Federal special education officials and civil rights advocates are pushing back against a recent study that takes on the conventional wisdom that minority students are steered into special education by biased educators.
Michael K. Yudin, the assistant secretary for the office of special education and rehabilitative services, says he wants to hear more about the methodology behind the study on minority underrepresentation in special education, but that the department is following its own statistics on the topic. And at least one civil rights activist and researcher calls the report “fatally flawed.”
The study was published online in the journal Educational Researcher, a publication of the American Educational Research Association. It was accompanied by a New York Times opinion piece with the provocative headline “Is Special Education Racist?” Paul L. Morgan, an associate professor of education at Pennsylvania State University, was the lead author of the underrepresentation study. He and his colleagues have done other work on minority enrollment in special education.
Looking at a national sample of about 20,000 children tracked from kindergarten to 8th grade, the researchers found that minority students are less likely than similar white peers to be in one of five common disability categories—emotional disturbance, intellectual disability, “other health impairment,” specific learning disability, and speech and language impairment.
In the study, the researchers controlled for variables they say are connected to educational outcomes: English-language status, birth weight, insurance status, household income, and mother’s marital status, among others. They also controlled for the child’s academic achievement and behavior.
When the children were compared after accounting for those confounding factors, the probability for being identified for special education was always lower for minority students.
Education Department Response to Special Education Study
“The approach to this study is absolutely interesting, and it’s different,” said Yudin on Thursday. “We’d love to have a conversation with them to learn more, and to understand better about some of the conclusions they drew.”
But the findings do not change the view of the department that significant overrepresentation of minorities in special education is a real problem, Yudin said. While the researchers worked with a sample of children, the Education Department gets data each year from each state about exactly how many children are identified with disabilities, and their race.
“We’re concerned that in the real counts, it is common in states to see significant overrepresentation,” Yudin said.
And that disconnect between the sample that the researchers used and the real numbers is just one of the reasons why this study and its conclusions are inaccurate, said Daniel J. Losen, one of the editors of the 2002 book Racial Inequity in Special Education. Losen is also the director of the Center for Civil Rights Remedies, an initiative at the Civil Rights Project/Proyecto Derechos Civiles at the University of California, Los Angeles.
Sampling Flaws in Special Education Study?
Looking at the statistics collected from the states in fall 2012, American Indian, black, and Native Hawaiian or Pacific Islander children ages 6 through 21 were identified by the Education Department as having a higher “risk” of being enrolled in special education compared to the special education student population as a whole.
But the sample the researchers used in the new paper started off with minority underrepresentation. “They started with a sample that proved their point,” Losen said.
Losen agreed that underrepresentation is a problem in some situations; it’s just not the national concern that has been suggested. And, “the idea that civil rights advocates are only on one side of this issue is simply wrong.”
Morgan understands the concerns about the different samples. The different data sets have strengths and weaknesses, he said, but the sample they used is intended to be nationally representative. The data collected by the states also don’t allow for correction of any confounding factors that the researchers used in their study.
Ultimately, Morgan said, educators “shouldn’t focus on the right proportion of children. We should focus on the individual child level, and let’s provide the help that child needs.”