Numbers matter. They track progress, regress, and
effectiveness. They allow us to understand populations fully and to make
changes for the better—and student data is no exception. Without student data,
there is no way to account for, and therefore no way to serve, students who
face barriers to education.
Last month, the Data Quality Campaign (DQC) released findings from its analysis of 50 state report cards, evaluating each individual state and the District of Columbia based on the data it tracks in state report cards and how accessible they are online. States are required by federal law to provide these annual report cards showing who their students are, and how both those students and the schools are performing—information that helps everyone from parents to policymakers.
The DQC analysis revealed that report cards are often “silent on whole groups of students,” with as many as 13 states not reporting enrollment data on student gender, 7 on disability status, and 6 on race or ethnicity.
These findings, when situated within the context of a broader conversation about information and privacy, reveal just how inconsistent student data collection and reporting can be, irregularities that have real consequences for the information it actually provides.
A model state report card should include the most recent assessment data, student performance data, and educator and student demographic data. Without this information, it is difficult for parents to make the right choice of school for their child, for federal leaders to allocate funding, and for the state itself to know where improvements must be made.
It’s impossible, however, for data to be a resource if they do not accurately reflect student populations. As collection and reporting currently stand, there are large gaps in identity and demographics that, if filled, would help data-collecting entities toward achieving their goal of serving all students.
Although many data-collecting entities gather student identity information, few do so in a way that accurately represents diverse student populations. The National Center for Education Statistics (NCES), for example, uses “mixed-race” as a catch-all category. Doing so groups together millions of students who may be learning under very different circumstances, in part because of their specific racial identities. If a parent has a child who is biologically multiracial but identifies as monoracial, they may select the race by which the child normally identifies. Or, if the child is able to “pass” as one race or another—that is, if they physically appear to the general public as monoracial—a parent may opt to select that race because of the scholarship or social opportunities it presents. Another parent in the same situation may select ‘mixed race’ for their child because they understand that to be the technically-correct choice. It is then impossible to know the actual racial makeup of students in each category, and therefore impossible to know exactly how they are performing or how policies affect them.
Similar technicality issues arise when selecting students’ gender. The Department of Education’s Office for Civil Rights (OCR), the largest body of publicly-available student identity information, collects enrollment, performance, and discipline data by sex and not gender, which does not account for transgender students. “Sex” is commonly understood as a biological category and “gender” as the way someone identifies and operates in the world. At present, different states have different laws regarding the way students are allowed to indicate gender. In some states, the parent of a transgender student may select the sex their child was assigned at birth, and the parent of another student may select their child’s preferred gender. In the states that do not allow transgender students to be represented as their actual gender, those students are miscategorized, invalidating the data. Information with these gaps does not serve transgender students because there is no way to know how many of them are enrolled, how they are performing, or whether they face a disability.
Students offered services under IDEA, or the Individuals with Disabilities Education Act, are also grouped together without nuance in most data collection. IDEA includes students with physical, intellectual, and learning disabilities, but both CRDC and NCES indicate only student IDEA status, not disability type. True, being required to report more specific data on students’ individual disabilities may present privacy concerns for parents, but data as they are collected now, could do more for the students they aim to serve.
As all of this suggests, student demographic, or identity, data counts among the information most critical to civil rights, providing information on how students of particular races, genders, and abilities are enrolling and performing in schools. It’s unclear from DQC’s findings which demographic data are collected but not reported, and which data are simply not collected.
Furthermore, DQC’s analysis highlights the importance of which data we collect and the way we collect it. The very way information is sought out simultaneously reflects and determines the way identity is conceptualized. Collecting it in a way that does not reflect students’ true identities will maintain a system that further excludes and disadvantages the very students it ignores.
These findings also point to a larger gap in education data where a focus on student representation should be: Are student data gathering operations achieving their goals? Do they work for the students for whom they’re designed?
According to OCR, the goals of Civil Rights Data Collection are simple: to “collect data on leading civil rights indicators related to access and barriers to educational opportunity at the early childhood through grade 12 levels,” and to be a resource for educators and parents who seek data on student equity and opportunity.
Now, on the eve of a Trump presidency, accurate representation is more important than ever. The incoming administration’s collective record of fighting against the students who most need representation is indicative of what’s to come. Trump will soon be in a place of power to act on the rhetoric that built this campaign, keeping students of diverse identities at a disadvantage. The sharp spike in identity-based violence in schools since the election underscores the dangerous attitudes and violence that this election has emboldened. If this pattern continues—as it likely will—without easily accessible data to advocate against it, the status of already-marginalized students will become worse. Unlike the violence that followed Obama’s election, this violence is in the name of the president. It is in the name of disdain for the very groups of students whom education data ignore.
Without representation in data, students remain invisible. They are invisible to the public, to their schools, and to policymakers who have control over their well-being. Just as identity is complex, so too are the answers to the issues it presents in data. Without more accurate categories for racial, gender, and disability identities, widely-accepted definitions of these categories, and directions for parents of children who may belong to more than one, millions of students will continue to face overwhelming barriers to education.