addressing the disproportionate representation of culturally and linguistically diverse students in special education through culturally responsive educational systems

In this article, we present a conceptual framework for addressing the disproportionate representation of culturally and linguistically diverse students in special education. The cornerstone of our approach to addressing disproportionate representation is through the creation of culturally responsive educational systems. Our goal is to assist practitioners, researchers, and policy makers in coalescing around culturally responsive, evidence-based interventions and strategic improvements in practice and policy to improve students’ educational opportunities in general education and reduce inappropriate referrals to and placement in special education. We envision this work as cutting across three interrelated domains: policies, practices, and people. Policies include those guidelines enacted at federal, state, district, and school levels that influence funding, resource allocation, accountability, and other key aspects of schooling. We use the notion of practice in two ways, in the instrumental sense of daily practices that all cultural beings engage in to navigate and survive their worlds, and also in a technical sense to describe the procedures and strategies devised for the purpose of maximizing students’ learning outcomes. People include all those in the broad educational system: administrators, teacher educators, teachers, community members, families, and the children whose opportunities we wish to improve.

indicator of disproportionate representation by the Office of Special Education Programs (OSEP).Donovan and Cross used White students' risk ratios as the denominators in calculating risk, as do we here.OSEP is now recommending using a risk ratio in the denominator that includes all racial or ethnic groups rather than Whites only.See Donovan and Cross for a detailed description of the different approaches to determining disproportionality as well as a discussion of the many challenges faced by those trying to follow placement patterns.41,677,158 11,564,606 9,804,643 2,517,754 647,581 66,211,74 2 62.95% 17.47% 14.81% 3.80% 0.98% Note: Disability categories are terms used by Donovan and Cross (2002).CI = composition index; RI = risk index; RR = risk ratio.Source: U.S. Department of Education, Office of Special Education Programs (2003).
Concern about disproportionate representation is focused on the "judgmental" categories of special education-those disabilities usually identified after the child starts school and by school personnel rather than a medical professional.To be eligible for services as LD, MR, or ED, a child must be found by school clinicians to qualify for one of these three disability categories, and school clinicians typically exercise wide latitude in deciding who "fits" (Gottlieb, Alter, Gottlieb, & Wishner, 1994).Children identified as having LD, MR, or ED usually do not exhibit obviously discernible features, yet are still considered to have internal deficits that affect their learning and/or behavior.It is noteworthy that overrepresentation does not exist in low-incidence disability categories (such as visual, auditory, or orthopedic impairment) (Donovan & Cross, 2002).
Over the years some researchers and practitioners have disputed the view that disproportionate representation is a problem, arguing that special education placement results in the provision of additional resources and support and should be considered a benefit.Yet, as Heller, Holtzman, and Messick (1982) contended in response to this assertion, if bias or inappropriate practice is found at any phase of the referral process that leads to special education placement, then disproportionality must be treated as problematic.There are numerous reasons for this.Students in special education may be denied access to the general education curriculum and, particularly if they have been placed inappropriately, may receive services that do not meet their needs.Furthermore, when disability labels stigmatize students as inferior, result in lowered expectations, potentially separate students from peers, and lead to poor educational and life outcomes, improper placement is of great concern (Patton, 1998).
In this article, we outline the theoretical assumptions and guiding principles that we believe should guide efforts to reduce the disproportionate representation of culturally and linguistically diverse students in special education.First, we describe our explicit theoretical understandings about the complex nature of disproportionate representation.Then we discuss our vision for addressing disproportionality and improving outcomes for all students through the creation of culturally responsive educational systems.Finally, we explain the domains of this work as they cut across policies, practices, and people.
We build on the findings and recommendations of the latest National Research Council's report (Donovan & Cross, 2002), the work of the Harvard Civil Rights Project (Losen & Orfield, 2002), and other key syntheses of the literature.Our approach to reducing disproportionality and improving educational outcomes for students focuses on how the different dimensions of disproportionality (i.e., child and structural factors) come together in practice.Though a few researchers (Garcia & Ortiz, 1988;Ortiz & Yates, 2001;Serna, Forness, & Nielsen, 1998) have urged the implementation of prereferral interventions as a way to reduce inappropriate referrals to special education, we hold that in general the field of special education has not adequately considered prevention and intervention strategies at the general education level as a viable means of addressing disproportionate representation.Moreover, compensatory educations programs that serve culturally and linguistically diverse students are often academically problematic.Many culturally and linguistically diverse students are isolated in schools that provide a compensatory education that is merely the regular curriculum "repeated, broken into meaningless segments, or 'dumbed down.'Because compensatory education programs are located in low-socioeconomic schools and are aimed at low-track students, the problems of a narrow, fragmented, measurement-driven curriculum that plague these schools also threaten the pedagogical utility of compensatory education" (Strickland & Ascher, 1996, p. 618).
Our goal is to assist practitioners, educational leaders, researchers, and policy makers to coalesce around culturally responsive, evidence-based interventions and strategic improvements in practice and policy to help close the achievement gap between culturally and linguistically diverse students and their peers.Our work is based on logic derived from existing literature (Donovan & Cross, 2002;Kilpatrick, Swafford, & Findell, 2001;Ortiz & Yates, 2001;Vaughn & Fuchs, 2003): If we can improve the instruction provided in general education classrooms and through general education support systems, then we will reduce the number of culturally and linguistically diverse students referred to and placed in special education programs.

