Understanding Racial Imbalances in Special Education


As another April has come and gone, so has another World Autism Month. The Light It Up Blue campaign celebrates each spring with a renewed push for greater understanding and acceptance of individuals with autism spectrum disorder.

And with greater attention to autism (hopefully) comes greater attention to learning and developmental disabilities more broadly. In the context of education, this means greater attention to the who, what, and why of special education (SPED) services.

Special education provides a public education, generally through implementation of individualized curricula, to students with intellectual, learning, developmental, and/or physical disabilities. [1]

Or does it? In the last decade, researchers and policymakers have begun to take a closer look at the students enrolled in SPED. Red flags have emerged, to say the least.

The Numbers Don’t Add Up

At the forefront of concern is evidence of substantial disproportionality in SPED enrollment. Disproportionality arises when a group, such as a racial or ethnic minority, is represented in SPED at a greater rate than they comprise within their school, community, or nation. For example, if a school is comprised of around 65% white students, we should expect that the SPED classrooms are also comprised of around 65% white students.

Yet in nearly every state, rates of SPED enrollment show evidence of overrepresentation of minority groups. [2]

Now, before delving into the possible factors contributing to such disproportionality, it is worth noting that special education is still relatively new to U.S. public education. SPED was first enacted in 1973 and has gone through several policy iterations to reach its current form: the Individuals with Disabilities Education Act (IDEA).

IDEA mandates that education services for children with disabilities must meet students’ individual needs and must take place in the least restrictive environment possible (ideally with non-disabled students). As well, and perhaps most important for the current discussion, is the mandate that SPED assignment can happen only after appropriate enrollment procedures have concluded. These procedures include aptitude and achievement tests, teacher recommendations, and considerations of the student’s cultural background. [1]

Despite IDEA’s requirements, however, SPED services do not appear to be distributed equitably. [3] Enrollment data show that students of color consistently experience disproportionate inclusion in SPED, and this issue has actually come to the attention of Congress more than once. During both the 19th and 22nd Annual Reports to Congress, the Office of Special Education Programs (OSEP) and the Office of Civil Rights (OCR) reported that students of color may be being misclassified or inappropriately placed in SPED, that such placement may be a form of discrimination, and that SPED students may be receiving services that do not meet their needs. [4]

Red Flags

What kind of disproportionality are we talking about? Let’s look at a snapshot of some of the numbers and contexts that researchers have been tracking:

  • despite Black children constituting only 17% of total school enrollment, they comprise 33% of children diagnosed with mental retardation (now referred to as intellectual disability, ID) [5]
  • Black boys are, on average, 5.5 times more likely to be diagnosed as emotionally disturbed (ED) than are white girls [6]
  • American-Indian boys are, on average, nearly three times more likely than white girls to be diagnosed with a learning disability (LD) [6]
  • among students with disabilities, 57% of Hispanics are educated in partially separate or substantially separate settings and denied access to inclusive settings, compared to 45% of whites [7]
  • English language learners (ELL) are 27% more likely to be placed in special education during the elementary school grades [8]

Where to Begin?

Those are some pretty troubling statistics, and researchers have endeavored to get to the bottom of them. But as one might expect, disproportionality is a complex, layered issue.

And a potentially misleading one. For example, a natural response to reading the numbers above might be that disproportionate overrepresentation is worse than disproportionate underrepresentation. We would be remiss to take that thought as a blanket statement, though. After all, while overrepresentation may reflect heightened disapproval of minority students’ behavioral or academic performance, underrepresentation may reflect minority students’ struggles going unnoticed. And that latter possibility isn’t any better than the former.

For example, the diagnosis of Intellectual Disability in Black students has been shown to decline as poverty increases. [2] In other words, the poorest Black students may be the least likely to be identified as having ID. But, there is still a disproportionately high rate of Black children in SPED who are diagnosed with ID overall.

