The disproportionate representation of specific groups of students in special education has been an area of concern since Dunn's (1968) seminal article over 40 years ago. Despite the longstanding acknowledgment of this problem, schools and districts across the United States continue to report the overrepresentation of African American and Native American students (e.g., Skiba et al., 2005, 2008; Sullivan & Bal, 2013) and boys (e.g. Bruce & Venkatesh, 2014; Coutinho & Oswald, 2005; Hibel, Farkas, & Morgan, 2010) in special education. Acknowledging the problems associated with disproportionality, reauthorizations of the Individuals with Disabilities Education Act (IDEA, 2004) have instituted policies that require schools to collect and report data on disproportionality in special education. Unfortunately, such policies have done little to remedy the problem to date, necessitating additional scholarship focused on potential contributing factors and solutions. Much of the existing research on disproportionality has focused on outcomes, rather than process factors. The present study sought to address disproportionate representation at the referral step of special education placement through the use of a data-driven process of identification, particularly for those who might qualify for services due to an emotional or behavioral difficulty.CURRENT DATA ON DISPROPORTIONALITYDisproportionality is defined as an overrepresentation or underrepresentation of a particular student group within a setting or outcome of interest, given that group's proportion in the total population. The overrepresentation of specific student groups in special education is problematic if the services provided by special education are not meeting the needs of, or are harming, those students (Algozzine, 2005; Bruce & Venkatesh, 2014). Special education services have been found to be most beneficial for students with milder academic-based disabilities served in minimally-restrictive environments (Fuchs & Fuchs, 1994). However, special education is largely perceived as ineffective for many students, with the consequences of being removed from general education outweighing the benefits of receiving additional services (Harry & Klingner, 2014; Hosp & Reschly, 2003; MacMillan & Reschly, 1998). The outcomes associated with special education placement indicate lower achievement across subject areas longitudinally (Wagner et al., 2006), with those qualified to receive special education services for emotional or behavioral problems often faring the worst academically (Bradley, Doolittle, & Bartolotta, 2008; Greenbaum et al., 1996; Wagner et al., 2005).Alarmingly, demographic characteristics such as race, gender, and primary home language are better predictors of special education placement than both academic performance and economic conditions (Bruce & Venkatesh, 2014; Hosp & Reschly, 2004; Skiba et al., 2005, 2008; Sullivan, 2011). When considering race as a predictor of placement, African American students are overrepresented in special education, particularly in the categories of intellectual disability (ID) and emotional disturbance (ED; Ahram, Fergus, & Noguera, 2011; Hosp & Reschly, 2003; Jasper & Bouck, 2013; MacMillan & Reschly, 1998; Skiba et al., 2005, 2008; Skiba, Simmons et al., 2006). In most studies, African American students are also more likely than their peers to be placed in more restrictive environments (Harry & Klingner, 2014; Serwatka, Deering, & Grant, 1995; Skiba et al., 2008), particularly for high incidence disabilities such as specific learning disabilities and mild emotional and behavioral disturbance (Albrect et al., 2012; Skiba, Poloni-Staudinger et al., 2006). Native American students are typically overrepresented in special education as well, particularly within the learning disabilities category (Harry & Kilngner, 2014; Skiba et al., 2008; Sullivan, 2011). …
Read full abstract