In recent years, schools have made progress in meeting the needs of culturally and linguistically diverse students with special education needs. Even so, a number of longstanding issues still plague this field. Researchers and practitioners today express concern that culturally and linguistically diverse students are disproportionately represented in special education (Baca & Cervantes, 1998). Research offers some helpful insights into the inappropriate assessment and placement practices that lead to these discrepancies.
Our discussion here of some of the current challenges in special education for culturally and linguistically diverse students takes place within the larger context of increasing minority enrollment. In schools across the United States, the percentage of students of diverse cultures is growing. The U.S. Department of Education (2003) estimates that more than 3.5 million students in U.S. schools have limited English proficiency. In 2000, 42.6 percent of the school-age population in California and 32.4 percent in Texas spoke a home language other than English. The figure was more than 20 percent in Arizona, Florida, Nevada, New Jersey, New Mexico, New York, and Rhode Island, and more than 10 percent in 12 additional states (U.S. Census Bureau, 2003). Hispanics are the fastest-growing ethnic group in U.S. schools, having passed African Americans as the largest minority group in the United States (U.S. Census Bureau, 2003).
Many cultural minority groups—Hispanic students in particular—continue to underachieve at alarming rates. Hispanic students have higher dropout rates than non-Hispanics (National Center for Education Statistics, 2000). In 1998, only 63 percent of 18- to 24-year-old Hispanics had finished high school or earned a GED, compared with 85 percent of the total population (U.S. Department of Education, 2003). Although the achievement gap between Hispanics and whites narrowed in the 1970s and 1980s, it widened in the late 1980s and 1990s and remains large today (Lee, 2002).
Disproportionate Representation
The Challenge
Educators have been concerned for more than three decades about the overrepresentation of culturally and linguistically diverse students in certain special education categories (learning disabilities, mental retardation, and emotional disturbance) and their underrepresentation in programs for the gifted and talented (Artiles, Trent, & Palmer, in press). The phenomenon has recently come under closer scrutiny, however, with the publication of reports by the National Research Council (Donovan & Cross, 2002) and the Harvard Civil Rights Project (Losen & Orfield, 2002).
Although nationally, Hispanic students are only slightly overrepresented in the learning disabilities category and not at all in the mental retardation or emotional disturbance categories (Donovan & Cross, 2002), national data do not reflect the wide variability at the level of individual states and school districts. As Donovan and Cross explain,The nationally aggregated data have been interpreted to suggest no overrepresentation of either black or Hispanic students in LD [learning disabled]. But state-level data tell a more complex story. For black students, for example, the risk index ranges from 2.33 percent in Georgia to 12.19 percent in Delaware. For Hispanic students, the risk index ranges from 2.43 in Georgia to 8.93 in Delaware. Clearly, there is overrepresentation for these two minorities in the LD category in some states. (2002, p. 67)
This extensive variation in placement rates across states (Artiles & Trent, 2000) and within school districts (Losen & Orfield, 2002) occurs chiefly in the three “judgmental” categories of special education: learning disabilities, mental retardation, and emotional disturbance. These conditions are usually identified after a child starts school and are diagnosed by school personnel rather than by medical professionals. School clinicians typically exercise wide latitude in determining whether a student is eligible for services in one of these categories (Gottlieb, Alter, Gottlieb, & Wishner, 1994). In contrast, Hispanics and African Americans are not overrepresented in low-incidence, nonjudgmental disability categories, such as visual, auditory, or orthopedic impairment (Donovan & Cross, 2002).
Although state departments of education collect data about the ethnicity of students in special education, they typically do not accumulate information about students' language proficiency. Thus, we know little about the representation of English-language learners (students not fully proficient in English) in special education programs. Emerging evidence from urban districts in California, however, suggests that this population is overrepresented in high-incidence disability categories, and that those English-language learners classified as lacking proficiency in both their first language and in English are heavily overrepresented (Artiles, Rueda, Salazar, & Higareda, 2003).
Prospective Solutions
Professional groups and the U.S. federal government have given more attention and invested more resources in recent years to understand and address disproportionate representation. For example, the U.S. Department of Education recently funded a technical assistance center to address disproportionate representation: the National Center for Culturally Responsive Educational Systems (see www.nccrest.org). New tools and resources for practitioners have been published (for example, see National Association for Bilingual Education & ILIAD Project, 2002). In addition, recent amendments to the Individuals with Disabilities Education Act (IDEA) have strengthened nondiscriminatory mandates and regulations in the prevention, identification, assessment, and treatment of disabilities (Hehir, 2002).
