directs policy initiatives including the sponsorship of the Ed-Fi data standard for the Michael & Susan Dell Foundation. The Ed-Fi solution is a universal educational data standard and tool suite
(unifying data model, data exchange framework, application framework, and sample dashboard source code) that aligns with Common Education Data Standards. The Ed-Fi tool suite is designed to integrate information from a broad
range of existing education data sources. It facilitates data comparisons and interoperability, and allows vendors to develop reusable products across
TVA: What’s new with the Ed-Fi initiative?
LF: The Ed-Fi initiative is experiencing great momentum on adoption. We launched with five states committed to Ed-Fi tools (DE, CO, LA, TN, TX.) These
early adopters are in various stages of putting dashboards to use in the classroom. Three more states have come on board (AZ, KS, NM) and signed licenses,
and others are in licensing discussions. We’re also working closely with the Shared Learning Collaborative, (SLC)
which extends the reach of Ed-Fi tools to states like North Carolina, New York, Illinois and Massachusetts. Progress has been faster than we initially
predicted. In one year, we’ve gone from announcing the existence of the standard to seeing states using the tools and realizing tangible benefits.
TVA: What’s in version 1.1?
LF: The new release, scheduled for November, includes improved dashboards, content tagging and student assessment tracking. A big part of our goal with the
release was to incorporate feedback from entrepreneurs who wanted better support for moving data in real time - we wanted vendors to be able to focus their
money on the product instead of the plumbing. So we focused the new release on providing them with the ability to easily address shifting technical
requirements and to design reusable products. The goal is to empower them to address the changing requirements of different districts, states and schools
without major code overhauls. The comment period for the new release is currently ongoing. We’re hoping to capture and address additional feedback as we
finalize the code for this new release.
TVA: Where do you see promising work?
LF: Delaware was the earliest adopter and has made really significant progress. In fact, within 12 months of contracting with vendors for major system
components, teachers in the state’s pilot districts had begun actively accessing critical insights using dashboards powered by the Ed-Fi solution. The
Colorado Department of Education has leveraged the Ed-Fi solution to streamline the data pipeline from district to state. They’ve overhauled data systems,
simplified business logic on collection, and reduced 18 static collections of data points to 8 data interchanges. Like Delaware, they have a very capable
team of internal and external partners, which has made implementation quick and low-cost.
A number of other states are evaluating how best to overhaul of their infrastructures in order to better support the work teachers are doing in the
classroom. Finally, we view SLC states as another a great case study. We’ve already learned an enormous amount from the work the SLC is doing to help
states personalize learning by putting actionable data to use. And we’ll build on that, and incorporate it into future releases.
But what’s maybe even more exciting is the way we’re starting to see vendors respond to the Ed-Fi solution - not just in terms of adoption, but in terms of
understanding that products need to be reshaped in order to meet new state requirements to provide data that’s actionable in the classroom. So, for
instance, we’re seeing vendors begin to relook at dashboard design from an end-user perspective. They’re no longer just thinking through what’s standard
practice. They’re looking through the lens of making something better than what’s freely available via Ed-Fi dashboards.
More critically, we’re seeing vendors begin to evaluate whether their products do more than capture the kind of compliance data that was required in the
past. Are they capable of capturing the level and variety of data teachers need in the classroom? Are their products focused on end-user applications? Can
they support integration of data from many applications? We’ve also started to see states and districts really beginning to think through some of the
implications of data ownership, and of maintaining rights to the data that’s collected rather than of defaulting to vendor ownership. That shift has real
implications in terms of budgets.
We’re also seeing sparks of more radical innovation, particularly in the blended learning space. Developers and vendors are identifying use cases and
developing new products that we hadn’t originally envisioned. Between SLC, state and district licensees, 34 percent of US students and 36 percent of US
teachers will soon be benefitting from the tool. Adoption at that scale has real potential to creating a crossing the chasm scenario.
TVA: We’re writing a paper with recommendations on the electronic student record; how much data needs to follow a child grade to grade and school to
LF: There are several ways to address that question. We start with a view from the classroom: What would a teacher need to know about a new student to do
the best job on day one? What data is most critical in helping teachers meet the needs of each child? We got input from more than 2600 educators when we
did our initial work. The data set they identified as most valuable is pretty broad. It includes attendance and behavior patterns, demographics, credits,
the results of formative and summative assessments, grades, and maybe even some example of student work. We aren’t alone in this view of what data needs to
follow kids, of course. States often identify this type of data set as an area of need. Arkansas and Colorado have shown specific interest in it.
TVA: Within an SLC state, a teacher should be able to access any of a student’s previous information, so the ‘what data’ issue may be as much about
permissions management as records management.
LF: We aren’t in a position to give specific insight into how SLC will handle student data, but what we can address is the critical role of permissions
management and good governance of student data. The fact is that state and district governance practices need to evolve as their data systems and data
practices evolve. Reevaluation of governance is a key aspect of any implementation process. We see districts and states undertaking very specific projects
to update their governance policies and practices to ensure they provide the highest levels of security and confidentiality for student data. In many
cases, Ed-Fi implementations provide a sort of forcing function in terms of helping states better define role-based authentication and permissions
TVA: What about out-of-school vendors?
LF: Careful vendor selection and vetting is critical no matter what data standard is used. It makes sense that some vendors have the ability to contribute
to learner profiles; to that extent giving a vendor access to limited data may be something a district wants to consider. But in terms of permissions and
governance, states and districts should apply careful due diligence to vendors in the same way that they do to their own employees.
TVA: The Michael & Susan Dell Foundation has made a big commitment to the Ed-Fi initiative.
LF: Improving the quality of education available to all kids is a major pillar of the foundation’s work. We see the commitment to rolling out a scalable
and sustainable data solution as key to success. We still find teachers using index cards to keep track of their kids. And even where there are information
systems in use at schools, teachers need to sign on six to twelve times and don’t have any way to compile the data. It’s crazy. Ed-Fi tools offer a clear
path out of that situation. And if you listen to project leaders involved in the states where the Ed-Fi solution is already in use, it’s better, cheaper
For an FAQ on the Ed-Fi solution see //www.ed-fi.org/faqs/#what-is-ed-fi.
The opinions expressed in Vander Ark on Innovation are strictly those of the author(s) and do not reflect the opinions or endorsement of Editorial Projects in Education, or any of its publications.