Ethnicity and Special Education Research: Identifying Questions and Methods
Coutinho, Martha J., Oswald, Donald P., Behavioral Disorders
There is enormous state-by-state variation in the percentages of students with SED served under the Individuals with Disabilities Education Act (IDEA, P.L. 105-17, as amended) and a substantial gap between estimates of the prevalence of SED and the actual number of students served (Oswald & Coutinho, 1995). Parents, teachers, mental health professionals, and policymakers continue to disagree about what represents serious emotional disturbance, one consequence of which is varied eligibility requirements across programs. Because IDEA is an education law, in order to be identified as a student with SED a child must demonstrate serious and ongoing behavioral or emotional problems and fail to make adequate academic progress. Unless there is a substantial impact on the child's educational performance, there is no disability under IDEA. This requirement, as well as the stigma and ambiguity associated with SED and other definitions (Reschly, 1995), can allow differences related to the individual (e.g., culture, ethnicity, socioeconomic status) or to the sociocultural beliefs of educators, evaluators, and parents to influence inappropriately who is identified as having SED in a particular community. When this occurs, the likelihood increases that a child will be misidentified or provided with an inappropriate education.
The importance to the public of preventing the potential for harm to a child who is misidentified or served inappropriately is evident in the equal protection clause of the Fourteenth Amendment, Title VI of the Civil Rights Act of 1964, the procedures for evaluating children suspected of a disability under IDEA, the data collection and monitoring responsibilities of the U.S. Department of Education's Office for Civil Rights (OCR), and court decisions spanning more than 20 years. Unfortunately, public concern remains that although America's student body is becoming more and more diverse, children who are nonwhite, non-native-English-speaking, or poor continue to be overidentified as having disabilities and to be served in more segregated placements than their peers (Artiles & Trent, 1994). The recent amendments to IDEA (P.L. 105-17) expand requirements to report and correct inappropriate identification, placement, or service patterns. In general, research has supported public concern, but the picture is unclear because studies have varied so much with respect to definitions of minority representation, findings across ethnic groups, and technical methods.
Obtaining Accurate and Defensible Estimates of Disproportionality
Considerable research has documented the overrepresentation of African-American students in special education (Carlson, Hirschorn, Ryaboy, & Zhao, 1996; Chin & Hughes, 1987; Coulter, 1996; Harry, 1992, 1994; Reschley, 1991; Reschley & Ward, 1991; Serwatka, Derring, & Grant, 1995), in particular as having SED or mild mental retardation (MMR). Many different definitions of the problem and various methods have been employed to investigate disproportionate representation, limiting the interpretation and generality of findings because of varying samples, unclear or questionable assumptions about disproportional ity, and inadequate technical methods. For example, many studies have used some form of a percentage to describe overrepresentation (Markowitz, 1996), but the use of percentage figures can be confusing. A district that includes 10% AfricanAmerican membership, in which 15% of the students with SED are African American, might be described as displaying 5% overrepresentation 0 5% - 10% = 5%) or 50% overrepresentation [(15% - 10%)/10% = .501.
An example from our current work illustrates the significance of the definition and the method used to calculate disproportionate representation. We are conducting analyses of the national database on ethnicity and disabilities that is collected by OCR. The sample consists of the districts selected for the Fall 1992 and 1994 Elementary and Secondary School Civil Rights Compliance Report survey, of which 4,692 responded. …