Policies on ethnic disparity awry compared to research
A recent article published in Educational Researcher underscores what many of us have long suspected – that our measures for ethnic and linguistic disproportionality have discounted crucial factors leading us to erroneous, even damaging, conclusions.
The carefully constructed study suggests that when eligibility rates in high incidence categories are examined in a cohort with adjustments for covariates – such as socioeconomic status – minorities are not over-represented in special education but, in fact, are under-represented. The article flies in the face of the perspective that has been advanced by the U.S. Department of Education for almost a decade and, ironically, comes to light as the national discussion on race has intensified in the wake of the Charleston tragedy and Ferguson shooting.
In 2007 Alexa Posny, then Director of the U.S. Department of Education Office of Special Education Programs wrote a memorandum to State Directors of Special Education saying, in part, “greater efforts are needed to prevent the intensification of problems connected with mislabeling minority children with disabilities; African-American children are identified as having mental retardation and emotional disturbance at rates greater than their white counterparts;…studies have found that schools with predominately white students have placed disproportionately high numbers of their minority students into special education.”
These statements take their cue from the 2004 reauthorization of the Individuals with Disabilities Education Act wherein various procedural and financial mechanisms were put in place to prevent and correct the disproportionate over identification of ethnic groups in special education.
The language in the law and Posny’s subsequent statements imply that special education has been used as a vehicle in an implicitly biased educational system to wherehouse children of color apart from the mainstream.
But the article, entitled, “Minorities Are Disproportionately Underrepresented in Special Education: Longitudinal Evidence Across Five Disability Categories” was posted online late last week, indicates the opposite conclusion. That is, that certain ethnic groups, when compared to the general population, have had not fully utilized the protections and supports offered through special education and that the disparity is due to socio-economic and environmental factors unrelated to race.
In other words, there is an equity issue but it has to do with lack of full access to specialized services rather than the use of those programs for segregation.
The results bring to the surface many of the other re-occurring issues about disproportionality and concomitant legal guidelines that have been part of an undercurrent of skeptical observation for years now.
One glaring contradiction, that reflects the findings in the aforementioned study, is that federal regulation of disproportionality only applies to over identification, not its equally evil twin, under identification. This willful neglect stands in contrast to federal regulations that require educational agencies to maintain a “comprehensive child find system.” The policy is even more glaring in light of the 2001 Compton Unified School District v. Addison decision by the Ninth Circuit Court essentially declaring that a failure to properly identify a child was equivalent to denial of a free appropriate public education.
Nevertheless, in 2012 the Education Department announced that it would no longer collect information about the underrepresentation of ethnicities in special education, the rationale, being in part, that it obscured the focus on overrepresentation, which, of course we now know, could be fallacious.
Equally disconcerting is the improper identification of English learners in programs for students with disabilities that has been the basis for litigation in a number of situations, most notably Diana v. California State Board of Education. While education agencies are required to submit data regarding English learners, there is no requirement that public agencies statistically measure disproportionate representation.
English learners, however, remain overrepresented in special education in school districts with a small number of EL students and underrepresented in those districts with a large number, according to a report entitled “Truth in Labeling” published by the National Education Association and the National Association of School Psychologists in 2007.
Another apparent oversight is the lack of rules for disproportionate gender representation, even though they would certainly be justified. The NEA and NASP study reported that nearly 75 percent of the students in special education and 76 percent of the students considered emotionally disturbed are male.
Regardless, in comments entered into the federal register, regulators provided the following rationale for not requiring proportional calculations with respect to gender: “Although States will be collecting data on the gender of children with disabilities for other purposes, the Act does not require an analysis for disproportionality on the basis of gender. We are concerned about increasing the burden on States. Given that there is no statement of congressional intent indicating the need to do this analysis, we do not believe it should be included in the regulations.”
Finally, as Howard, Dresser, and Dunklee document in “Poverty is not a Learning Disability: Equalizing Opportunities for Low SES Students,” a “substantially higher percentage of children” with learning disabilities come from families with incomes below $25,000 than from other economic strata. In spite of this and similar evidence, there are no requirements to calculate or report on eligibility determinations in relation to socio-economic status.
That national policy is subject to political preferences should not be news to anybody but it is particularly disturbing when it flies in the face of artfully designed research from reputable sources.