Assumptions about the Nature of Disproportionate Representation
Disproportionate representation is a complex phenomenon, not explained simply nor understood easily.In summarizing the extant literature on disproportionate representation, we draw several conclusions (see Artiles, 2003;Artiles et al., 2004;Donovan & Cross, 2002;Losen & Orfield, 2002 for syntheses and emergent research on this topic).The following synopsis is not meant to be exhaustive, but rather represents what we consider to be the most salient issues to consider across a wide spectrum.
Intrinsic deficits?We reject the notion that culturally and linguistically diverse students are overrepresented in special education because they are more likely to have true disabilities.The bulk of the literature emphasizes child factors to explain and address the problem of disproportionate representation.We know, for example, that reading difficulties and behavior problems are the main reasons for special education referrals (Donovan & Cross, 2002).Child poverty and associated risk factors, such as low birth weight, exposure to alcohol during pregnancy, tobacco and drug use, malnourishment, and exposure to lead, are often described as causal factors in the development of language or cognitive deficits or maladaptive behaviors (Donovan & Cross).When framed in this way, the problem is reduced to a discussion of technical issues related to presumed intrinsic child deficits, with little attention to contextual, historical, or institutional issues (Artiles, 2003;Artiles, Osher, & Ortiz, 2003;Daniels, 1998;Patton, 1998).One consequence of this perspective is a growing literature base that overemphasizes student placement patterns in which deliberations about technical matters dominate (e.g., definitions of overrepresentation, accuracy of indicators to monitor the problem); preventive or intervention models to tackle this problem have been largely ignored.Also, it is important to note that poverty itself does not automatically result in low learning potential, as witnessed by a significant number of children and schools who "beat the odds" (Donovan & Cross, 2002;Taylor, Pearson, Clark, & Walpole, 2000).
Contextual issues.Emergent research evidence suggests that contextual factors play an important role in contributing to disproportionality (Artiles et al., 2004;Harry & Klingner, in press;Losen & Orfield, 2002).Some of the specific factors that shape disproportionality include the following: (a) decision-making processes by which eligibility for special education is determined; (b) placement in special education programs with uneven levels of restrictiveness; (c) administrative decisions related to hiring practices and resource allocation that result in disparities; (d) interactions among school location, disability, ethnicity, poverty, and density of culturally and linguistically diverse populations; (e) the lack of availability of alternative programs (e.g., early intervention, bilingual education, Title I); (f) the presence of subtle forms of bias at various stages of the referral process; (g) the uneven quality of instruction and management in general education classrooms; and (h) the effects of various discipline policies (e.g., suspensions).Donovan and Cross (2002) discuss the significance of classroom context in terms of teacher effectiveness, "[T]he same child can perform very differently depending on the level of teacher support, and aggressive behavior can be reversed or exacerbated by effective or ineffective classroom management.In practice, it can be quite difficult to distinguish internal child traits that require the ongoing support of special education from inadequate opportunity or contextual support for learning and behavior" (p.3).A national dilemma is that teachers' degrees, qualifications, and licensing or certification status in affluent communities are impressive and increasingly improving, while teachers in high-poverty schools are under-prepared and know too little about teaching culturally and linguistically diverse learners (Villegas & Lucas, 2002).While cultural diversity in the student population is increasing, the composition of the teaching or professional force is becoming less diverse (American Association of Colleges for Teacher Education, 1994;Snyder, 2002).During the 1980s, commissioned reports (e.g., A Nation at Risk, National Commission of Excellence in Education, 1983) and research findings (e.g., Kozol, 1991) focused the nation's attention on the failure of U.S. schools to improve the status of education for culturally and linguistically diverse children from low socioeconomic backgrounds.More recently, in their investigation of the disproportionate representation of culturally and linguistically diverse students in special education in one of the nation's largest school districts, Harry and Klingner (in press) noted that teachers in inner-city schools with predominantly Black populations had fewer qualifications and advanced degrees and were more likely to exhibit ineffective instructional and classroom management skills than teachers in other schools.
Power and hegemony in the education of culturally diverse students.A key premise of our conceptual framework is that power and hegemony play a significant role in the educational experiences of culturally and linguistically diverse students in U.S. schools.McLaren (1989) defines hegemony as the "maintenance of the domination not by sheer exercise of force but primarily through consensual social practices, social forms, and social structures produced in specific sites such as the church, the state, the school, the mass media, the political system, and the family" (p.173; emphasis in original).We draw from various theoretical frameworks (critical race theory, Latino/a critical theory, e.g., Ladson-Billings & Tate, 1995;Solorzano & Yosso, 2001) to understand the central role of power in human affairs and examine how presumed race-neutral structures in education actually reinforce racial borders and hierarchies (Marvin & Adams, 2002).As such, it is important to examine the impact of oppression and hegemony in culturally diverse students' opportunities to learn in general education (Tate, 1995) as well as in the special education referral and placement process.
In this vein, Patton (1998) asserted that basic assumptions about race, worldviews, beliefs, and epistemologies serve to perpetuate disproportionate representation.Mainstream educators generally interpret culturally diverse students' performance through white middle-class normative parameters of competence.Because culturally diverse students' performance does not always align with such parameters, it is often regarded as deficient.These hegemonic processes are further complicated by the fact that current educational reforms accept substantial inequality in practice as a baseline that actually serves to perpetuate the status quo (Gutierrez, Asato, Santos, & Gotanda, 2002).
It is a challenging endeavor to trace how hegemony works, particularly because it is generally invisible to members of cultural communities.An insidious feature of hegemony is that it requires that the "oppressed unknowingly participat[e] in their own oppression" (McLaren, 1989, p. 173).This consensual dimension of hegemony is an understudied notion in the special education field.There is an urgent need for research on this aspect of hegemony as it affects the disproportionate representation of culturally and linguistically diverse students in special education.At the same time, however, we must be reminded that we cannot assume an overly deterministic perspective when understanding the roles of power.Indeed, people resist the weight of hegemonic codes and terms of reference.Thus, future special education scholarship should also be concerned with documenting how culturally and linguistically diverse students and families resist and overcome hegemonic systems of oppression.
It is not enough to merely acknowledge that power and hegemony play a role in the lives of culturally and linguistically diverse students.Our goal is not only to make transparent the ideological barriers to optimal conditions of learning, but also to offer real solutions vis-à-vis technical support and advanced tool development with the goal of student advancement in mind.How this plays out is complex.At times the social practices of individuals in authority are built on deficit model assumptions, thus the vision of what might be technically possible in terms of improving opportunity to learn structures are not discussed or considered.Or those in power may view educational opportunity as a zero-sum game where investments in some groups are deemed essential and investments in other groups are thought to be superfluous.The point is that technical support and invention typically have not been forthcoming to students and communities viewed as outside the power structure.We assert that this must change.
Assumptions about intelligence.Categorical views of intelligence as a measurable construct affect the way teachers and schools think about students.Deeply held assumptions about inferior intelligence among students of color represent one of the most enduring legacies of Western racism.
Despite exposure of their fallacious nature by Gould (1981) and numerous other scholars, these beliefs have been institutionalized in the policies and practices of our public schools (Steele, Perry, & Hilliard, 2004).That the construct of eligibility for high incidence disabilities is tied to IQ measurement means that cultural and linguistic minorities continue to be more likely to be found deficient, since there is little doubt that these measures reflect the cultural, social, and linguistic knowledge of society's mainstream.The decontextualized IQ testing for the identification of high incidence disabilities that typifies assessment for special education is based on a narrow view of intelligence that fails to take into account the social and cultural nature of learning (e.g., Hilliard, 1994;Rogoff, 2003;Rowe, 1991, Samuda, 1998).Thus, as Hilliard (1995) has argued, what is needed is "either a paradigm shift or no mental measurement" (p. 6).The National Research Council (Donovan & Cross, 2002), in concluding its consideration of assessment issues, called for a focus on children's intervention needs rather than a search for intrinsic disability, and for an end to the requirement for IQ tests as a "primary criterion" for eligibility (p. 313).This report also emphasized that children's academic achievement falls along a continuum, the cut-off points for "disability" or "giftedness" are "artificial and variable" (p.26).We agree with these statements and recommendations.
Assumptions about behavior.The notion of a continuum of performance is equally central to an understanding of children's behavior in school settings.Many scholars have emphasized that personal and cultural norms are inextricable from decisions about which behaviors are acceptable, to whom, and under what circumstances (e.g., Cartledge, et al., 2002;Obiakor et al., 2002;Townsend, 2000).Norms regarding what behaviors are considered appropriate vary across cultures, and yet school personnel tend to judge students' actions through a narrow, white, mainstream lens.In terms of formal assessment of children's behaviors, even the application of well-designed rating scales cannot exclude subjectivity in judgment and, in many states, final judgments rely on projective testing, a set of procedures that have been the subject of much debate related to unreliability and subjectivity (Gresham, 1993;Motta, Little, & Tobin, 1993).The ambiguity of the process is exacerbated by historical racist beliefs and practices (Children's Defense Fund, 1975) reflected in a "punishment paradigm" (Maag, 2001), which includes zero tolerance policies, corporal punishment, suspension, and expulsion.These strategies target African American students at disproportionately high rates (McFadden, Marsh, Price, & Huang, 1992;Skiba, 2002) and contribute to their overrepresentation in disproportionately segregated programs for emotional disturbance (Massachusetts Advocacy Center, 1986).The combination of historical racism and extremely ambiguous definitions, policies, and practices places the most vulnerable students at increased risk of inappropriate labeling and isolation.These serious outcomes reveal the fallacy of applying special education's categorical mind set to what is essentially a continuum of human behavior.
Wait to fail model.The educational system works on the assumption that failure must be documented first to secure assistance for struggling learners (President's Commission on Special Education, 2002).Most students who show signs in kindergarten or first grade of falling behind are not provided with the early intervention in reading or behavior that might enable them to "catch up" with their peers.Also, currently no mechanisms are in place to guarantee that students will receive adequate opportunities to learn through exposure to state-of-the-art reading instruction or classroom management before they are identified as having a "within-child" problem (Donovan & Cross, 2002).
Research to practice gap.The availability of research-based intervention approaches in general and special education does not guarantee their adoption in professional practice.Indeed, a major challenge in addressing disproportionate representation is the widespread gap between what we know from research about "what works" and what actually gets implemented by teachers in practice (Gersten, Vaughn, Deshler, & Schiller, 1997;Malouf & Schiller, 1995).The NRC report recognizes that "between the articulation of what we know from research and best practice and a change in everyday practice lies a wide chasm" (p.382).In part, we believe this gap exists because research has not sufficiently addressed issues of language and culture and the varying contexts within which practice takes place (Artiles, Trent, & Kuan, 1997), or the role of the teacher as a knowledge generator as well as a knowledge user (Cochran- Smith & Lytle, 1999).We address these issues in more detail later in this paper.
To conclude, the genesis of disproportionate representation is located beyond the borders of special education and requires a solid understanding of the intersection of culture, learning, disability, and the socio-historical constitution of educational processes and outcomes.Two issues are associated with the persistence of culturally and linguistically diverse overrepresentation in special education, namely the issues related to understanding the complexity of this problem and also difficulties associated with the use of research knowledge to address it.Ultimately, what is needed is the transformation and improvement of educational systems in culturally responsive ways.

