Advocates Say Education Officials Hide Too Much Data in Name of Student Privacy
Transparency advocates say Colorado is hiding too much information about how schools perform, but state education officials counter that they’re just following laws to protect student privacy.
There’s no dispute that how much information can be found about how students are performing in local school depends a great deal on the size and composition of the student body. The question is whether the trade-off is necessary to protect student privacy.
A report from the Colorado Right to Know Coalition found the state didn’t release information about how students who qualified for free and reduced-price lunches were performing on state standardized tests in one out of every four schools. Some schools also lacked information by racial group, English language learner status and, in very small student bodies, grade level.
Van Schoales, president of A+ Colorado, a coalition partner, said the state hasn’t shown any real-world examples of people using the data to identify specific children. The coalition supports suppressing scores when an individual student could be identified, he said, but parents need to know how well schools are serving children in their kid’s subgroups.
“We think it’s really important that parents and communities be able to understand,” he said. “We’re still far away from knowing how subgroups of kids are performing in Colorado.”
For years, the Colorado Department of Education has hidden scores from public view when any student subgroup has fewer than 16 students. Since 2015, state law has forbidden revealing information that could identify students when used in combination with other data, which the department determined requires it to hide the scores of some larger groups.
Parents receive their child’s scores, and teachers and administrators receive the full data for their schools.
Joyce Zurkoski, chief assessment officer at the education department, gave an example of a fictional district with three schools named A, B, and C. If the public had access to data about standardized test scores in the district and schools A and C, they could figure out how many students scored at each level in school B, even if school B had so few students that its data normally would be hidden, she said.
The state also can’t report larger groups if every student falls in the same category—for example, if no one in school has a passing score on the annual achievement tests, Zurkowski said. Parents would still receive the average score for their child’s school, which is enough to know that most students are performing poorly, unless the school had fewer than 16 students—but in that case, the community likely knows how the school is doing, she said.
“We decided at some point that people knew how to subtract,” she said. “There are a whole bunch of parents that say, ‘I don’t want my neighbor to know (my kid’s score),’ and that seems like a reasonable position.”
States can adopt slightly different data policies to try to balance transparency and privacy, but there’s no plan that will satisfy both demands in every situation, said Ben Shear, an assistant professor of research and evaluation methodology at University of Colorado Boulder’s School of Education.
For example, many people think it’s important to know if all subgroups of students are benefiting equally from their education, but if a school has only one Hispanic student, you wouldn’t want to violate the student’s privacy by breaking down the data by ethnicity, he said.
“There’s very much tension and a trade-off,” he said.