ELL Misidentification and Special Education Referral Processes

Teacher Understandings of Referral Processes for ELL and LD Students

There is a clear need to objectively distinguish the needs of students who may be indicated for special services intervention. Garcia and Ortiz (2004) state that academic failure distinctions can be generalized into three categories: Type I, where students are in inappropriate classrooms for their particular learning needs (such as ELLs); Type II, where students have learning and achievement issues that do not fall into the special education/learning disabled category and therefore must be served in the general education classroom; and Type III, where students clearly have a functionally severe limitation such as a major disorder that limits the learning process (Garcia & Ortiz, 2004). Failure to distinguish between the three types may result in a student being inappropriately referred for special education when in fact they should not be. The question arises as to why the referral process may be dysfunctional, allowing this type of failure to occur in needs-assessments of students.

Various issues in the assessment process have been identified in the research literature. There are issues related to professional practices, with the spectrum ranging from one end where teacher perceptions and attitudes influence the assessment process — including their prejudices as well as their lack of awareness and training in properly assessing ELL students — to the other end, which represents inappropriate and inefficient professional methods of assessment, implementation, and follow-through. Additional barriers relate to the structure of assessment testing, with much criticism leveled at the use of IQ and achievement testing in assessing students for learning disabilities. This methodology does not adequately reflect the individual learning styles of students, and so may inappropriately identify students as learning disabled when in fact their issues are not due to learning disabilities.

Teachers may fail to understand the referral process. Ortiz et al. (1985) found, in assessing data gathered as part of referral and pre-referral processes on English language deficient learners in LD classrooms, that education professionals involved in the data-gathering phase of the referral process may not understand the learning issues underlying students with limited English language proficiency (Ortiz et al., 1985).

Misidentification of ELL Students as Learning Disabled

Wagner et al. (2005) report that the need for different approaches to identifying ELL students who have learning disabilities is paramount and increasingly recognized in the field of education. They state that the existing models of assessment and intervention do not effectively address the need for proper identification of students at risk. The needs of ELL students within the education system point to a need for a larger conceptualization of what constitutes a learning disability, including the proper distinction between learning disabled and language proficiency issues (Wagner, Francis, & Morris, 2005).

Spinelli (2008) points to the fact that misidentification of ELL students as learning disabled leads to those students being unable to receive appropriate services, and hence those students face greater emotional, social, behavioral, and educational challenges as time progresses and their needs continue to go unmet. Spinelli holds that an informal assessment measure is a viable option in the evaluation process that can be adapted to culturally and linguistically diverse students, their learning styles, and their unique characteristics which underlie their challenges (Spinelli, 2008).

Guiberson (2009) finds that not only are ELL students misidentified as learning disabled, but the forms in which the misidentification manifests are varied. These forms include overrepresentation of minority students in special education compared to the whole school population, underrepresentation when those ELL students with disabilities do not get correctly identified for intervention services, and misidentification when the ELL student is categorized as having a disability different from the one they actually possess (Guiberson, 2009).

Sandberg and Reschly (2010) state that the growing achievement gap between ELL students and language majority students is increasing, and that traditional standardized testing fails to bridge this gap. The use of curriculum-based measurement may be a better standard for assessing the academic needs of ELL students and correctly identifying those with true learning disabilities (Sandberg & Reschly, 2010).

Vaughn et al. (2003) report that the identification of LD students has increased by upward of 200% since 1977, with explanations ranging from a likely outcome of the growing knowledge field, to LD as a field serving as a catch-all for the failures of general education to meet the needs of students of varying abilities. The study investigators find not only that the heterogeneity of identified students is quite wide, but also that many students are overrepresented (misidentified) or underrepresented (unidentified). One major problem is the use of IQ tests to identify students as learning disabled. Using standardized tests fails to accurately identify those students who have reading difficulties or whose first language is not English. More emphasis is needed on response-to-instruction models of assessment and intervention to replace ineffective normalized standards for identifying students at risk and properly placing them for early intervention (Vaughn, Linan-Thompson, & Hickman, 2003).

Abedi (2008) reports that administering assessment tests in English for non-native English speakers may not provide reliable information about the academic levels of ELL students. Additionally, linguistic issues in the wording of tests that are outside of knowledge content further confound the issue. Language variables not pertinent to the academic content diminish the strength of testing outcomes. Test results used for identification of students at risk for learning disability may not be reliable, especially for ELL students at the low end of the English language proficiency curve. ELL students risk being misidentified by tests that do not accurately measure their knowledge or reflect their current academic levels (Abedi, 2008).

Waitoller et al. (2009) reviewed research on overrepresentation between 1968 and 2006. The study was intended to identify the characteristics of overrepresentation studies and to elucidate how those studies framed the problem of overrepresentation. The investigators found that most of the literature came from journals on special education, with study numbers increasing over time, with particular emphasis after the year 2000. Most studies focused on learning disabilities and African Americans. The studies were typically framed in one of three ways: sociodemographic models focusing on individuals and social contexts; critical perspectives involving power issues revolving around race; and professional practices leading to both the creation and maintenance of overrepresentation. This finding indicates that while awareness is increasing, the spheres in which it is occurring may need redressing, especially to capture the cultural factor of misidentifying ELL students as learning disabled (Waitoller, Artiles, & Cheney, 2009).