Addressing Disproportionate Representation through the Creation of Culturally Responsive Educational Systems
The cornerstone of our approach is the assumption that disproportionate representation should be addressed through the creation of culturally responsive educational systems.Instead of determining how to "fix" culturally and linguistically diverse students' "deficits," professionals' biases, or society as a whole, we aim to promote the creation of conditions, produce resources and tools, and support multiple stakeholders in the creation of educational systems that are responsive to cultural diversity.Our work draws from scholarship on culturally relevant and culturally responsive pedagogy (Gallego, Cole, & the Laboratory of Comparative Human Cognition, 2001;Gay, 2000;Hilliard, 1997a;Ladson-Billings, 1994, 1995;Nieto, 1999Nieto, , 2002Nieto, /2003;;Villegas, 1991).Through our theoretical assumptions and actions, we seek to advance the core values of care, respect, and responsibility across all levels of educational systems.

Culturally Responsive Educational Systems
Culturally responsive educational systems are grounded in the beliefs that all culturally and linguistically diverse students can excel in academic endeavors when their culture, language, heritage, and experiences are valued and used to facilitate their learning and development, and they are provided access to high quality teachers, programs, and resources (Gay, 2000;Nieto, 1999;Valenzuela, 1999).As Sonia Nieto (2002Nieto ( /2003) notes, it is not enough to help students celebrate their own and others' cultural traditions.We must ask tough questions about who is being taught by the best teachers.Who is taking advanced placement courses?Where and for what purposes are resources allocated?
Culturally responsive educational systems instill ethics of care, respect, and responsibility in the professionals who serve culturally and linguistically diverse students.These systems have a transformative goal in all their activities and nurture the creation of school cultures that are concerned with deliberative and participatory discourse practices (Gay, 2000).Culturally responsive educational systems create spaces for teacher reflection, inquiry, and mutual support around issues of cultural differences (Beauboeuf-Lafontant, 1999;Ladson-Billings, 1995;Villegas, 1991).Key questions that are consistently researched and debated include the following: (a) What are the explanations for the differential achievement of culturally and linguistically diverse students?(b) What are the conceptions of "self "and "others" that inform pedagogical practices?(c) How are social relations structured in the cultures of schools and classrooms?(d) What are the conceptions of knowledge that inform pedagogical, curricular, and assessment practices?(e) What are the consequences of the aforementioned assumptions for academic and social outcomes?
Culturally responsive educational systems benefit all children.When educators strive to develop the individual self-worth of each child, everyone gains.Also, exposure to a variety of experiences enriches lives by broadening perspectives and validating each person's uniqueness and sense of belonging to a larger whole (Nichols, Rupley, Webb-Johnson, & Tlusty, 2000).

Systemic Change: Building the Context for Culturally Responsive Practice
Even if teachers themselves are able to explore their own cultural boundaries and learn to reach out to their students and connect and engage them in learning, they often do so in spite of the systems that surround them (Miramontes, Nadeau, & Commins, 1997;Townsend & Patton, 2000).Individual excellence in culturally responsive teaching can only become collective tradition when the contexts in which teachers practice and learn are able to support, sustain and expect culturally responsive practice.While school leaders often seek teachers from culturally and linguistically diverse backgrounds, they do so without a deep understanding of what culturally responsive practices and systems could be and accomplish.To engage in substantive transformation of our current educational systems requires changes in fundamental assumptions, practices and relationships, both within school systems and between school systems and the outside world (Abrams & Gibbs, 2000;Conley, 1997;Elmore, 1996Elmore, , 2000;;Henig, Hula, Orr, & Pedscleaux, 1999).Christensen and Dorn (1997) emphasize that such efforts must begin with an examination of assumptions about social justice.One path to creating culturally responsive systems is by working at systemic reform (Townsend, 2002;Utley & Obiakor, 2000).
System characteristics are often so familiar to the people involved in them that they seem invisible.Systems have a life and dynamic of their own that resists change (Bateson, 1972).To think about culturally responsive educational systems requires looking at the processes, decision and communication paths that are used to make and sustain changed practice (Beyer, 1996).It should be no surprise to observe that the hierarchical systems that have been created in most public school districts and schools mirror our military and traditional business organizations.Policy decisions are made by a few individuals at the top of a pyramid of workers and are conveyed to the workers or practitioners who, in turn, convey them to families and students.Even in enlightened and reformed educational systems, where site-based decision-making prevails, the kinds of participatory communication and decision making that mark some cultures are rarely present (Bondy, Ross, Sindelar & Griffin, 1995).Thus, teachers who may engage in culturally responsive teaching practices receive rather than construct policy and practice around teacher development, assessment and evaluation.This mismatch between expected practice in the classroom and systems of administration and leadership in the school can create tension and signal the system's preference for conformity over diversity.This helps explain why classroom and school practices may be so intractable.Educational systems try to maintain equilibrium in order to sustain familiar, and therefore predictable, routines and practices.
Conversely, practitioners thrive and are better able to innovate, support student effort, and generate improved outcomes when their organizations support and encourage their cultural responsivity through systems of leadership that also meet standards of culturally responsive practice (Miramontes, Nadeau, & Commins, 1997).Organizational support for culturally responsive practice must, in turn, be supported by initial educator preparation and ongoing professional development opportunities that enable educators to acquire and build this capacity.Teachers and other school personnel are able to engage in sustained, thoughtful, continually improving and reflective practices if the school organization is able to provide a milieu or environment that supports professional practice (Beyer, 1996).Schools that organize themselves to create time for these rich conversations are able to sustain this kind of dialog over time (Ferguson, Kozleski & Smith, 2003).Further, their approaches to professional development create the contexts for continued growth and awareness of culturally responsive teaching.
The key to systemic reform is the coherence and alignment of activities across and within levels.As Miramontes, Nadeau, and Commins note (1997, p. 15), "Schools can make a positive and significant difference for students when educators account for the complex interaction of language, culture, and context, and decisions are made within a coherent theoretical framework."In this paper, we put forth such a framework.Because we are focused on systemic change, this work must cut across different interrelated domains to be successful.These we describe next.

Interrelated Domains in Addressing Disproportionality
We believe that to be fruitful, efforts to address disproportionality must cut across three interrelated domains: policies, practices, and people."Policies" include those guidelines enacted at federal, state, district, and school levels that influence funding, resource allocation, accountability, and other key aspects of schooling.We distinguish between intended and enacted policies.We also contend it is imperative to understand the ideological premises and histories (e.g., original goals, specific circumstances) leading to the ratification of the policies that guide educators' work (Cole, 1996).We use the notion of "practice" in two ways.First, we use it in the instrumental sense of daily practices that all cultural beings engage in to navigate and survive their worlds.Second, we use "practice" in a technical sense to describe the procedures, models, or strategies devised by educators and researchers for the purpose of maximizing learning outcomes.However, our framework also suggests that people use practices in complex institutional contexts in which policies play a crucial mediating role."People" include all those in the broad educational system, administrators, teacher educators, teachers, community members, families, and the children whose opportunities we wish to improve.We attempt to convey the complex interplay within and across these domains.Every level in the system potentially influences and is influenced by other levels through a complex blend of top down and bottom up systemic reform strategies (Fullan, 2001).

