The placement of students of color in special education classes is perhaps one of the most complex problems facing educators as we move into the new millennium.(n1) The issue itself is not new. In 1968, Dunn, citing U. S. Office of Education statistics, reported that about 60 to 80 percent of the pupils taught by [teachers in mild mental retardation or MMR classes] are children from low status backgrounds--including Afro-Americans, American Indians, Mexicans, and Puerto Rican Americans; those from nonstandard English speaking, broken, disorganized, and inadequate homes; and children from other non-middle class environments. (p. 6)

In 1968, the Office for Civil Rights (OCR) began surveying special education placement in school districts. The survey has since been taken approximately every 2 years.(n2) Meanwhile, several studies had a significant impact on our awareness and understanding of this problem. For instance, Mercer published in 1973 the results of her circa 1968 study of students in Riverside, California. Hispanics comprised 7% of the age 6-15 school sample in Riverside, but 12% of the similar age students placed in classes for students with MMR--at that time, these students were referred to as "educable mentally retarded" or EMR. Conversely, while white students made up 82% of the school population, they represented only 53% of students placed in EMR programs. African Americans represented 9.5% of the district population although they made up 32% of students in MMR classes (Reschly, 1996).

With the passage of Public Law 94-142 in 1975, the subsequent Individuals with Disabilities Education Act, continuous monitoring by OCR, and years of litigation, some argue the degree of overrepresentation has receded slightly, though the basic findings have remained fairly consistent: Overrepresentation affects the so-called high incidence (emotional disturbances, MMR, and learning disabilities) or "judgmental" disabilities and the students that are consistently affected at the national level are African Americans, particularly males in mental retardation (MR) and emotional disturbance (ED) programs.(n3) Latinos and American Indian students are overrepresented in some disability categories in selected school districts and states.

The misplacement of students in special education is problematic in that it is not only stigmatizing, but it can also deny individuals the high quality and life enhancing education to which they are entitled. But what are the factors that shape this problem? And have there been changes since the problem was first identified over 3 decades ago? Before we answer these questions, however, we must first define how overrepresentation is calculated.

There are two equally valid ways to present overrepresentation figures (Reschly, 1997). The first, favored by OCR, looks at special education enrollment by (ethnic) group. For example, if 33% of the MR enrollment is African American, but 17% of the overall school population is African American, the representation of these students is about twice the level expected. Clearly there is overrepresentation that is of concern. However, what percent of African American students are classified as MR?

The answer to that question often is surprising. In fact, it is neither 17% nor is it 33%. The actual percentage of African American students classified as MR according to the 1997 OCR survey was 2.54%. The latter percent is the percent of the (ethnic) group in the special education program. The confusion about these two indicators is harmful because it may establish or perpetuate the myths that large proportions of African Americans and other minority children have disabilities, that something is fundamentally wrong with certain minority children, or that special education is primarily used to deny educational rights to minority children. In fact, small, but clearly disproportionate, percentages of minority children are placed in special education.

Understanding the difference between the proportion of program by group and group in program is very important to gauge the magnitude of the problem. Both indicators are valid and useful. For example, looking at the percentages in special education placements of African American in MMR (e.g., 33%) or emotionally disturbed (ED; e.g., 28%) classes allows us to see the overidentification problem in relation to the students served in these disability programs. This figure, however, does not tell us anything about the actual percentage of the African American population placed in these programs. Such information is obtained by looking at the percent of group indicator, which suggests the percent of all African American students that is placed in MR classes is 2.54% and in ED programs is 1.29%. The 2.54% figure may not appear to be very large. However, when one realizes that it is five times that of Asian/Pacific Island group and over twice that for Whites, the problem should be deemed as serious. The following summary of data will illustrate both approaches.