Contreras (2006) found that Spanish-speaking Latino students in a South Texas school district were twice as likely to be receiving special education services as their English-speaking counterparts. Further analysis revealed that the overrepresentation of ELL students receiving special education was at 77%. This clearly reveals an inappropriate method in those school districts where ELL students were misidentified as learning disabled, likely due to language proficiency issues (Contreras, 2006).

Sullivan (2009) reports that there has been a disproportionate representation of linguistic minorities in the research on misidentification of students for special education services, with most studies focusing on racial minorities. Sullivan finds that ELL students are not only underrepresented in the literature, they are also overrepresented in actuality for special education services (Sullivan, 2009).

August et al. (2005) report that ELL students who display slow development of their vocabulary cannot comprehend text to the same degree as their English-speaking counterparts. Outcomes for these students are that they are likely to perform poorly on assessment tests and hence risk being misidentified as learning disabled. The investigators suggest using native language testing, basic word-meaning tests, and appropriate support and reinforcement in instruction to mitigate the language deficits of ELL students (August, Carlo, Dressler, & Snow, 2005).

ELL students are at particular risk for being misidentified as learning disabled in the typical English-speaking educational system. Cultural and linguistic differences in testing and intervention do not account for the actual academic knowledge of students, with static models of testing and intervention potentially exacerbating and prolonging the situation for ELL students in terms of their academic needs not being met.

Issues in the Pre-Referral Process

Educators and school personnel are the people involved in identifying children at risk and in possible need of intervention services. Teachers may be affected by existing processes for referring students for intervention services, as well as be unfavorably influenced by factors relating to the implementation of intervention programs in the typical classroom setting. The problem is not only limited to the referral process; it is also related to the implementation process, and this falls within the concerns regarding assessment. The need is to identify those areas of concern held by teachers that may affect both the referral process and the implementation process for intervention services. The issue becomes even more concerning when dealing with a child who is an English Language Learner (ELL) (Wood, 1998).

There is a strong need for collaboration between teachers who provide intervention services and the general education teacher. This is a significant problem area. Hardin et al. (2007) found that teaching staff felt they were inadequately trained to work with ELL students of diverse cultural and ethnic backgrounds (Hardin, Roach-Scott, & Peisner-Feinberg, 2007). If English is the primary language of the staff and there is little or no resource staff that speak the native language of the child in need of services, the problem is significantly compounded and may result in inappropriate placement. Placing ELL students in the special education classroom due to misidentification in the referral process actually causes further deterioration in ELL students' achievement in the long run (Buysse, Castro, & Peisner-Feinberg, 2010), and is therefore of primary concern in determining the underlying factors in the pre-referral and referral process.

In the pre-referral process, there are discrepancies between what the family believes is being communicated and what the interpreter is actually saying. This problem is likely due to the interpreter's lack of knowledge regarding not only the referral process but also their lack of training in early childhood education terminology (Hardin, Roach-Scott, & Peisner-Feinberg, 2007). The data methods used across cultures and languages also vary, making it difficult to obtain reliable data upon which to base referral assessments (Collier, 2007).

The federal government has addressed students whose first language is not English, defining them as Limited English Proficient (LEP) students in Title VII of the Improving America's Schools Act of 1994. While the government offered a definition of what qualifies as an LEP student, the burden falls on individual states to determine exactly how they will address students with ELL needs (Schon, Shaftel, & Markham, 2008). As such, the lack of an institutionalized methodology for data-gathering has contributed to the problem of properly assessing and placing students with ELL needs. Such data-gathering in the pre-referral and referral stages includes home-language surveys, teacher observations, language test assessments, informal assessments, role-playing, physical demonstrations, pictorial illustrations, student portfolio assessments, and other methods (Norris, 1998, p. 28). The diversity of assessments calls for a need for best practices in serving language minority students.

Best Practices Theory in Assessing Language Minority Students

The lack of clarity in objective measures for assessing ELL students' abilities and determining their needs reflects a failure in the U.S. education system to take into account not only the resources needed by educators to adequately address the issue, but also a failure to address the needs of the growing ELL population in the United States. There is a mainstream bias that exists in the social and educational realms of American culture, which is manifesting in the inability to serve the underserved in the nation (National Research Council Committee on Developing a Research Agenda on the Education of Limited English Proficient and Bilingual Students, 2001).

Adjusting for the Language Barrier

Various examinations have addressed the need for best practices in serving language minority students. Bainter and Tollefson (2003), in a survey of 195 school psychologists who assess language minority students, found that using native-language-based testing, using a bilingual school psychologist, and using tests in English when the student was dominant in English were always acceptable methods for testing the cognitive skills of ELL students. In contrast, administering non-verbal tests requiring oral instructions without the use of an interpreter, and using tests in English when the student was dominant in another language, were never acceptable (Bainter & Tollefson, 2003).

