Reasons for the Misidentification of Special Needs among ELLs

Assessment

The first and most significant reason for the tendency to overidentify ELLs as in need of special education is that the assessment of proficiency and its manifestations among ELLs is fraught with difficulty. Assessment as a process is a complex enterprise that requires consideration of multiple factors, including symbolic proficiency, affect, previous experience, cultural and linguistic learning and application, expectations, and contextual variables (Carroll, 1993; Damico, 1992; Gutkin & Nemeth, 1997; Hamayan & Damico, 1991; Lubinski, 2000; Müller, 2003; Wentzel & Wigfield, 1998). However, this complexity and the methods to address it are not typically considered in school settings. Consequently, a number of biases that orient the assessment process away from the best interests of the ELL become operational. These biases include:

• a focus on superficial behaviors rather than underlying factors as indices of difficulty
• the collection of inadequate data — often in the form of norm-referenced and standardized test results — that do not enable sufficient descriptions of proficiency
• a lack of recognition of several consequences of bilingualism during assessment, and
• the application of inappropriate discrepancy formulas for interpretation purposes.

Behavioral manifestations: Intrinsic vs. extrinsic factors

At a superficial level, the way that academic and language difficulties manifest among ELLs is very similar to the way such difficulties manifest among students with long-term disabilities or special needs (Crago & Paradis, 2003; Cummins, 2000; Damico & Damico, 1993a; Salameh, Håkansson, & Nettelbladt, 2004). For example, Paradis (2005) examined English grammatical morphemes in 24 typically developing ELLs and compared their productions with those of monolingual English-speaking children exhibiting specific language impairment (SLI). She found that the ELLs' accuracy rates and error patterns were similar to those reported for same-age children with SLI. Similarly, both ELLs and monolingual English-speaking learning-disabled students may frequently search for words in English even though they have understood the concept. That is not to say that the causes of these difficulties are the same in the two populations. Rather, there are only so many ways that language and learning difficulties manifest behaviorally, despite the underlying causal variables (i.e., disability or difference). It is not enough to focus only on the superficial indices of difficulty; one must also determine whether deeper or more complex variables are operating (e.g., Ahlsén, 2005; Armstrong, 2005; Damico, Smith, & Augustine, 1996; Holm & Dodd, 1999; Perkins, 2005).

Language and learning disabilities are generally due to factors intrinsic to the learner, such as a neurological impairment or a problem with symbolic processing (Perkins, 2005), whereas second language learning difficulties are typically due to factors extrinsic to the learner, such as the language learning process itself or cross-cultural differences (Damico & Hamayan, 1992). In the case of vocabulary usage, for example, if an ELL frequently forgets a common word that has been taught, it is possible that the visual aid used to represent the concept may have been culturally irrelevant for that student (for example, the Liberty Bell representing the concept of freedom or independence that is specific to American history); hence the visual symbol would not provide any help for that student in learning new vocabulary. For learning-disabled students, the same difficulty — that is, forgetting common words that have been taught — may result from a completely different set of reasons. The student may have oral language comprehension or production difficulties as a consequence of word retrieval problems, or the student may have memory problems. In such cases, the pedagogical needs of the two populations are different: learning-disabled students need support in creating compensations to overcome their problems (e.g., Damico, Smith, & Augustine, 1996; Dunaway, 2004; Genesee, Paradis, & Crago, 2004; Westby, 1997), whereas second language learners need to develop further proficiency in academic language.

Data gathering

Another difficulty in assessment is that the data gathered as part of the referral and evaluation process are frequently inadequate. There is a tendency to rely too heavily on norm-referenced and standardized test scores, which give only a very narrow and potentially quite inaccurate view of what a student is able to do (Flynn, 2000; Gutiérrez-Clellen & Peña, 2001; Laing & Kamhi, 2003; Tetnowski & Franklin, 2003). As discussed by Damico (1991), norm-referenced and standardized tests typically:

• focus on superficial aspects of language structure
• have validity and authenticity concerns
• provide numbers that have a differentiating function rather than an interpretive function and
• focus on identifying students' weaknesses.

Consequently, these tests do not give us a good sense of what ELLs are able to do. Further, ELLs, who are often unfamiliar with the cultural context of test items, are likely to give the wrong answers not because they do not have the specific skill being assessed but because they do not understand the question. This constellation of factors promotes unreliable test results that give an inaccurate view of the student (Aspel, Willis, & Faust, 1998; Gunderson & Siegel, 2001; Müller, 2003; Portes, 1999).