The LD Definition and the Exclusionary Clause
The Challenge
IDEA's definition of learning disabilities contains an exclusionary clause stipulating that before a school determines that a student has a learning disability, it must ascertain that the student has had sufficient opportunity to learn—including adequate instruction in a language that the student can understand. Yet school personnel involved in placement decisions tend to disregard this exclusionary clause, paying insufficient attention to the influence of a student's native language, ignoring the ecology of the student's regular education classroom, and only sporadically implementing prereferral strategies (Klingner, Harry, Sturges, Artiles, & Wimes, 2003).
Ochoa, Rivera, and Powell (1997) surveyed 859 school psychologists who had conducted psychoeducational assessments of bilingual students. These psychologists indicated that they considered a variety of factors in their efforts to ascertain that students had had sufficient opportunity to learn, but frequently omitted consideration of the student's native language and the number of years of English instruction that the student had received—two crucial factors. Only 6 percent of psychologists reported asking about the student's home language, and only 1 percent attempted to determine whether a learning problem occurred in both English and the student's home language. The researchers called these findings “especially disheartening” (1997, p. 165). Disregard for the potential influence of language and culture on students' school performance can increase the incidence of false positive diagnoses, with devastating consequences for culturally and linguistically diverse students.
In an ethnographic study of the referral processes in 12 schools, we found that members of child study teams and Individualized Education Program (IEP) committees gave little weight to factors related to language acquisition when making decisions about special education eligibility and placement. Even for students who were not yet considered fully proficient in English and who had been tested bilingually, only English test results were typically included in psychologists' evaluation reports or discussed at IEP team meetings. The teams did not address the possible influence of second-language acquisition on the students' performance (Klingner et al., 2003).
We also found that the teams paid almost no attention to the ecology of the classrooms from which students were referred. Although many students were referred by teachers with weak instructional and classroom management skills, frequently neither the evaluating psychologist nor anyone else on the IEP team conducted classroom observations. From the start, team members seemed to assume that the student must have an internal deficit of some kind and that classroom observations were unnecessary. When asked whether she observed students whom she assessed in their classrooms, one psychologist said that she rarely did, noting,The whole morning with them [spent testing] gives you a good idea; you know it does. And you rely on the teacher's comments.
Some psychologists did indicate that they would like to conduct classroom observations but had insufficient time in their busy schedules. Without classroom observations, evaluation teams cannot know whether a student has had adequate opportunity to learn in an appropriate, culturally responsive environment.
In addition, we found that school personnel were inconsistent in their use of prereferral interventions designed to provide students with additional assistance before evaluating them for special education. Generally, this step in the referral process was not taken seriously by teachers, many of whom felt that these strategies were “just what we do anyway,” or “something checked on a form to meet referral requirements.” Similarly, in an investigation of the schooling characteristics of 46 Hispanic elementary students with limited English proficiency referred to or participating in bilingual special education in New York City, Carrasquillo and Rodriguez (1997) found that few prereferral interventions had been tried with students prior to their placement.
Prospective Solutions
We recommend a three-pronged approach for addressing this problem. First, schools should provide professional development to enhance professionals' understanding of the exclusionary clause, the factors that school personnel must consider in determining whether a student has received adequate opportunity to learn, and how to implement meaningful prereferral strategies.
Second, a professional with expertise in English-language acquisition should be present when IEP committees meet and when child study teams discuss prereferral interventions and eligibility evaluations (Klingner et al., 2003; Ortiz, 1997). This person should be knowledgeable about language and cultural issues and able to point out when language could be contributing to a student's difficulties.
Third, the child study team and IEP team members should consider classroom context when discussing students' behavior and learning (Harry, Klingner, Sturges, & Moore, 2002). To ensure that the student is receiving an adequate opportunity to learn, evaluators should conduct observations in the student's regular classroom and other settings. Someone other than the classroom teacher should complete these observations to ascertain whether the student's difficulties are being exacerbated by ineffective or culturally insensitive instruction.
Inappropriate Testing Practices
The Challenge
The assessment of culturally and linguistically diverse students is fraught with theoretical misunderstandings and flawed practices. Schools that use the same few tests with most students and fail to take language proficiency into account are setting up diverse students for assessment failure.