Culturally Responsive Policies
A policy can be culturally responsive only when policy makers truly take into account how its enactment affects all students.Such policies are proactive in their attempts to provide equitable opportunities to those students who historically have had the least access to high quality schooling.We describe selected federal and state, district, and school policies that affect the practices and people involved in educating culturally and linguistically diverse students.We acknowledge that there are tremendous barriers to actually carrying out these potentially controversial policy changes.

Federal and State Level Policies
Policies at the federal level are the most likely to be disconnected from the work of practitioners simply because they are the result of multiple experiences and multiple viewpoints sifted and distilled over time to meet competing political, social, and professional agendas.Further, federal policy makers in Washington, D.C. and practitioners in classrooms across the country operate in very different worlds.While the intent of law and regulation may be to level the playing field for specific groups or increase the access to goods and services for traditionally marginalized populations, the interpretation of law through regulation and action at the federal, state and local level may obfuscate or neutralize those goals.Hence, practitioners who carry out policy act on information that has been distilled through many layers of bureaucracy.What is intended, interpreted, required, and finally, enacted, is the product of reinvention (Ferguson, Kozleski & Smith, 2002).
There is great wisdom in carefully considering the role of the federal government in determining education policy.We suggest that federal and state policies and practices should be reexamined and revised to promote culturally responsive educational systems.Though there are several policies worthy of scrutiny, here we discuss just a few.
First, we support current efforts to amend legal requirements at federal and state levels concerning the determination of eligibility for special education.Though there are those who would prefer to retain intelligence quotients and discrepancy formulae for identification purposes, others question the validity and utility of such approaches.The newly reauthorized Individuals with Disabilities Education Act, now called the Individuals with Disabilities Education Improvement Act (IDEA 2004) allows states to discontinue the use of an IQ-Achievement discrepancy formula for identifying students with LD and permits the use of response to intervention (RTI) criteria as part of the special education identification process.This shift represents a dramatic shift in how disabilities are conceptualized.While we see the potential in such a transformation to reduce disproportionate representation, we urge caution and thorough consideration of issues related to cultural and linguistic diversity when making eligibility determinations as part of an RTI model.We discuss these issues further in a later section of this paper.
Second, we advise careful scrutiny of governmental policies and mandates related to school financing and the allocation of resources.Currently, educational resources are not equitably distributed across schools and districts (Darling-Hammond, 2000;Donovan & Cross, 2002;Parrish, Hikido, & Fowler, 1998).Culturally and linguistically diverse children living in high-poverty areas are more likely to attend schools that are inadequately funded and staffed.Yet more money allows districts to hire better-prepared teachers who use more effective instructional strategies, reduce class size, offer more college preparatory classes, and provide teachers and students with more resources.These all serve to increase students' opportunities to learn (Elliott, 1998), thereby potentially reducing excessive referrals to special education.
Third, we suggest reexamining accountability measures, including how high stakes test results are used to evaluate schools.Like others, we are concerned that high stakes testing is affecting culturally and linguistically diverse schools in disproportionately negative ways (Hilliard, 2000;Kohn, 2000;Smith & Fey, 2000;Townsend, 2002;Valencia & Villarreal, 2003).As an alternative, we support standards-based reforms that are culturally responsive and not premised on high stakes (Townsend, 2002).We agree with Hilliard (2000), who suggests placing greater focus on teachers who excel, in spite of barriers, and providing additional support for teachers and schools in need of assistance.
Finally, we suggest that all states should review their teacher certification/licensure requirements to make sure they include standards specific to teaching culturally and linguistically diverse students, and that they require evidence from teacher preparation programs indicating they are addressing diversity in significant ways.Miller, Strosnider, and Dooley (2000) investigated the state teacher licensure requirements regarding diversity among the 50 states and District of Columbia.They found that 67 percent of respondents required some level of diversity preparation, though specific requirements varied substantially from state to state.This percentage should be closer to 100%.

District Level Policies
School systems in which all students are successful create policies that are based on a thorough and timely analysis of data related to student learning, teacher quality, beliefs about teacher learning and change, the school contexts in which students are expected to learn and teachers teach, and changing community demographics (Cochran- Smith & Lytle, 1999;Elmore, 1996;Miramontes, Nadeau, & Commins, 1997;Spillane, 2002).Effective systems are built on several features: (a) partnerships with local community resources including museums, business, non-governmental nonprofits, mental health, police and social services, (b) alliances with families, (c) productive working relationships with teachers' unions, (d) collaboration with faith-based organizations, and (e) commitments to inform and involve the community through local media as well as the internet (Shanklin, Kozleski, et al., 2003).Having information and making informed choices is critical to school system success.Capital investment, technical assistance, transportation, professional development, the size and quality of the teaching force, salary compensation, maintenance, and infrastructure capacity can be allocated more effectively and reinvested more readily in a system that collects, manages and uses information streams well.How data that drive policy, regulation and practice are assessed and acted upon signals the values of the organization.In culturally responsive systems, questions about who benefits from current policy and practice and who is being marginalized or disadvantaged are essential to any discussion of resource allocation (Townsend & Patton, 2000).District administrators who reallocate resources based on need and equity and are more likely to bring about real change.
Inequities in the quality of leadership and instruction in inner-city schools exacerbate efforts to reduce the disproportionate placement of culturally and linguistically diverse students in special education.Many schools that serve students from racially, ethnically, economically and linguistically diverse neighborhoods have the least qualified teachers and administrators, inadequate physical resources, and are buffeted by violence in their communities (Ansell & McCabe, 2003;Darling-Hammond, 1995;Klingner, Harry, & Felton, 2003;Krei, 1998;Oakes, Franke, Quartz, & Rogers, 2002;Orfield, 2000;Schneider, 1985).Problems related to the recruitment and retention of highly qualified principals and teachers in inner-city schools must be addressed if disproportionality is to be addressed effectively (Harry & Klingner, in press;Orfield, 2000).School district administrators who examine these issues can reorganize their resources in an effort to reassign the most skilled teachers and administrators to these buildings.In addition, by examining which students are being identified for special services such as gifted and talented programs, special education, and Title 1, district personnel can assess the degree to which their general education system is well-prepared to build upon the assets children bring to school and support students' various needs.
Collaboration between district level special and general education administrators provides another path to reducing disproportionality.By forming partnerships with general education administrators, special education leaders can play a role in developing effective intervention models designed to reduce inappropriate referrals to special education (for an example, see Klingner, Harry, & Felton, 2003).Recent recommendations at the federal level include the permissive use of a portion of IDEA funds to support early interventions (i.e., pre-referral) within general education.By leveraging these resources, special education administrators can ensure they are part of conversations that previously were the purview of general educators only.
Community organizations have the potential to provide additional resources for improving educational outcomes for all students.School systems can create expectations and on-going processes that enlist and sustain productive working relationships with local community leaders.Creating a policy environment that invites and enlists this kind of community ownership for the outcomes of public education can create the milieu in which community members and organizations from multiple perspectives and experiences can come together to educate each other as well as work on improving schools (Abrams & Gibbs, 2000;Keith, 1999).
Finally, many school systems are fortunate to have local universities and teacher preparation institutions in their communities (Abdal-Haqq, 1998;Levine, 2002).Strong partnerships for teacher preparation, professional development, research and inquiry and exemplary practice can be established (Murrell, 2000).Partner schools have the potential to ground the next generations of practitioners in culturally responsive teaching and learning and, in doing so, transform a one size fits all approach to teaching and learning to a diverse and personalized program for each child.However, for these partnerships to be successful, higher education must itself become more culturally responsive and implement powerful teacher education programs that promote models of culturally responsive practice that are committed to goals of social justice (Cochran- Smith, 2004).