OCR always presents data as the special education enrollment by (ethnic) group. In 1978, African Americans made up 16% of the school population. However, 38% of the students in EMR classes, 27% of the students in "trainable (moderate) mentally retarded" (TMR) classes, and 24% of the students in ED classes were African American. In the 1997 OCR survey, African Americans made up 17% of the student population. Yet, African American students made up 33% of the classes for students with MR and 28% of ED classes. This means that, at these two moments in time, African Americans had the greatest risk of being placed in MR and ED classes and that such representation was almost twice as large as their representation in the general education system. Let us contrast the aforementioned figures on special education enrollment by group with the data on percent of (ethnic) group in special education programs: The actual proportion of all African American students that is placed in MR classes was 2.54% versus 0.49% for Asian/Pacific, 0.74% for Latino, 1.31% for American Indian, and 1.13% for white students. This means that while only a small percentage of all African Americans are actually classified as MR, clearly, these students are overrepresented in this category.

The OCR survey in 1997 included only 3 of the 13 categories of disabilities recognized in federal law (MR, learning disabilities or LD, and ED). The other categories were not included because prior OCR surveys did not find overrepresentation in those categories or the other categories have had relatively equal representation across all groups; however, there is preliminary evidence that overrepresentation of English Learners may be emerging in the Speech and Language Impairment category in some states.

In conclusion, two commonly used indicators of overrepresentation are the special education enrollment by group and the percent of group in special education. Each indicator offers important yet partial information. We recommend the use of both indicators to get a clearer idea about the magnitude of the problem. At the same time, these indicators do not offer a comprehensive perspective on the problem unless other aspects are taken into account such as whether placement data are examined cross-sectionally or longitudinally and the need to disaggregate the data by disability, diverse subgroups, and geographical region.

One problem with the examination of national cross-sectional data is that it masks potentially important trends. For instance, the difference between the percent of African American and White students placed in high incidence disability classes declined in the late 1980s but it seems to be rising in the 1990s (Artiles & Trent, 2000). These trends vary as we disaggregate the data by disability category; more specifically, the Black-White difference declined in the MR category from 1986 to 1992 but it begun to rise again in 1994. In contrast, the Black-White difference in ED placement has remained relatively stable after a decline in the late 1980s (Artiles & Trent).

Another problem associated with reporting national data and general group categories is that statistics on subgroups are obscured. For example, Asian and Pacific Islanders are reported as a single category. This Federal category includes Asian groups such as Chinese, Japanese, and Pakistanis. It also includes Pacific Island groups such as Hawaiians, Samoans, and Tongans. Native Hawaiians are placed in special education classes at disproportionately high percentages. However, the categorical reporting of Asian/Pacific Islanders obscures this fact (see also Artiles, Rueda, Salazar, & Higareda, in press, for an analysis of English learner subgroups).

National data also obscure individual state variability. For example, 1997 OCR national data indicate African Americans and American Indians are overrepresented in high incidence programs; in the case of African Americans it is particularly noticeable in ED classes. However, the problem is even more acute in states such as Kentucky: African Americans made up 9% of the total student population, and an estimated 26% of the ED classes. These data could suggest that African Americans in Kentucky are more prone to ED at a disproportionately higher level than national trends. This variability in placement rates raises questions about flaws or discrepancies in the identification process, the diagnostic criteria utilized, as well as the definition of the ED category. State variations in placement rates are also evident for Latino, Asian/Pacific, Native American, and White students in states such as Arizona, Hawaii, Texas, and Alaska (see also Artiles et al., in press). Another factor that is reportedly associated with overrepresentation is the availability of alternative programs, such as bilingual education.

Multiple factors are entangled in this complex predicament. Explanations range from the pervasive impact of poverty on minority children's development to institutional discrimination that may result in lower expectations, over-referrals, and overidentification. The following discussion illustrates the complex interconnections between the multiple factors that shape this problem.

Although poverty and special education placement are associated, it is important to note that poverty can contribute both directly and indirectly to the risk of school failure, special education placement, or both. This has important implications for people of color because, although there are more white Americans living in poverty than any of the racial/ethnic groups of color, the percentages of the latter living in poverty far exceeds that of Whites. The booming United States economy in the late 1990s enabled the United States Census Bureau (2001) to report poverty rates that were below or equal to the lowest recorded rates for all groups except Whites. In 1999, the overall poverty rate was 11.9%, the lowest since 1979. The poverty rate for Whites was 8%, compared to 11% of Asian/Pacific Americans, 23% of Latinos, 24% of African Americans, and 26% of American Indians.