Consequences of bilingualism

In addition, norm-referenced and standardized assessment tools typically do not account for the fact that the students being tested are bilingual. Instead, tests designed for monolingual students are developed or translated from one language to another, and different norms are collected. Even assuming that the tests developed are valid and reliable (a dangerous assumption), factors relating to bilingualism as a process are not always carefully considered (Chamberlain & Mederios-Landurand, 1991; Genesee, Paradis, & Crago, 2004; Grosjean, 1998; Oller & Eilers, 2002). For example, although many studies have demonstrated that ELLs are highly variable in their performance, based on individual, contextual, affective, and developmental factors as well as proficiency (e.g., Bialystok, 2001; Genesee, Paradis, & Crago, 2004; Goldstein, 2004; Thordardottir, Rothenberg, Rivard, et al., 2006), these variables are rarely considered in the scoring criteria or in the interpretive instructions for these tools (Damico, 1991, 1992; Lubinski, 2000; Wilson, Wilson, & Coleman, 2000).

Similarly, although numerous studies have indicated that an ELL's test performance should be interpreted using both languages via conceptual scoring — that is, scoring the meaning of a response regardless of the language in which it is produced (e.g., Bedore, Peña, García, et al., 2005; Marchman & Martinez-Sussman, 2002; Oller & Eilers, 2002; Pearson, Fernández, & Oller, 1993, 1995) — this is not typically done. Failure to take into account the complexity of the bilingual process during assessment further reduces the effectiveness of traditional assessment and diagnostic tools when focusing on ELLs.

The discrepancy model

Finally, the discrepancy model for identifying learning disabilities, which was initially useful as a gatekeeping mechanism, does not serve ELLs well, since it is normal for ELLs to have lower scores on verbal tasks than on tasks that do not require as much verbal processing (Campbell, 1996; Campbell & Kenny, 1999; Cummins, 1996; Kohnert, 2004). This discrepancy may provide an accurate description of monolingual students with a learning disability (Fletcher et al., 1998; Sternberg & Grigorenko, 2002). However, it also describes quite precisely a major characteristic of normally developing ELLs!

Perhaps because of these problems, the Individuals with Disabilities Education Act (IDEA) of 2004 has given districts permission to go beyond the discrepancy formula to embrace a model that assesses how students respond to different interventions and instructional activities. This moves the definition of learning disability out of the domain of a discrepancy between performance and potential and into the domain of response to intervention (RTI). The RTI model seeks to improve the learning environment for all students in the classroom by supporting both teachers and students and subsequently keeping track of students who resist these interventions. We discuss the potential advantages and disadvantages of using this model to identify ELLs with disabilities in Chapter 4.

Time and resources

To get a better sense of what an ELL is able to do and what he or she has difficulty with, a considerable amount of information needs to be gathered. In many instances, staff members do not have the time or resources to gather the qualitative information necessary to determine ELLs' needs. Unless they are trained in the process and provided with appropriate tools, gathering accurate and useful qualitative information is a difficulty in many schools. There is often a lack of know-how of how to use ethnographic approaches to collect data in such a way that the information obtained is meaningful and useful. It is important to adhere to certain standards in order to collect qualitative data that are authentic, and to have the conviction that these data are not inferior to data obtained through more standardized means.

Neither can staff always collect the amount of information needed for such a determination. At the end of the assessment process, we are often left with something akin to a huge jigsaw puzzle with several pieces missing. Further, these data are often gathered and interpreted by school personnel who are not adequately versed in ELL education issues, second language acquisition, or cross-cultural education and the implications of these factors for the assessment and instruction of ELLs (Goldstein, 2004; Müller, 2003; Oller & Eilers, 2002; Ortiz, et al., 1986; García & Ortiz, 1988). Often the bilingual or the English as a second language (ESL) specialist is not involved in making sense out of the gathered information, let alone in the data collection (Chronicle 1.1 in the manual describes an instance in which language proficiency testing was overlooked completely).

Prevalence of a Medical Model

A second fundamental reason for the tendency to overidentify ELLs as needing special education services lies in the way in which the special education system has evolved in the United States. In the last quarter-century, a medical model has prevailed in the way that many people attempt to understand the special needs that some students have in school. This model is based on the notion that disabilities or challenges can be identified much like the list of ailments that are officially recognized as diseases in the medical field (Coles, 1987; McDermott & Varenne, 1995; Skrtic, 1991; Taylor, 1991, 1993). This tendency to view academic and language learning difficulties from the perspective of human pathology can be seen in the medicalization of learning problems in the Diagnostic and Statistical Manual of Mental Disorders (American Psychiatric Association, 2000) and the tenth edition of the International Classification of Diseases (World Health Organization, 2002). Whether these medical designations are defensible or not, they do function to create an explicit grounding for the discipline of special education (Skrtic, 1991).