School-Level Policies
All of our collective hopes and dreams about a quality education for every child come to fruition or wither over time in the corridors and classrooms of our nation's schools.While federal, state, and district policies and regulations greatly influence what happens in schools, it is also evident that in the same community, only neighborhoods apart, one school can take a group of children and help them excel while another fails year after year to accomplish this goal.The most important aspect of renewing our nation's schools is extending our understanding of schools in which students "beat the odds."Beacons of excellence in inner cities exist throughout the country.In fact, in most of the 100 largest systems in this country, researchers have found schools that work well for disenfranchised students living in poverty, many of whom speak languages other than English (Charles A. Dana Center, 1999;Jerald, 2001).Much can be learned by studying these highly successful schools (Scheurich, 1998).
Numerous key policies are established at the school level.Other policies are officially set at the district level, yet vary substantially in how they are implemented in schools (Harry & Klingner, in press).As the leader at the school-site, principals exert tremendous influence over hiring practices, the assignment of teachers to classes, whether students are grouped by ability level or heterogeneously across classrooms, discipline policies, student retention policies, class size and scheduling decisions, whether paraprofessionals are hired and how they are utilized, visitor policies, the extent to which interruptions to instructional time are allowed, whether students are permitted to take school books and other materials home, how resources are allocated, and curricular decisions.In their ethnographic investigation of 12 schools in one district, Harry and Klingner (in press) noted a great deal of variability across sites in each of these policies.For example, in one inner-city school 102 students out of a total school population of 603 were given out-of-school suspensions in one year, whereas in a nearby school with similar demographics 20 students out of 780 were suspended in the same time period.
Culturally responsive school-level policies take into account how decisions affect all students, even those who are typically marginalized.Effective school leaders are able to see the "big picture" and make sure the conglomeration of different programs and policies they enact make sense when implemented simultaneously (Bryk, Sebring, Kerbow, Rollow, & Easton, 1998;Fullan, 2002a).They give a high priority to quality instructional time in class without interruptions.Teachers are assigned to classrooms in equitable ways that ensure that all students have access to the most effective teachers.Teachers are provided with the support they need to succeed.Culturally responsive leaders consider alternatives to suspension and put into practice discipline policies that are proactive, such as in-house counseling support for anger management and other emotional/behavioral needs, positive reinforcement systems and behavioral supports (discussed in depth later), and increased relationship building with students and their families (Townsend, 2000).They implement alternative programs (other than special education) that provide students with intensive, early assistance within general education, such as by using Title I funds.

Culturally Responsive Practices
For each of the policies we describe there is a corresponding set of practices or actions that must be carried out to bring the policies to fruition.In our discussion of these practices we focus on ways that teacher education, professional development, literacy instruction, positive behavior supports, and early intervention can support systemic reform efforts focused on reducing unnecessary referrals to special education.We do not intend this to be an exhaustive list but rather wish to exemplify a few key practices that are currently under national scrutiny and promoted in IDEA 2005.

Teacher Education: Rethinking the Context
Teacher development is a process of growth over time.Sanders and Rivers (1996) assert that teachers' ability to impact student achievement occurs over years of practice and that their influence on student learning begins a strong growth trajectory in about the fifth year of their teaching career and continues to improve until about the 12 th year of teaching.This construct of continuing teacher growth over an extended career timeline suggests that the initiation of teachers into the practice community through preparation, induction, and ongoing mentoring and coaching requires significant investment of resources by both universities and school systems.Supporting teacher learning over time is further complicated by the need to engage teachers in a critical dialogue about issues of social justice and the importance of paying conscious attention to culturally responsive dispositions and practices (Cochran- Smith, 2004;McLaren & Fischman, 1998;Zeichner & Miller, 1997).
School systems and university preparation programs are becoming more adept at creating practice arenas for teacher candidates and teachers through teaching schools or professional development schools (Kozleski, Sobel & Taylor, in press).The Holmes Group and the National Network of Educational Renewal have been instrumental in developing this approach to teacher preparation and renewal (Wise & Leibbrand, 2000).The American Association of Colleges of Teacher Education has published a set of monographs highlighting the strengths of these newer conceptions of teacher learning (Kozleski, Pugach, Bellamy, & Yinger, 2002).Emerging data suggest that teachers who are prepared in urban schools are more likely to develop culturally responsive practices and continue to work in schools and classrooms with students who are culturally and linguistically diverse (Ladson-Billings, 2001).Similarly, Grossman, Smagorinsky, and Valencia (1999) use the construct of activity theory to focus on the contexts and settings in which teachers learn their craft and sustain their own growth over an extended career.Where teacher candidates, teachers and university faculty practice together, they develop a set of tools for informing their discourse, their actions, and their reflections that help nourish improvement over time (Chandler, Kozleski, et al., in press).The preparation and development of culturally responsive teachers requires attention to the context in which this preparation occurs, the nature of the skills and dispositions that characterize culturally responsive teachers, and the elements of multicultural education (Ford, 1992;Kea & Utley, 1998;Obiakor, 2001).It also requires that teachers develop knowledge of specific instructional practices and how to implement these practices in culturally responsive ways that enhance students' opportunities to learn and reduce the likelihood they will underachieve and be referred to special education.
Another issue to consider in the context of teacher education is the recruitment and selection of prospective teachers.Haberman (1995;Haberman & Post, 1998) noted that there are certain attributes that predispose individuals to be more likely to be effective in high-poverty, culturally and linguistically diverse schools and more likely to stay on the job.Older, non-traditional students with more life experiences tend to be better suited for such settings than younger, lessexperienced candidates.

Professional Development that Promotes Culturally Responsive Teaching
The ultimate test of sound intervention research is its application in the contexts where practitioners and students work and learn.We advocate for on-going professional development that provides teachers with the support they need to implement new practices and to advance reform efforts (Hargreaves, Earl, Moore, & Manning, 2001;Hilliard, 1997b;Quartz, 2003).We envision this professional development as two-pronged, on the one hand focused on helping teachers learn how to implement instructional practices that have been validated with students similar to those they are teaching, and on the other hand designed to help them develop the attributes of culturally responsive teachers.
This second prong is one that is relatively new and untested.We suggest that professional development programs should include experiences intended to help teachers understand the central role of culture in learning and think deeply about their views of culture.This development of selfawareness is an important part of the process of becoming multicultural.On-going discussions with others about diversity and what it means to be culturally responsive should also be key aspects of professional development programs (Gay, 2000;Ladson-Billings, 2001).
Our approach to professional development relies on the growing literature on teacher change (Borko & Putnam, 2000).Practitioners work in complex social milieus and thus, the application of research knowledge requires that they change what they think and do and transform the contexts in which they work.Teachers are faced with many challenges, including the need to reconcile differences between long-term goals and short-term needs and to balance their experiential knowledge with new research knowledge (Artiles, Barreto, Peña, & McClafferty, 1998;Lieberman, 2000).Beliefs, feelings of self-efficacy, attitudes, and perceptions all affect the extent to which teachers try new strategies and persist in using them even when confronted with challenges (Artiles, 1996;Sparks, 1988).Change can be difficult even in the best of circumstances, and for teachers implementing new instructional practices or trying to teach in novel ways, it can be daunting.Hargreaves, Earl, Moore, and Manning (2001) highlight the intense emotional demands that school change imposes on teachers.A supportive community of teachers and researchers can provide needed assistance while teachers make the shift towards improved practice (Klingner, Arguelles, Hughes, & Vaughn, 2001;Pressley & El Dinary, 1997).Within this community, it is important for diverse perspectives to be respected (Grossman, Wineburg, & Woolworth, 2001) and for teachers' needs to be validated (Silin & Schwartz, 2003).
The two-pronged professional development programs we endorse will be most successful if they are part of a systemic change model that includes the following: (a) the program promotes practices aligned with the district's curriculum and standards; (b) the program provides student outcome data showing the practices are effective with culturally and linguistically diverse students similar to those they teach; (c) school administrators facilitate implementation; (d) teachers receive long-term support (including demonstrations and coaching); (e) teachers take ownership of the practices, adapting them to fit their students' needs and the local context; and (f) teachers take responsibility for mentoring their peers as part of a community of practice (Klingner, 2004).Teachers need to see concrete examples of how a new theory relates to their students and their circumstances.If teachers do not see the relevance of the new approach to their situation, little change is likely to occur (Englert & Tarrant, 1995).
Induction programs offer a promising approach for mentoring beginning teachers in culturally and linguistically diverse settings and helping them teach in culturally responsive ways (Haberman & Post, 1998;Murrell, 2000;Quartz, 2003).Such programs not only lead to enhanced effectiveness, but also have the potential to improve teacher retention (Quartz, 2003).A key component of successful induction programs is the partnering of novices with true master teachers (Hilliard, 1997b).

