Gordon Freedman is president of the National Laboratory for Education Transformation, which includes partnership centers at the University of Texas Austin, the University of California Santa Cruz, and Los Alamos National Laboratory.
How is data important to policymaking in education?
Education data is hugely misunderstood and misapplied. The use of data in education needs to come into parity with how big data is understood and used in other fields. This can be done through a better understanding of two policy frameworks which are currently not joined.
The first is a human capital framework, or a macro-education frame. The United States has to compete globally, but education is delivered and guided locally. This is a significant disadvantage given that most countries have a single education system where policy, funding, and performance are managed centrally.
The second framework is an public investment framework, or a micro-education frame. Education averages at least 25% of every state budget and a similar percentage comes from municipal budgets. Since more than half of the states in the US and many localities have reduced their annual expenditures, and cost-paring continues while pension payouts rise, we are cutting instruction at a time when we should be investing in it to create more jobs and new industries. The impact is acute in low-performing urban and rural schools.
What are the big gaps in the data or research related to education?
Data in education, like every other sector, is extremely important. However, educational data usage is constrained in ways that would seem odd and inefficient to the consumer, IT, financial, pharmaceutical, or security sectors. Large-scale data collection, as mandated by the Federal No Child Left Behind law, requires states to report on annual yearly progress (AYP) of students in order to receive Federal funds. While the merits of annual "summative" testing are often debated, what is not often discussed is the fact that this data primarily travels in a single direction—away from the student.
While data gets fed back to the school districts, it does not come back in time or in a form useful for addressing individual learning needs of students. Unlike consumer data and other use-data, in the case of education, the end user is not in the data loop.
Daily and weekly "formative" tests are given frequently without any standard reporting nor correlations to the annual "summative" test. So, from a data-driven policymaking perspective, there are many disconnects.
What's the long-term vision for National Laboratory for Education Transformation?
NLET is organized to bring the best minds, methods and organizations—from outside education—to address the problems mentioned in the two earlier questions. We believe that the actual stakeholders in commerce, industry, and security need to be part of the community that reconstructs education for the challenges of this century—the same way the leading industrialists helped organize education in the early 1900s.
Our vision for NLET is four-fold, but starts with data: constructing actionable data ecosystems that can be understood and utilized by policymakers, educators, parents and students.
NLET is intent on creating a dynamic map that displays macro-education data and related data in a visualized system. Like Google Earth, a user can zero down to the micro-education data for a local school or district. We need to have a highly interactive data landscape system to better understand education across all the localities.
Data properly conceived and used in the macro framework could help guide states to create competitive human capital, while the same data could help states in the micro framework better spend and deploy funds locally.