The test performance of culturally and linguistically diverse students may be affected by their differential interpretation of questions, lack of familiarity with vocabulary, limited English language proficiency, and issues of language dominance (Garcia & Pearson, 1994). Even students who have demonstrated English fluency on oral language measures may not be ready to demonstrate their achievement on tests at higher cognitive levels in English.
When is an English language learner ready to be tested only in English? We have not yet developed a test of language proficiency that can adequately answer this question (Ortiz, 1997). Even students who demonstrate English proficiency on language assessment measures typically demonstrate a low verbal IQ/high performance IQ profile (Figueroa, 1990).
If a school transitions a student prematurely from a bilingual or ESOL program to a regular classroom, the student's achievement and scores on tests of intelligence will likely suffer. English-language learners are often moved to English-only programs after one or two years, when they have attained basic interpersonal communication skills but have not yet developed the cognitive academic language proficiency required for demanding learning tasks, which often takes from four to seven years (Cummins, 1984; see MacSwan, 2000, for a critique of Cummins's distinction between cognitive and interpersonal skills).
Many standardized tests (for example, intelligence, oral language proficiency, and achievement tests) yield valid scores for most students but tend to underestimate the potential of culturally and linguistically diverse students (Abedi, 2002). This limitation was documented in a longitudinal study that compared students' grade point averages, standardized reading scores, and standardized math scores in 1982 with their 1972 scores on the Full-Scale WISC-R, an IQ test (Valdez & Figueroa, 1994). Students who achieved at higher levels than predicted by their IQ scores were considered to be “overachievers”; those who achieved at lower levels than predicted were considered to be “underachievers.” Hispanic students who in 1972 had scored at or below the mean on the WISC-R were more likely than their Anglo counterparts to show above-expected school grades and achievement, thus placing them in the “overachiever” category, especially if they lived in homes where Spanish was spoken. Valdez and Figueroa (1994) concluded that decisions based on IQ tests can lead to inaccuracies in decision making for Hispanic pupils.
Prospective Solutions
Moll (1990) recommends looking for the cultural, linguistic, and social resources—funds of knowledge—that students and their families bring to the school setting and viewing these as strengths on which to build problem-solving abilities. Harry and colleagues (2002) suggest acknowledging the arbitrariness of assessment and placement decisions, which would pave the way for more useful, less expensive, and less stigmatizing ways of helping students with learning and behavioral difficulties.
One prospective solution that the reauthorization of IDEA may promote is to stop using IQ tests and discrepancy formulas to determine who qualifies for special education services (Fletcher et al., 2002). Instead, students would be found eligible using a response to intervention model that provides students with increasingly intensive levels, or tiers, of support (Fuchs, Fuchs, & Speece, 2002; Gresham, 2002).
In three-tiered models, the first tier consists of high-quality instruction in a general education classroom. Students who do not reach expected curriculum benchmarks then receive intensive second-tier assistance, either through tutoring or in small groups, still as part of a general education support system. Students who continue to struggle are provided with a third tier of assistance, which many would consider to be special education. Linan-Thompson and colleagues (2003) are investigating the three-tiered model with English-language learners, with promising results.
This approach may solve some of the problems associated with the biased testing of culturally and linguistically diverse students, but only if we ensure that the interventions offered to students are culturally and linguistically appropriate. Like previous eligibility criteria, this model seems to assume that if a student does not make adequate progress, he or she must have an internal deficit. As with earlier identification criteria, this model must be based on students having received an adequate opportunity to learn.
Schools should also find alternative procedures for conducting linguistically and culturally sensitive assessments, such as portfolios (for example, Brown, Campione, Webber, & McGilly, 1992; Gonzalez, Brusca-Vega, & Yawkey, 1997; Rueda, 1997).
The Challenges Ahead
The special education referral process for culturally and linguistically diverse students remains a complex and challenging task. The field has not yet adequately determined how to distinguish between disabilities and normal second-language learning development, nor have we discovered how to assess students' true learning potential rather than knowledge acquired through previous home and school experiences.
The special education profession needs to overcome the widespread tendency to view culturally and linguistically diverse children from a deficit perspective. This shift holds the key to building a knowledge base and a professional culture that account for the role of culture in learning and for the complex historical contexts within which educators, students, and families live and labor.