Using Cultural Intelligence Theory for the Classroom

Chang et al. (2007) found that there was a strong correlation between the social context of the classroom setting and peer-to-peer and student-to-teacher relationships. In classrooms where some Spanish speaking occurred and where educators spoke some Spanish, students developed better social skills and had a more favorable learning atmosphere (Chang et al., 2007). The implication is that policies regarding ELL students should always take into account the context in which the assessment is taking place.

The concept of cultural intelligence can apply to the field of education. Typically used in global business management to teach managers how to best address other cultures' needs for optimal business strategizing, cultural intelligence is not limited to the business realm. It is useful in understanding how best practices for ELL students can benefit from incorporating cultural intelligence into the sphere of policy making. Using cultural intelligence management psychology would benefit the professionals who are assessing ELL students, which would ultimately benefit the students (Peterson, 2004, pp. 166–172).

Cultural intelligence deals with how people relate to each other across cultures. Through understanding a person's cultural background and the particular culture's norms, ideologies, and ways of viewing the world, the interested party can become more successful in navigating that culture's worldview. Simply put, understanding the norms, language semiotics, and society of the culture an ELL student is native to will result in a more favorable educational outcome for the ELL student (Ang et al., 2007).

This is especially true in multicultural societies, where an understanding of various cultural aspects — religion, language, and social customs — can enable the educational professional to best formulate a plan for an ELL student's entry into the classroom (Earley & Ang, 2003). Cultural intelligence, also called cultural quotient (CQ), focuses on three main areas of development:

Cognitive means (understanding cultural diversity)
Physical means (using one's senses and body language to facilitate cross-cultural communication)
Motivational means (becoming empowered through successful outcomes)

People who score high in CQ are viewed as better able to merge and engage with other cultures than those who score low (Earley & Ang, 2003). Through using this three-pronged model, education professionals can explore opportunities via a cultural-awareness assessment approach. By utilizing CQ, they can have the proper tools to address the different cultural landscapes of ELL students in assessment, thereby preventing the misidentification of those students as learning disabled (DuBrin, 2008).

Multiple Intelligence Model for ELL Students

One way to approach policy making, the referral process, and assessments of students of diverse cultural and ethnic origins whose first language is not English is through the framework of the Theory of Multiple Intelligences. This theory holds that there is no single "right" way to measure how intelligent someone is, nor any one test that can accurately do so (Gardner, 1993). In the field of education, the traditional structure of teaching and measuring learning outcomes has been based on reading and writing. This can work well for many students; however, there will be those for whom logical intelligence (reading) and linguistic intelligence (writing) standards of teaching and testing do not work well. Gardner proposed an eight-degree model of intelligences (Gardner, 1983) that accounts for this diversity of cognitive strengths.

To better serve all students, a broader methodology for teaching and testing is required — one that understands that not all students are intelligent in the same way (Stanford, 2003). This model of multiple intelligences is viewed by many educators as a matter of course; educators simply adjust when the requirements of the student call for it. However, for ELL students who are at risk for being misidentified as learning disabled, the theory of multiple intelligences takes on a new significance.

Haley (2008) found that teachers using Multiple Intelligence (MI) teaching strategies for ELL students improved those students' success rates on assessment tests (Haley, 2008). Campbell and Campbell (1999) studied six schools that employed MI strategies with culturally and linguistically diverse student bodies. All six schools reported that using MI strategies in teaching was advantageous for the student population. A notable finding was that the influence of MI on teachers' attitudes toward teaching was a key factor in the academic success of the students (Campbell & Campbell, 1999).

Kim (2009) found that use of MI theory in ESL classrooms improved language learning skills (Kim, 2009). Nolen (2003) found that teachers who adjust their teaching methods to account for MI have better classroom outcomes, simply by being aware that students do not learn alike and that varying teaching methods reaches more students than a linear instructional model (Nolen, 2003).

A combination of adjusting for the language barrier — allowing assessment processes that take into account the primary language of the student being assessed — and employing teachers who are professionally trained in cultural intelligence theory and multiple intelligence strategy opens the door to creating assessment models that genuinely address student needs. Assessment models that bridge the gap between student needs and teacher understanding are a necessary step in preventing the misidentification of ELL students as learning disabled.

Teacher attitudes and perceptions strongly influence the educational setting and learning outcomes of all students. General education teachers faced with the possibility of being involved in an academic plan for a special education student may be affected by their lack of training in special education, lack of resources for accommodating students with special needs, and other beliefs about disability in general. When the question arises as to whether a student has a linguistic deficiency or a learning disability, those same factors may negatively influence the pre-referral and implementation outcome for those students (Buell, Hallam, Gamel-McCormick, & Scheer, 1999).

Not only is the general educator dealing with a potential learning disabled student, but they may also be facing the issue of having a student who does not speak English. The teacher is best supported through this process by identifying "differences" rather than "problems," to allow for proper assessment procedures (Buysse, Castro, & Peisner-Feinberg, 2010).