Such nuanced findings may suggest that Black students are being over- and under-monitored based on their socioeconomic background in addition to, or in lieu of, their academic profile.

Thus, the story that needs to be uncovered is not only the extent of disproportionality (i.e., the raw numbers) but also the forms (i.e., the diagnoses) and the causes.

What’s Happening with SPED Assessment?

Turning to causality in particular, some researchers have hypothesized that the assessment procedures required by IDEA for SPED enrollment may be less rigorous in practice than on paper. In an investigation of how qualitative factors, such as personal beliefs, may affect the rigor of psychological/educational evaluation, Harry, Klinger, Sturges, & Moore (2005) investigated community perceptions of the validity of SPED referrals throughout urban schools in Southern Florida. [9] Extensive interviews with teachers, administrators, and families uncovered a high level of confidence in school-ordered assessments.

That is, the interviewees believed that students would be referred for, and enrolled in, SPED only after a true need for such services was found. Which sounds good! But, paradoxically, this high level of confidence may actually lead to harmful results.

Because from the get-go, students may be vulnerable to inappropriate SPED placement if members of their family, school, and community are unlikely to examine a referral critically. Further, given that studies have found Limited English-Proficient students to be more likely to be placed in SPED, families that experience a language or cultural barrier to their child’s school may face particular disadvantage in advocating for their child.

These same researchers also found that teachers’ perceptions of a student’s learning difficulties, as well as their perception of dysfunction existing within a student’s family, predicted their students’ SPED assessment results. This may indicate a complex process through which a teacher’s perception of a student influences the nature of their interactions (e.g., challenging the student less due to lower expectations), which in turn contributes to lower levels of student achievement and, eventually, consideration for SPED.

Other researchers, however, have suggested that psycho-educational assessment is not the main event in SPED placement at all. [8] Rather, disproportionate referrals may arise from the ongoing failure of regular education classrooms to serve racial and ethnic minority students. They argue that it is the quality of a student’s classroom instruction, and the level of management within the classroom, that should be most emphasized during student assessment. This emphasis would allow for underachievement to be seen as the result of a poor learning environment rather than individual student failure.

What Other Factors Underlie Disproportionate Representation?

Overall, most researchers have concluded that disproportionality in SPED is the result of:

  • subjective student identification practices (e.g., teachers’ interpretation of the same behaviors differently depending on the student);
  • blatant violations of IDEA’s guidelines;
  • and antiquated systems of SPED funding based on category of disability (i.e., schools receive more money if a student is diagnosed with Intellectual Disability than if diagnosed with Dyslexia). [10]

Yet other studies have begun to take a new, ecological approach in their investigations. For example, based on the assumption that low-income students are more likely to be students of color, several researchers have asked: is poverty is associated with increased risk for SPED enrollment?

In one such study, Strand & Geoff (2009) analyzed the 2005 Pupil Level Annual School Census – a data set of 6.5 million students in England. [11] The authors found that poverty and gender explained more disproportionality in SPED enrollment than did ethnicity; but, the overrepresentation of students of color in SPED was still significant even after controlling for poverty. It appears, then, that some degree of interplay between individual (e.g., academic strengths and weaknesses, learning support at home) and environmental (e.g., socioeconomic conditions, teacher and societal beliefs) factors significantly contribute to placement in SPED classrooms.

Getting to the Bottom of it

So far, researchers seem to have a lot of pieces of the disproportionality puzzle in a lot of places. How do we put them all together–at least enough so that we can begin to do something about it?

To start, Oswald, Coutinho, & Best (2005) recommend a new research agenda. They advocate for disproportionality studies to focus specifically on disentangling social factors (such as systemic bias) from individual factors (such as differential susceptibility) as an underlying cause of over- or underrepresentation in SPED. [6]

These authors argue that studies should investigate whether students of certain racial or ethnic groups are differently susceptible to schooling contexts such as low-quality instruction, loose classroom management, or particular academic interventions. Under the theory of differential susceptibility, it is perhaps so that some students fare better in special education classrooms than others, making them more likely to be placed back into regular education.

