The disproportionate representation of culturally and linguistically diverse students in special education has been a persistent problem in education for more than four decades. The core issue concerns the possibility that some students may be misidentified, thereby receiving inappropriate educational services and being unnecessarily placed at-risk for the negative outcomes associated with disability labels, while others may fail to be identified for much needed services. The literature on disproportionality has been dominated by analyses of identification patterns for students identified as racial/ethnic minorities, particularly those who are Black and Native American, and, to a lesser extent, Latino/a, at both the national, state, and local levels. There has been considerably less attention to the continued disproportionate representation of linguistic minority students among those identified as disabled. Moreover, while there have been several studies investigating the predictors of disproportionality, few have included students identified as English language learners (ELLs). This study addresses these gaps in the literature by examining the extent and context of ELL disproportionality in special education in a state with a large population of students identified as ELLs. Utilizing local education agency (LEA) data obtained from the Arizona Department of Education, this study examines identification and placement patterns for the 1998-1999 to 2005-2006 academic years in order to understand the extent of disproportionality in special education and the high-incidence disability categories and in each of the educational environments in which students with disabilities are served. Additionally, the study examines how certain characteristics of LEAs predict these patterns. Results show that students identified as ELLs are overrepresented in special education overall and in the high-incidence categories of specific learning disability, mild mental retardation, and speech language impairment at the state-level and in many LEAs These students are more likely than students identified as White to be served in the least restrictive environments, and are increasingly less likely to be removed for the majority of the school day. Predictors of disproportionality varied by identification and placement categories. The implications for research and practice are addressed.