Poverty contributes to a significant number of problems such as less than optimal medical care both at the prenatal stage for expectant mothers, as well as postnatal care for the newborn. Physicians providing medical care through governmentally supported clinics and agencies are often overburdened with large patient loads, and are unable to provide the more thorough care purportedly offered by private physicians or health maintenance organizations. Because of the long waits, language differences, immigration issues, and limited access (e.g., inefficient public transportation) that many poor people may face, even the medical care to which they may be entitled is perceived as less accessible and may not be utilized as often as necessary. Furthermore, appropriate nutrition for expectant mothers and for newborns, infants, and children is problematic for those living in poverty. Expectant mothers who out of necessity work late into term, and who have inadequate nutrition and prenatal care, are more prone to having children born at risk. Children born preterm (i.e., those who weigh less than 3 lbs. 5 oz.) may be at risk of cognitive and sensory impairments (Drew, Logan, & Hardman, 2000).

Although more closely aligned with socioeconomic status (SES), preterm births have been associated with ethnicity. Gelfand, Jenson, and Drew (1988) report that 51% of non-White births have complications, as compared to 5% of White upper class births. Children living in older homes (which may be the case of many living in poverty) may be at greater risk for lead poisoning; it has been reported that one in six children suffers from lead poisoning (Carolina Environment, Inc., 1999). Lead poisoning can be associated with reading and LD, language impairments, lowered I.Q., neurological deficits, ED, MR, kidney disease, heart disease, stroke, and death (Carolina Environment, Inc.). Other medically related problems include crack babies and babies with fetal alcohol syndrome being born to poor mothers involved in substance abuse (Drew et al., 2000).

Aside from the pervasive potential impact of poverty on children's development, it should be acknowledged that poverty is associated with lower academic achievement, which in turn exacerbates the chances of special education placement. At the same time, recent research suggests the need to assess the contexts of schools and communities to better understand overrepresentation patterns. For example, Oswald, Coutinho, Best, and Sing (1999) found that African American overrepresentation in MMR classes worsens as the school poverty level increases, whereas African Americans have a greater chance to be placed in ED programs in low-poverty schools.

Furthermore, the latest research on the link between socioeconomic disadvantage and child development is moving from descriptions of the impact of poverty to more dynamic analyses of the processes by which these effects come about (McLoyd, 1998). Alternative measures of poverty, the duration, timing, and the context of poverty (e.g., neighborhood-level poverty) are important considerations. Recent findings suggest early and persistent exposure to poverty has negative developmental outcomes. Moreover, research has traditionally focused on African Americans or racially diverse samples; McLoyd argues these practices might limit the generalizability of findings to groups with high rates of poverty such as Latinos and some Asian American subgroups.

It is equally important to examine the processes and factors that protect children from the negative influence of poverty given the growing knowledge base that suggests "competence develops in the midst of adversity when, despite the situation at hand, fundamental systems that generally foster competence in development are operating to protect the child or counteract the threats to development" (Masten & Coatsworth, 1998, p. 212). Examples of such protective systems-influences include bonds to prosocial adults outside the family and effective schools. Indeed, we must strive to craft a knowledge base that emphasizes "possibility" for poor minority students.

As stated above, an overwhelming number of special education students are poor, male, and ethnic minority; educators are primarily middle class, female, and White. When the cultural backgrounds of students and teachers are incongruent, it may result in interpersonal misunderstandings, which may have consequences for special education placement. Ysseldyke, Vanderwood, and Shriner (1997) suggest a large percentage of students who are referred to special education are eventually placed in such programs. Teachers do the vast majority of referrals. Although the existing empirical evidence on referral bias is equivocal, there is some evidence middle class elementary school teachers possess greater proclivity to racial biases with regard to achievement expectations and perceptions of student maturity; also, teacher perceptions and expectations of poor students are less positive, "largely on the basis of noncognitive considerations (e.g., speech patterns and dress)" (McLoyd, 1998, p. 193). Ladner and Hammons (2001) reported that districts with more White teachers had a greater rate of minority special education enrollment while White student special education placement was unaffected by the racial composition of the faculty.