They also advocate for assessment procedures that compare an individual’s performance to the performance of students of similar characteristics. For example, the achievement of a Hispanic female youth should be compared to the average performance of similar female students within their school or district (i.e., not their non-Hispanic classmates). If differential susceptibility to an aspect of the educational environment exists for some racial or ethnic minorities, assessment procedures that compare similar students may provide the most accurate depiction of an individual successes or challenges.

No Time Like the Present

It is clear that disproportionality exists within SPED. But what it less clear is why, or how to fix it. Given that it is a relatively new addition to public education, however, we can hope that the inequity currently seen in SPED may not yet be as deeply rooted as some of the challenges that regular education faces (e.g., school segregation).

Nonetheless, time is of the essence for new research! It is only with a better understanding of the roles that various factors play in SPED disproportionality that the development (and enforcement) of policy interventions can commence.

[Editor’s note: this post was written by Lindsay Clements. The initial byline, saying that it had been written by me, was incorrect. My apologies for the mistake.]


[1] U.S. Department of Education Office for Civil Rights (2010). Free Appropriate Public Education for Students With Disabilities: Requirements Under Section 504 of the Rehabilitation Act of 1973, Washington, D.C.

[2] Parrish, T. (2005). Racial disparities in the identification, funding, and provision of Special Education. In D.J. Losen & G. Orfield (Eds.), Racial Inequity in Special Education (pp.15-37). Cambridge, MA: Harvard Education Press.

[3] McDonald, K.E., Keys, C.B., & Balcazar, F.E. (2007). Disability, race/ethnicity and gender: Themes of cultural oppression, acts of individual resistance. American Journal of Community Psychology, 39, 145-161. doi:10.1007/s10464-007-9094-3 [link]

[4] U.S. Department of Education (1997). Nineteenth annual report to Congress. Washington, DC: Author.; U.S. Department of Education (2000). Twenty-second annual report to Congress. Washington, DC: Author. 

[5] Losen, D.J. & Orfield, G. (2005). Racial inequity in special education. In D.J. Losen & G. Orfield (Eds.), Racial Inequity in Special Education (pp.xv-xxxvii). Cambridge, MA: Harvard Education Press.

[6] Oswald, D.P., Coutinho, M.J., & Best, A.L.M. (2005). Community and school predictors of overrepresentation of minority children in Special Education. In D.J. Losen & G. Orfield (Eds.), Racial Inequity in Special Education (pp.1-13). Cambridge, MA: Harvard Education Press.

[7] Garcia Ferros, E. & Conroy, J.W. (2005). Double jeopardy: An exploration of restrictiveness and race in special education. In D.J. Losen & G. Orfield (Eds.), Racial Inequity in Special Education (pp.39.70). Cambridge, MA: Harvard Education Press.

[8] Artiles, A.J., Rueda, R., Salazar, J.J. & Higareda, I. (2005). Within-group diversity in minority disproportionate representation: English language learners in urban school districts. Exceptional Children, 71(3), 283-300. [link]

[9] Harry, B., Klinger, J.K., Sturges, K.M., & Moore, R.F. (2005). Of rocks and soft places: Using qualitative methods to investigate disproportionality. In D.J. Losen & G. Orfield (Eds.), Racial Inequity in Special Education (pp.71-92). Cambridge, MA: Harvard Education Press.

[10] Reschly, D.J. (2000). IQ and Special Education: History, current status, and alternatives. Unpublished paper, National Academy of Sciences, National Research Council, Washington, DC.

[11] Strand, S., Geoff, L. (2009). Evidence of ethnic disproportionality in special education in an english population. The Journal of Special Education, 43(3), 174-190. [link]

tags: category: L&B Blog

Leave a Reply

Your email address will not be published. Required fields are marked *