This prevailing medical orientation has as a result created a set of beliefs about the nature of special education and even about ELLs' role within special education, and these beliefs inform both our perceptions of academic and language-learning problems and our assessment practices (Damico, Müller, & Ball, 2004; Frick & Lahey, 1991; Klasen, 2000; McDermott, 1993; Taylor, 1993). This situation has falsely suggested to educators that disabilities can be easily identified in a valid and reliable manner, which results in misidentification, because the complexity of the phenomenon and the individuality of the ELL are often overlooked.

In reality, exceptionality is not an objective and easily verifiable empirical construct. Rather, it is a social construct, and the diagnostic criteria employed for various exceptional traits have been constructed on the basis of a number of sociocultural factors that mirror whatever ideologies are in vogue at any given time (e.g., McDermott & Varenne, 1995; Reid, Maag, & Vasa, 1994; Skrtic, 1991). These diagnostic categories may be further complicated by ambiguity and subjectivity, and this lack of definitional rigor may result in various types of problems. Indeed, the extreme overidentification of ELLs as learning disabled and language disordered is a clear demonstration of the problems (e.g., Jitendra & Rohena-Diaz, 1996; McNamara, 1998; Ortiz, García, Holtzman, et al., 1985; Stow & Dodd, 2005).

Availability of Categorical Funding

The third reason for the misidentification of ELLs as having disabilities lies in the categorical status of special education. The legal status of special education gives impetus to the tendency to formally identify separate categories of disabilities. When Public Law 94-142 (The Education Act for All Handicapped, also known as the IDEA) passed in 1975, it provided legal support for special education and subsequently created a separately funded category within the educational system. Since then, students have had to be formally identified as having special education needs before funding can be received by the school for that individual student. Thus, the identification of these disabilities is connected with additional funding that gives the school much-needed resources. Having additional funding attached to the categorization of students in special education may make it less objectionable for schools to identify special education needs. The gate or hurdle that would keep students from being needlessly categorized into special education is simply not there. It is too tempting to place students in special education when they are perceived to fall outside of what is considered "normal," regardless of the underlying reasons.

This categorization of special education could also have the opposite effect. In some districts, because of fear of monitoring agencies or potential litigation, the gatekeeping mechanisms are overused. The pre-referral process becomes so complex and time-consuming that students who do need support never get it (Stow & Dodd, 2005). Students have to fail or wait beyond an assimilation period before they are even referred to special education.

The Need to Provide Relevant Services in a Timely Manner

The accurate identification of the special needs of ELLs is important for two reasons. The first reason has to do with the stigma of the label (Cummins, 2000; Goffman, 1964; McDermott & Varenne, 1995). As much as we try to make it sound as though a special education placement is within acceptable limits, the label still carries a certain stigma that will remain with a student, formally and informally, for a long time (Coles, 1987; McDermott, 1993; Taylor, 1991). This stigma is especially problematic insofar as many in mainstream society perceive ELLs as having lower status; many in the general society do not value the languages and cultures that ELLs and their families bring with them. Immigrants and refugees are often marginalized and do not participate fully in what the larger society has to offer. So the added burden of yet another stigmatizing label is not helpful at all (Harry & Klingner, 2006).

The more compelling reason for not erroneously placing ELLs into special education is that we must ensure the most effective learning environment for these students. If an ELL is having difficulties in school because of second language issues, the best support for that student would come from expanding proficiency in English as a second language, and not from special education interventions (e.g., Damico & Hamayan, 1992; Freeman & Freeman, 2001; Genesee, Paradis, & Crago, 2004). Studies have shown that Hispanic students who were classified as learning disabled performed at a lower level after three years of special education placement (Wilkinson & Ortiz, 1986). Thus, these students may not benefit at all from the support they receive in that setting.

Another difficulty with the way special education services are provided to students in general is that it often takes a long time before any additional support is given to ELLs who are having special difficulties in school. Many school districts feel compelled to wait until a full-fledged referral has resulted in a formal categorization into special education before providing specialized support. Thus, the student remains without those specialized interventions for too long. This state of affairs is significantly influenced by some myths regarding ELLs in need of special education services.