The assessment process has been controversial in the special education process. Major court decisions have raised red flags, and have resulted in mandates for the use of culturally responsive procedures and tools with African Americans and English learners in special education eligibility decisions. Two California court cases shaped largely these regulations, namely Diana v. State Board of Education (1970) and Larry P. v. Riles (1979; see Reschly, 1997 for a discussion of these cases). Some of the most important outcomes of Diana include a mandate to test in students' primary language, use nonverbal tests, and use extensive supporting data in future placement decisions. A major outcome of Larry P. was a ban on the use of I.Q. tests for identification and placement purposes with African American students in California.

There is considerable evidence about the inadequacies of traditional assessment models with culturally diverse groups (García & Pearson, 1994); however, research about the role of norming, content, linguistic, and cultural biases and test result uses on minority placement in special education is scarce. Future research on assessment and placement of students of color should focus on the complexities associated with these processes and, thus, research should document not only the impact of instruments' technical inadequacies, but also the institutional, historical, and political contexts of assessment and placement practices. Such focus will imply a shift in assessment practices from an exclusive attention to children's individual factors to an understanding of children's development as situated in cultural contexts.

In this vein, evidence from a 3-year ethnographic study in Florida (Harry, Klingner, Sturges, & Moore, in press) points to the many ways in which the assessment process is influenced by unofficial, undocumented practices. These include informal pressures from school administrators, referring teachers (or both), teachers' and psychologists' unacknowledged biases regarding children's family structures and practices, and widely varying choice and implementation of psychological assessment tools. Furthermore, this research revealed that child study teams seldom take into account information regarding the atmosphere and practices obtained in the classrooms of referring teachers. In the study, several children were referred from classrooms where very poor instruction and classroom management were the norm, making it impossible to know if the children's difficulties might have been mitigated in more effective classroom environments.

Overrepresentation is a multifaceted problem; thus more and better research is the foremost priority. Future inquiries must be comprehensive, interdisciplinary, and should transcend analyses of placement figures. Efforts should concentrate on at least two broad domains: (a) structural antecedents and mediating forces, and (b) the contexts and activities associated with the special education process.

Because poverty rate is higher among minority people, some scholars argue this is the main explanation of minority overrepresentation. Poverty indeed can have a brutal impact on people's lives. At the broadest level, therefore, we should invest in the prevention of the nefarious consequences of poverty. Educators and public health professionals need to develop curricula and public education campaigns that emphasize good nutritional practices throughout one's life, but especially during the prenatal period and with infants and young children. Such efforts should also emphasize the hazards of lead poisoning and other health dangers. In addition, it is necessary to empower students and their families to access available health care.

At the same time, the complexities associated with poverty have been understudied in the context of overrepresentation. For instance, we know little about the potential mediating effects of the duration, timing, context, and various definitions of poverty on special education placement. Moreover, the literature on poverty cited above suggests that, just as poverty brings poor health care and numerous environmental hazards, it also brings poor schooling. Some of the evidence cited earlier hints at the need to study the politics of special education placement in various socioeconomic contexts. For example, we must examine the dynamics behind the differential placement of African Americans in high-versus low-poverty districts. Indeed, research on the effects of poverty must include measures or descriptions of the quality of schooling to which poor children are exposed, including funding, availability of resources, and the quality of instruction.

Another compelling mediating force is society's notion of "difference" and the ways in which the educational system responds to "different" people. Let us remember "difference" is a comparative notion. Patton (1998) suggests that those who produce the special education knowledge base have attempted to understand and explain the behavior and life experiences of African Americans through their own ethnocentric perspectives and against an equally narrowly constructed cultural-ethnic standard. Thus, it is imperative future research focuses on questions that are relevant to issues of difference, such as What are the assumptions about difference that inform decisions to place students in special education? What functions are served by the maintenance of a rigid demarcation between general and special education? When (under what circumstances) is overrepresentation a problem? What are the consequences of overrepresentation? For whom? and What is the function of special education in an increasingly diverse society?