Culturally Responsive Evidence-Based Instructional Practices
A core principle of our conceptual framework is that disproportionate representation will be reduced if more students receive validated culturally responsive instructional practices.Thus, we agree with those who emphasize that instructional practices should be based on scientific evidence about "what works."However, we would add that it is essential to find out specifically "what works" with whom, in what contexts, and under what circumstances.We are concerned with issues of population validity and ecological validity.Was the practice validated with culturally and linguistically diverse students similar to those for whom the practice is being considered?Were other aspects of the environment similar (Bracht & Glass, 1968)?We value findings from carefully designed experimental, quasi-experimental, mixed methods, and single subject research studies, but also emphasize that much can and should be learned through qualitative means.Qualitative approaches are ideally suited to answering questions about "how" and helping us to understand essential contextual variables that impact the effectiveness of an approach.A well-known example of this is the Kamehameha Elementary Education Program (KEEP) in Hawaii (Au, 1980;Au & Mason, 1981).KEEP was a research-and-development program that spanned more than a decade.Lessons were observed over several years while school personnel implemented different reading programs that had been found to be effective in other settings.It was not until the discourse of reading lessons was allowed to become more like the style of day-to-day Hawaiian conversation (in other words, more culturally responsive) that reading achievement improved dramatically.
The issue of population validity is an important one.We advise caution when interpreting research findings when applied to culturally and linguistically diverse students, particularly when diverse students have not been included in participant samples, or demographic and other relevant information has been under-reported (Artiles, Trent, & Kuan, 1997;Donovan & Cross, 2002;Gersten, Baker, Pugach, Scanlon, & Chard, 2001).We emphasize the importance of research reports' and articles' including sufficient information about the language proficiency, ethnicity, socio-economic level, immigration status, and other characteristics of participants (Bos & Fletcher, 1997;Keogh, Gallimore, & Weisner, 1997), as well as more information about environmental variables (e.g., school and community environment and history).
We believe researchers and educators must continue to ask tough questions about whether we are doing all we can to improve outcomes for culturally and linguistically diverse students who seem to be "left behind."If Treatment A is found to be better than Treatment B (or nothing), we must not assume that Treatment A is the best we can do.What if we were to adapt Treatment A to be culturally responsive to a particular group of students and then compare Culturally Responsive Treatment A with Traditional Treatment A? How do we know when increased outcomes with a given intervention are "good enough?"We must continue to ask if we are truly providing an optimal learning environment for all students.If we believe that African American, Hispanic, and American Indian students really cannot achieve as well as their white mainstream peers, we will settle for less than satisfactory performance (Valencia, 1997).Similarly, if we expect children from low-income households to perform lower than middle-class children, we can use this as justification to withhold resources (Brantlinger, 2003).Yet if we believe that culturally and linguistically diverse students truly can achieve at least as well as their white counterparts when provided with an appropriate education, we will not give up until we have facilitated this level of learning.Only by so doing can we reduce the disproportionate representation of culturally and linguistically diverse students in special education.
Although theses issues apply across subject areas and disciplines, in this article we zero in on how they pertain to behavior supports and literacy instruction.We focus on these areas because reading difficulties and behavior problems are the main reasons for special education referrals (Donovan & Cross, 2002).Also, they are the areas targeted in IDEA 2005.
Culturally responsive positive behavior supports.Positive behavior supports (PBS) has a rich, empirically derived database, steeped in applied behavior analysis and more recently in the application of functional behavior analysis to solve school-related behavioral issues (Utley, Kozleski, Smith & Draper, 2002).Sugai and colleagues (2000) define PBS as "a general term that refers to the application of positive behavioral interventions and systems to achieve socially important behavior change" (p.133).All PBS models are based on three fundamental principles: (a) behavior is affected by internal and external factors; (b) behavior is constantly shaped by unintended and intended responses in context; and (c) all behavior is learned and therefore can be changed (Sugai et. al., 2000).By examining behavior in context, it is possible to plan intentionally to encourage certain kinds of behaviors and to create disincentives for other, less desirable kinds of behaviors.According to Sugai et al. (2000), PBS takes the principles of applied behavior analysis and sets them within an arena that considers practicality, social values, and the interaction of variables within systems.In a school context, this means understanding the nature of student behavior, individually and in groups, the social contexts in which behaviors occur, and practical responses to shaping desirable behaviors.By studying the behavior of students across many schools, researchers have found that 80 to 90% of students in schools fall into a typical range of social behaviors (Sugai et al., 2000).Another 5 to 15% of students have more serious behavioral issues that can be successfully addressed in specific, group interventions.And a much smaller group of students (from 1 to 7%) of a school's population require ongoing, individualized support and intervention to improve their behavior.
It is in the latter situation that various communities criticize behavioral approaches, in spite of protestations by PBS researchers that their methodology incorporates considerations of social contexts and cultural expectations.At the heart of concerns are questions about who makes decisions about what is appropriate and inappropriate.In schools, where the vast majority of teachers are white (92%) and female (87%) (Snyder, 2002), and the students who are most likely to be sent to the office, suspended and expelled, are boys of African American descent, this criticism takes on heightened import (Osher, Woodruff, & Sims, 2000).Decisions to refer students for disciplinary reasons rely on the observations of teachers who themselves may not have the social and cultural contexts to translate and respond to student behavior in their classrooms.Sheldon and Epstein (2002) examined school officials' efforts to implement family and community involvement activities to reduce the number of disciplinary actions and ensure school climates focused on learning.They found that the more family and community involvement activities were implemented, the fewer students were disciplined.
Despite these limitations, we see potential in the approach espoused by the proponents of PBS.First, they acknowledge the importance of context.Second, they recognize the different levels of systematic intervention required to support and develop pro-social behaviors.Third, they offer strong and effective strategies for behavior change.To learn from critics of PBS, it might be instructive to think about the constructs inherent in multicultural communities.
From a multicultural perspective, the process of identifying students to target for intervention itself is problematic because identification of problematic behavior focuses on descriptions of difference.Appropriately, multicultural proponents ask the question, "Different from what?" Difference can hardly be conceptualized in the contemporary world as existing on a unitary dimension of normalcy and deviancy since that bimodal perspective assumes homogeneity and monolithic cultural norms rather than heterogeneous and indigenous norms.By naming and focusing on what is different, the process runs the risk of marginalizing some individuals.The rub of "otherness" gives rise to other, sociological, or anthropological phenomena: resistance, abandonment, and exclusion.These phenomena play out in the high numbers of students of African American and Hispanic descent who drop out of school.Therefore, identifying problems cannot be a unitary act by teachers or groups of professionals.Problem identification must be led not only by a multi-disciplinary professional team but by a multidisciplinary professional and community team.
To improve school success for every student, issues that are not typically considered as part of behavioral education must be addressed by general and special educators.Researchers and practitioners must work with families to examine issues related to classroom discipline, cultural diversity, and culturally responsive teaching to develop successful approaches for teaching pro-social skills and reducing antisocial behavior (Townsend, 1994).Teachers should spend time observing their students and the ways they interact while simultaneously engaging in self-reflection about their reactions and beliefs.When students do not behave in expected ways, it is important that teachers not jump to conclusions about what this means.For example, Cynthia Ballenger (1998) shared that when her Haitian preschoolers did not respond as she thought they would to her classroom management strategies, she examined her own assumptions about teaching and behavior and worked with Haitian teachers to learn more about their approaches.She was thereby able to adapt her style of discourse to better support her students in the ways they were accustomed.We consider this to be a culturally responsive approach to behavior management.Sugai and colleagues (2000) note that: The use of culturally appropriate interventions also is emphasized in the PBS approach.Culturally appropriate describes interventions that consider the unique and individualized learning histories (social, community, historical, familial, racial, gender, etc.) of all individuals (children with problem behaviors, families, teachers, community agents, etc.) who participate in the PBS process and approach.Data-based problem solving and individualized planning processes can help to establish culturally appropriate interventions.(p.134) For these reasons, we take the stance that school-wide PBS interventions should be proactive and promote a positive, culturally responsive climate that is conducive to learning by all.Teachers, administrators, and support staff should understand that perceptions of behavioral appropriateness are influenced by cultural expectations, that what is perceived as inappropriate varies across cultures, and that behaviors occur within larger socio-cultural contexts; connect with their students in ways that convey respect and caring; explicitly teach rules and expected behaviors within a culture of care; provide a continuum of support; and involve families and the community in positive, mutually supportive ways.
Culturally responsive literacy instruction.All children should receive culturally responsive literacy instruction that builds on their prior knowledge, interests, motivation, and home language, and emphasizes cultural relevance (August & Hakuta, 1998;Au, 2000;Rueda, MacGillivray, Monzó, 2001).We support a balanced approach to literacy instruction that promotes authentic literacy experiences in a supportive learning environment while providing the high level of explicit instruction needed for students to gain important skills and strategies (Delpit, 1995).This instruction should include frequent opportunities to practice reading with a variety of rich materials in meaningful contexts (Pressley, 2001;Pressley, Allington, Wharton-McDonald, Block, & Morrow 2001).It should also include explicit instruction in phonological awareness, the alphabetic code, fluency, and vocabulary development (National Reading Panel, 2000;Snow, Burns, & Griffin, 1998) as well as comprehension strategies (Snow, 2002).We believe that a focus on the complete literacy event does not mean that traditional skills are unimportant."Rather these skills are situated within a holistic context that is intimately linked with goals and conditions of reading" (Roller, 1996, p. 34).If students do not receive such instruction, how can we be ensured they have in fact received an adequate opportunity to learn and are appropriate candidates for a special education referral?
Literacy instruction should take into account the sociocultural contexts within which students learn (Artiles, 2002;Ruiz, 1998).A fundamental assumption of our approach is that culture matters-we believe that disproportionate representation is due in part to the inadequate attention to culture by researchers and practitioners.Culture is not a unitary construct but rather is complex and dynamic.In any given classroom, there are multiple cultures "as embodied in the cultural toolkit that each person brings to school and the cultures that are created as students, teachers, and school staff interact over time" (Artiles, 2002, p. 696).
But what does it mean to account for culture when teaching children to read?First, it means taking a broad view of what counts as literate in a multiethnic, diverse society.It means understanding the complex sociocultural, institutional, and political contexts that influence students' acquisition of literate behaviors (Artiles, 2003).It means recognizing that when children begin school, they may not have experienced all the same interactions with print as their mainstream peers, but they still have had valuable experiences that teachers can and should build upon.It means explicitly connecting home, community, and school literacy practices.It means recognizing that students' discourse and behavioral styles may not match school-expected ways, but they are still to be validated (Brice Heath, 1983;Cazden, 1988).It means recognizing that bilingualism is an asset and that learning English should be an additive rather than a subtractive process (August & Hakuta, 1997).Finally, it means making sure students are motivated and engaged in reading activities (Guthrie & Wigfield, 2000;Rueda, MacGillivray, Monzó, 2001).Although teachers need not be "insiders" in a particular culture to offer culturally responsive instruction, they should make an effort to learn about the cultures represented in their classrooms, respect students' values, and view differences in students' literacies as strengths, not deficits (Alvermann, 2003;Ladson-Billings, 1994).
Culturally responsive literacy instruction requires choosing relevant multicultural literature and other reading materials (Bieger, 1995(Bieger, /1996;;Godina & McCoy, 2000).Multicultural literature should be used in transformative ways that reconstruct the curriculum so that students are able to view concerns, themes, problems, and concepts from the perspectives of diverse groups (Banks & Banks, 1997).Literature should also be selected that allows students to identify social problems and to read about how the main character takes action to solve these problems.This approach helps students realize that all ethnic groups have roots in the past and a strong heritage (Bieger, 1995(Bieger, /1996)).
Culturally responsive literacy programs tap into community resources that promote children's literacy.One way to do this is by enlisting volunteers to serve as reading tutors (Baker, Gersten, & Keating, 2000;Fitzgerald, 2001;Invernizzi, Juel, & Rosemary, 1997;Wasik, 1998;Wasik & Slavin, 1993).Another is to invite parents and others in the neighborhood to share their expertise or "funds of knowledge" on a multitude of topics (Moll & González, 1994).For example, an effective model includes local elders in the schooling of American Indian youth (Aguilera, 2003).Programs that focus on developing partnerships with parents and other caregivers to enhance home literacy experiences also are beneficial.Parents can learn to interact with their children in ways that promote literacy achievement (Arnold, Lonigan, Whitehurst, & Epstein, 1994;Dickinson & Smith, 1994;Valdez-Menchaca & Whitehurst, 1992;Whitehurst et al., 1994).

