In part 1 of my anniversary blog, I documented IES' progress on many fronts; but there's a lot more to do.
First, and foremost, I hope that the three NASEM studies we have commissioned will provide a strong foundation for the next five years at IES. These reports—one on updating the research mission of IES, one on the future shape of NCES, and one on modernizing NAEP technology—will be delivered in time to help IES celebrate its 20th anniversary in 2022. I hope the NASEM research report will be immediately actionable as we fashion the following year's RFAs. The NCES report will help us modernize the center's data collections and approaches. And reforming NAEP's technological infrastructure is essential to keeping NAEP the gold standard of large-scale assessments. I asked NASEM to concentrate not on the past—but to develop blueprints that IES could follow in the near and mid-term future. I also asked them to think about what products, data, reports, etc. the nation needs and only then to think about the structures that will best get us there. A focus on desired outcomes, rather than existing practices and products, will hopefully allow us to find creative ways to better meet the needs of the American public, rather than simply continuing to fine tune what we already do.
I hope to continue to focus on better identifying the audiences for our work—and making sure that we produce data and reports that speak to their needs. We have done audience segmentation studies (excuse the jargon), and I think we have a much better idea about whom we need to reach to increase the impact of our work. Part of this effort will be a continued push to make the What Works Clearinghouse and practice guides more accessible and usable. And I hope to see the Regional Education Labs (IES' "boots on the ground") continue to work closely with states to solve pressing education issues.
Related to keeping our promise to make our data more accessible and usable, over the next several years, IES will redo our (rightfully unloved) website. Some of the changes will be less visible functions, like a digital portfolio governance and an updated content management system, while other changes will dramatically affect user search and the site's look and feel; together our new website should make IES' resources easier to find and use.
I hope to establish a Center for Excellence in Education Data Sciences within IES. While this center would not have the statutory standing of the other centers, it should create a powerful force in IES propelling all our work into the modern era of data collection, integration, and analysis. I am particularly hopeful that an infusion of data science and data scientists will have a big effect on our administrative data division. Under the Evidence Act, we are already charged with linking data across the federal government to provide better insights into current conditions and policy—data science should help us meet that mandate better.
There are so many buzzwords associated with data science—and many of the promises of data science are yet to be fully realized—so it might be tempting to just dismiss it as the newest bright shiny object. But data science feels like the "real deal"—and IES needs to assume a leadership role in using the ever-expanding world of administrative data, including what's sometimes called "fast data." To be sure, these data should be governed by strict privacy policies—and much of it may be beyond the purview of government agencies altogether—but IES needs to improve the information we collect and disseminate to the American people using modern data techniques.
I hope to help IES be a good shepherd of the $100 million in new funding from the American Rescue Plan. I have written about some of our plans elsewhere, so I will just highlight a few here. IES will work hard at building evidence about the effectiveness of the many programs and initiatives launched to help in learning recovery. This will give us the opportunity to focus on partnerships with state and local education agencies in ways that focus on outcomes rather than processes. I hope to see the State Longitudinal Data System (SLDS) program play a role in that work.
SLDS can play an important role in other initiatives as well. A recent NCES survey of state practices showed that many states have linked their K–12 data systems with, for example, early childhood and postsecondary education systems. Fewer have linked their SLDS to workforce data—but given the important tie between a good education and a good job, I'm hopeful that we can work with more states to achieve these needed links. Given the legislative foundation of SLDS in K–12 systems, we will need to be creative in how we support expanding the links between postsecondary education and workforce data.
I also hope to create new programs to continue supporting use of SLDS data to answer questions raised by researchers, by state policy makers, and by parents and students. With around $1 billion of federal money invested in building those state longitudinal data systems (and who knows how much money states spent), we need to increase the return on investment. This will require expanding the types of users who gain access to these data, as well as helping those new users employ the data to answer important education research and policy questions.
Ultimately, we will likely need another big investment in the SLDS infrastructure to reflect all the changes in data management and data science over the last decade—and this time around, I hope that these sites are built not as data warehouses (which was the original lens informing so much of the first generation of SLDS build) but rather as much more user-friendly sites. Amazon is a master of logistics and runs warehouses better than most. But we never visit their warehouses; rather, they have a front-end consumer interface that draws data seamlessly from their warehouse. But their interface is designed with consumers, not the warehouse, in mind. Rethinking the underlying model of SLDS with a broader audience in mind could help make those data sets more useful and increase the ROI of these data systems.
I hope that IES can continue to evolve into an organization that learns from failure. I have tried to emphasize to IES staff that the only failure in failure is the failure to learn. Further, if we don't fail that means we didn't try to reach far enough. I also keep emphasizing that just about everything we try should be viewed as an experiment—and as with most education science experiments, the most likely outcome is that what we want will fall short (often far short) of what we expected. Learning from projects that fall short is as important as learning from projects that reach their goal.
I hope to continue our investment in transformative research, through partnering with NSF, private foundations, and state/local governments. The transformative RFA (which we will continue to recompete) is just an indicator of deeper structural and cultural changes I hope to see in IES. And I haven't given up on the ever-elusive idea of an ARPA-ED (with, of course, IES as a central actor).
Like Fermat's last theorem, I have solutions to many of these problems but don't have the room right now to write them out.
Finally, I hope I haven't come across as excessively self-indulgent. But I think these developments (past and future) affect you all, so you deserve to be kept informed. As always, feel free to contact me: email@example.com