with special needs.
“We see this as a bridge,” said Lillian Pace, the senior director of national policy at KnowledgeWorks, a non-profit that works to better “personalize” learning for students. Pace has studied the pilot and its implications closely.
When lawmakers sat down to write ESSA, the field wasn’t ready, she said, to “paint the picture of the next generation of assessments”—the kinds of tests that can give teachers a fuller, real-time picture of what their students know while still helping states and district improve schools.
“While states may not be successful on the first try, we expect this to be a learning process for all,” Pace said. “The goal is for participating states to build and refine a system over time that really transforms student learning.”
Meanwhile, in the Harvard Law Review, Betheny Gross and Paul T. Hill of the Center on Reinventing Public Education argued the change in federal law could promote “democratic experimentalism” — meaning states could make a concerted, systematic effort to tinker with new ways of doing things like measuring student performance.
Under democratic experimentalism, independent communities (e.g., states) engage in four related activities: (a) they work together to test out alternative approaches to reach a goal that none has yet attained, (b) they rigorously assess the results and evidence about conditions leading to success and failure, (c) they create a mechanism by which these results are fully shared, and (d) they commit to using past results as the starting point for further experimentation.
…
Democratic experimentalism goes beyond mutual imitation and sharing of fads in two ways. First, entities (e.g., states) have to create internal conditions under which diverse experimental initiatives can be fully imple- mented and tested. In K–12 education, states need to make it possible for localities, and even individual schools, to experiment with factors that in the past have been standardized by law and regulation. In many respects, this approach is a departure from the inputs-driven policy of the first part of the twentieth century, but potentially consistent with the more recent performance orientation sparked in part by NCLB.
Second, states must commit to serious analysis of results and sharing them through an objective mechanism that no single set of states or policy advocates controls. Professors Dorf and Sabel suggest that federal entities like the European Union or the United States national government can perform this function. Again, many states are currently well positioned to participate in this function, given the data and analytic capacity they have developed [under No Child Left Behind and Race to the Top].
As Wicks put it during our interview, “there’s kind of recognition that, yes, there’s a lot to learn” when it comes to new kinds of assessments. “You have to start on a small scale. Some of them will be promising, and others won’t.”
He added the coming wave of experimentation could be “something that, over the next three to five years especially, will be very much worth watching.”