Culturally Responsive Early Intervention
Students who show early signs of struggling should receive supplemental, intensive instruction that is culturally and developmentally appropriate.We advocate for universal screening models that identify those students who may benefit from such support as soon as they first enter school, as well as progress monitoring that identifies students who do not advance at expected rates.Early intervention should be provided before students have had a chance to fail, as part of a general education support system.Ideally this instruction should take place in small groups, such as one-tothree, or one-to-four (Elbaum, Vaughn, Hughes, & Moody, 2000).Now that IDEA has been reauthorized and the ways we conceptualize and define disabilities are evolving, eligibility and identification criteria are changing as well.States now have the option to use other methods of identifying students as having learning disabilities rather than IQ scores and discrepancy formulae.Many states are moving towards a "Response to Intervention" (RTI) model.Yet what should the interventions in an RTI model look like?A popular current model has three "tiers": the first tier consists of quality instruction in a general education classroom based on evidence-based practices.The second tier is only for those students who do not reach expected benchmarks using an assessment instrument such as the DIBELS-Dynamic Indicators of Basic Early Literacy Skills.These students are then provided with intensive assistance, either as part of a tutoring model or in small groups, still as part of a general education support system.Students who make adequate progress in this second tier return to the first tier where they continue to receive general instruction.Students who continue to struggle are then provided with a third tier level of assistance.It is this third tier that many would consider to be special education.
Yet, like previous eligibility criteria, this model presumes that if a child does not make adequate progress, he or she must have an internal deficit of some kind (Vaughn & Fuchs, 2003).Our position is that we must ensure that the child has in fact received culturally responsive, appropriate, quality instruction.As with earlier identification criteria, this model must be based on students having received an adequate "opportunity to learn" through validated and wellimplemented instructional practices by skilled teachers before they are placed in special education.What should the first-tier interventions be for English language learners?What should the first tier look like for African American students, particularly those living in high-poverty areas?What interventions should the second tier include?Should the interventions be the same for all?If not, how should they vary, and how should this be determined?How can we ensure that practices have been validated with culturally and linguistically diverse students?How can we make sure that teachers are implementing practices with fidelity and making well-informed and appropriate modifications, when needed, to accommodate their students and the local educational context?These are important questions to consider as we move forward with RTI models (for further discussion of these and related issues, see Klingner & Edwards, in press; Klingner, Sorrells, & Barrera, in press).

People
It is people who enact policies and implement practices.In this final section, we discuss the roles of various key players in creating culturally responsive education systems.We describe school administrators, teachers, family and community members, and students.

