Conor Williams: Shut Down Student Data and Shut Out Lights on Ed Woes
If you think about it, you start every day with data. You roll out of bed into the dim pre-dawn shadows, trip over some unknown detritus on the floor, and fumble for the switch on the wall. Flip it, and lo, everything gets clearer.
You can see the [expletive] problem now — your son’s toy fire engine, your copy of “Infinite Jest”, your partner’s empty wine bottle, whatever clutters your particular floor—because you have a clear view.
It’s basic human behavior. It’s how we develop experiences and make sense of the world: We can only solve the problems we can outline, and we can only outline problems when we have clear data on what they are.
As (my favorite) progressive philosopher John Dewey once put it, “To mistake the problem involved is to cause subsequent inquiry to be irrelevant or to go astray. Without a problem, there is blind groping in the dark.”
This is a particular difficulty in education, where the data are pretty limited. Want to know how many dual language learners there are in the United States? We have a few piles of dated, incompatible, and contradictory data for you. Good luck crafting policies and divvying up resources to support this population of growing and shrinking students.
Want to know if minority students are over or underrepresented in special education classes? Well, as far as we know, the answer is … both. Good luck adjusting our screening processes and updating our regulations to rectify those, erm, mutually exclusive problems.
Want to study American students’ reading abilities in grades K–2? There’s essentially no comprehensive national data for you—even though third-grade reading proficiency is a key priority for many policymakers.
Is it any wonder that education debates are so ideologically-driven, so frequently unfocused, and so prone to ad hominem? The absence of good data usually makes it difficult to get basic agreement on what we’re arguing about. In education, for every study showing that Policy A works, there seems to be a counter-study drawing on different data to show that Policy A is catastrophic.
So we fall prey to a basic rule of politics: The less clear the facts, the wilder the rhetoric.
Which is why it’s truly crazy that the current push to decrease federal education accountability is morphing into an attack on federal data requirements. This threat is most prominent in the ongoing attempt to rewrite No Child Left Behind (NCLB), the federal government's main K–12 legislation—but it’s not the only place good data is under assault.
Sen. David Vitter, R-LA, has floated a new bill to replace the Family Educational Rights and Privacy Act (FERPA) that has researchers very worried.
If passed, the bill would require every family to “opt-in” to permitting their students’ data to be used by principals, districts, and the state to design programs to improve their learning. So without specific family permission, students would remain in the shadows of any research effort to evaluate which instructional methods, curricula, or policies are working in their schools.
Vitter’s bill would make these changes in the name of privacy, but the bill is so broad that it would make it much harder for states to efficiently maintain basic oversight of their education systems. Instead of providing targeted changes to meaningfully protect children’s personal information, it provides vague, blanket prohibitions that would stymie the collection of basic student achievement data by educators who need it to evaluate their own work.
According to the Data Quality Campaign’s Rachel Anderson, “The biggest thing that this bill would do would be to keep states from using student-level data in their own data systems … to do most of the things that they do now, like early warning systems, high school feedback reports showing how graduates do after high school, and so on.”
In other words, it would mandate a mountain of opt-in paperwork for schools and undermine states’ efforts to design data systems that provide useful information for the education policies they’re implementing.
A letter signed by more than 1,000 education researchers and organizations responding to Vitter’s bill warned, “This legislation would have a devastating impact on education research essential to best serving students across the country and improving the quality of teaching and learning in our schools.”
They should know: Good student achievement data is the backbone of their work. Or, better, to stick with the visual imagery I used in the opening, data is the source of light that makes good research possible. That's how researchers know what works: Is a district's pre-K program affecting kindergarten readiness? Is another district's pre-K program supporting better reading abilities in third grade? Is a new teacher evaluation system influencing student achievement? And so on.
Data make it possible to outline problems — and identify solutions. There are a number of ways to collect the data and do this work, but the best research requires data from common assessments that can be compared across schools and districts.
But data aren't just a source of light. They're also leverage. Which is why civil rights groups advocating on behalf of families of color have pushed hard to protect federal testing rules and accountability for underperforming schools.
These groups know that American schools, districts, and states are prone to ignoring children of color. They know that this systemic racism is harder to perpetuate when there’s transparent data on schools’ performance—and consequences for chronic failure. (Note: Information on African-American and Latino families’ views on testing and data are scarce, but one California poll showed strong support.)
So — with all this in mind — why is education data so unpopular? Well, when it comes to what students know and can do, good data depends on those common tests. But tests make most everyone uncomfortable—and they actually make a core of (mostly) privileged activists angry. Data can be threatening to those who are comfortable with the status quo.
Good data on problems with our education system is discomfiting. Parents — especially those who are accustomed to being served well by the status quo — don’t like hearing that their kids’ schools are performing below standards. Educators don’t like hearing that a test built by some distant company has determined that their students are behind, or that there are big achievement gaps between various groups of students in their schools. And most everyone in education (and beyond) hates facing consequences for performing poorly over a period of time.
That’s what’s driving this all-out, bipartisan assault on tests. That’s why federal accountability is withering in Congress.
Among people who work in education right now, there are two big pieces of related conventional wisdom: 1) federal education accountability is too unpopular to maintain and 2) the data and transparency gains from federal testing rules are still valuable.
But it's hard to keep those two things from collapsing into one another. Efforts to eliminate federal mandates requiring states and districts to intervene in failing schools are straying into efforts like Vitter’s bill.
That’s no accident. Reform opponents have realized that it’s hard to advance the cause of accountability if the public doesn’t have data illuminating the extent of educational inequity — both in terms of resources allocated and student outcomes achieved.
In other words, undermine good, comprehensive, and comparable data, and they’ll have shut out the lights. They’ll have stopped educational research, reform, accountability, and (maybe?) civil rights in their tracks.
Get stories like these delivered straight to your inbox. Sign up for The 74 Newsletter