Culturally Responsive School Leaders
Twenty years ago, effective schools research found that the principal was the key to creating schools of opportunity and possibility (Edmonds & Frederickson, 1978;Jackson, Logsdon, & Taylor, 1983;Weber, 1971).As Fullan points out (2002b), the principal must lead change efforts as an agent who transforms the teaching and learning culture of the school.Even the principal, however, operates within a larger culture, that of the school district, which, in turn, responds to state and federal mandates and policies (Bridgeland & Duane, 1987).How a principal responds to these mandates and policies affects their impact on students.The principals' beliefs, values, educational philosophies, and interpersonal as well as management skills have a great influence on the climate and culture of a school.Haberman and Dill (1999) noted that principals of successful urban schools have certain attributes that heighten their effectiveness.These principals see teacher motivation and nurturing as a top priority.They are skilled in recognizing and building upon each person's strengths (Goldman, 1998).They clearly exemplify a culturally responsive approach and model positive interactions and communications with parents (Danridge, Edwards, & Pleasants, 2000).Overall, effective school leaders empower teachers and students to succeed in school regardless of constraints-they exemplify a sí se puede ("it can be done") attitude.We suggest that for school leaders to be optimally effective in reducing the disproportionate representation of culturally and linguistically diverse students in special education, they must become knowledgeable about the nuanced factors that affect students' opportunities to learn in general education as well as experts in the referral process.
One promising model of school leadership is called "distributed leadership."For this approach to work in schools, the capacity of the organization to reform, renew, and improve is thought to lie in its ability to distribute the responsibilities for leadership throughout the community.In these school communities, learning becomes the central galvanizing theme for the group.As a result, formal leaders, such as principals, can come and go and the work of the community remains strong and responsive to the changing needs of the families and children who attend that school (Elmore, 2000;Spillane, Halverson, & Diamond, 2001).Similarly, other researchers emphasize the value of teachers as leaders in efforts to bring about and sustain change (Katzenmeyer, & Moller, 2001;Lieberman & Miller, 2004).

Culturally Responsive Teachers
Culturally responsive teachers have developed the dispositions that support all students' learning and are knowledgeable and skilled in implementing effective instructional practices.Gay (2000) describes such teachers as cultural organizers, mediators and orchestrators of social contexts.
To act in such a manner requires conscious attention to the ways in which students interact among themselves as well as with teachers.These teachers help their students bridge borders between their home and school cultures, recognize and understand differences in the social milieus, and build on the knowledge and skills that their students bring with them to school learning.In doing so, these teachers demonstrate their care, respect and commitment to each student's learning abilities, desires, and potentialities.
Culturally responsive teachers specifically acknowledge the need for students to find relevant connections among themselves, the subject matter, and the tasks they are asked to perform (Montgomery, 2001;Salend, Garrick Duhaney, & Montgomery, 2002).They know that students learn best when their experiences and interests serve as the basis for curriculum connections, making learning relevant to their lives (Ladson-Billings, 1994).Delpit (1995) helps to expand these concepts by asserting that culturally responsive teachers must explicitly teach skills and cultural capital, or, in other words, the knowledge and behaviors valued as being of high status by the dominant culture.Nieto (1999) describes the expertise of culturally responsive teachers in instruction and management and their ability to challenge and simultaneously support their students.Finally, culturally responsive teachers feel a strong sense of responsibility for all students, including students referred for or already placed in special education (Villegas & Lucas, 2002).We assert that unless teachers behave in the culturally responsive manner we describe, their students may be miseducated and underachieve not because they have internal deficits and should be in special education, but because they have not been taught in ways that promote their learning.

Families and Communities
We begin with the premise that many of the difficulties faced by culturally and linguistically diverse students in schools are not caused by intrinsic deficits.Nor do they indicate failures of families.We see families and communities as encircling students inside the layers of school-related personnel.Families and communities possess resources and abilities, or rich "funds of knowledge," that can promote student learning and enrich the context of the school and classroom (Moll, 1992;Nieto, 1999).
Historically, deficit views of culturally and linguistically diverse families and of children's learning difficulties have combined to discourage family participation in schooling and in the special education referral process (Harry, 1992).When families are not perceived to be valued partners by school personnel, students may feel forced to choose loyalty to family over school success (Delpit, 2002).Delpit (1995) suggests that the answers to the issues faced by culturally and linguistically diverse students and families "lie not in a proliferation of new reform programs, but in some basic understandings of who we are and how we are connected to and disconnected from one another" (p.xv).We recommend professional development for teachers and administrators to change a pervasive negative attitude towards culturally and linguistically diverse families living in poverty and a propensity to blame them for their children's struggles (Harry & Klingner, in press).We see great potential in proactive collaborative models that focus on family strengths and finding common ground upon which to build, and include parents in assessment, placement, and policy-making decisions, thereby restoring the balance of power in parent-professional discourse.
Schools that successfully make a positive difference in the lives and learning of culturally and linguistically diverse students work closely with families and communities as valued, respected partners (Epstein & Dauber, 1991;Giles, 1998;Harry, 1992;Kalyanpur & Harry, 1999).Sands, Kozleski, and French (2000) suggest that schools are becoming increasingly aware of the importance of parental involvement and are actively working to develop a process for establishing collaborative, reciprocal relationships.Davis and colleagues (2002) found many strategies being used to increase parental participation, including improving interactions between teachers and parents, making home visits, bringing in extended family, and involving parents in school cultural activities and in planning parent training programs.They suggest involving parents in new roles as assessors, policymakers, and advocates.This level of reform requires a framework of systemic change that is inclusive of all participants, and impacts change across all parts of the system.

Students
Our focus is on children's potential and promise, not risk factors, emphasizing students' strengths and amazing abilities to be resilient and to overcome adversity (Spencer, Noll, Stolzfus, & Harpalani, 2001).Culturally and linguistically diverse children living in poverty come to school with a variety of background experiences and from complex circumstances.Students' backgrounds are assets that teachers can and should use in the service of their learning.
Too often, students feel that they must choose to be academically successful at the cost of their cultural identity (Gay, 2000;Valenzuela, 1999).Yet academic success and cultural identity can and must be simultaneously achieved, not presented as dichotomous choices.O' Connor's (1997Connor's ( , 1999) ) studies of high-achieving African American students illustrate valuable strategies for maintaining identity.She examined how African American high school students living in low-income neighborhoods situated race, class, and gender while struggling to attain status.She found that the students emphasized the importance of hard work, individual effort, and education.However, students' overall views were complicated by their interpretations of how race, class, and gender affect life chances.
How can we bring teachers to the recognition that strong cultural identity supports, rather than detracts from, academic success?We must do a better job of helping teachers come to this understanding in our pre-service and in-service teacher education programs by emphasizing the importance of connecting with students' identities across the curriculum and of developing strong reciprocal relationships.Eleuterio (1997), in language arts, and Hoelscher (1999), in social studies, observed that classrooms filled with teachers and students who openly shared their lives, cultural identities, and life experiences built trust and fostered stronger relationships.This climate led to student engagement and excitement about learning together.Lave and Wenger (1991) emphasized the worth in acknowledging and valuing the cultural and linguistic backgrounds of students and the development of relationships based on care, respect, and responsibility.

Conclusion
Our intent is to decrease inappropriate referrals to special education and promote a more equitable education system that results in proportionate representation in special education among culturally and linguistically diverse students.Clearly, to accomplish this monumental task will not be easy.Yet through networking with others involved in similar work, raising awareness and discussing tough questions about who benefits from current practice, and instituting interventions designed to bring about real change, we are optimistic that with relentless effort we can make a difference in the lives of culturally and linguistically diverse students with and without disabilities.We emphasize the importance of collaborating with state departments of education and school districts across the nation, national organizations, technical assistance centers, research and personnel preparation projects, and institutions of higher education to ensure that culturally responsive evidence-based early interventions, reading programs, and behavioral practices are included in teacher preparation and professional development programs.We are engaged in a process of consciousness-raising designed to lead to transformative change (Freire, 1984).Together, we not only can change the world (Oakes & Lipton, 2002), we must.representation in special education, English Language Learners (ELLs), gender equity, assessment, and curriculum and instruction.David P. Riley, Ph.D., is Executive Director of the Urban Special Education Leadership Collaborative at Education Development Center, Inc., in Newton, MA.A national network of special and general education administrators, the mission of the Collaborative is to improve outcomes for children and youth with disabilities in urban school districts through leadership development.

Table 1
Ethnic Representation in Special Education, in Numbers and Percentages (National Aggregates)