An elusive factor with deep structural roots is bias. Discussions about the contributions of bias to school failure and to the likelihood of special education placement should acknowledge that bias is more than the personal decisions and acts of individuals. Rather, bias against minorities should also be thought of in terms of historical residua that are layered in social structures and that may lead to various forms of institutional discrimination. It is necessary that educators learn to recognize biases in educational settings and institutional practices and learn to deal with them appropriately.

The special education process encompasses multiple aspects that include the preparation of school personnel, the social organization of learning in general education classrooms, prereferral interventions, referrals, assessment, eligibility, and placement practices. The literature on solutions in each of these stages is growing. A review of solutions for each of these phases is beyond the scope of this article. A reference list on this topic is available from the first author. In this section, we only highlight a couple of basic ideas as they pertain to personnel preparation and prereferrals for these are among the least developed in the literature on overrepresentation.

(n1) Scholars refer to this problem as "disproportionate representation," which is defined as "unequal proportions of culturally diverse students in [special education] programs" (Artiles & Trent, 2000, p. 514). This problem includes both over and underrepresentation. The former is observed in high incidence disability programs, whereas underrepresentation is observed in programs for students with gifts and talents. Due to space constraints, we focus on the overrepresentation problem only.

(n2) Although the OCR data is perhaps the most widely used source, researchers also rely on various databases such as the U.S. Department of Education's data included in the Office of Special Education Programs' annual report to Congress and the National Longitudinal Transition Study. These databases have limitations that must be considered when interpreting research reports. For instance, the OCR data include 50 large urban schools plus a selection of districts throughout the nation; although this is not a nationally representative database, projections can be made from these data at the state and national level. Also, the database is based on school self-reports, there is some variation in data collection across regions, sampling procedures may vary from year to year, and critical information such as social class and language proficiency is omitted.

(n3) MacMillan and Reschly (1998) argue that overrepresentation is a problem restricted to the "judgmental" disability categories because ethnic overrepresentation in certain biologically based conditions (e.g., Phenylketonuria among White, Tay Sachs among Jewish, and sickle cell anemia among African American children) is not debated.

??? Artiles, A. J., Rueda, R., Salazar, I., & Higareda, J. (in press). Factors associated with English learner representation in special education: Emerging evidence from urban school districts in California. In D. Losen & G. Orfield (Eds.), Minority issues in special education in the public schools. Cambridge, MA: Harvard Publishing Group.

??? Artiles, A. J., & Trent, S. C. (2000). Representation of culturally/linguistically diverse students. In C. R. Reynolds, & E. Fletcher-Jantzen (Eds.), Encyclopedia of special education, Vol. 1 (2nd ed., pp. 513-517). New York: John Wiley & Sons.

??? Carolina Environment, Inc. (1999). How lead affect your child's health [On-line]. Available: http://www.knowlead.com/affect.htm

??? Darling-Hammond, L., & Post, L. (2000). Inequality in teaching and schooling: Supporting high quality teaching and leadership in low income schools. In R. D. Kahlenberg (Ed.), A notion at risk: Preserving public education as an engine for social mobility (pp. 127-168). New York: The Century Foundation Press.

??? Diana v. State Board of Education, Civil Action No. C-7037RFP (N. D. Cal. Jan. 7, 1970 & June 18, 1973).

??? Drew. C. J., Logan, D. R., & Hardman, M. L. (2000). Mental retardation, a life cycle approach. Columbus, OH: Pearson/Prentice-Hall.

??? Dunn, L. (1968). Special education for the mildly mentally retarded: Is much of it justifiable? Exceptional Children, 23, 5-21.

??? García, G. E., & Pearson, P. D. (1994). Assessment and diversity. Review of Research in Education, 20, 337-391.

??? Gelfand, D. M., Jenson, W. R., & Drew, C. J. (1988). Understanding child behavior disorders (2nd ed.). New York: Holt, Rinehart and Winston.

??? Harry, B., Klingner, J., Sturges, K., & Moore, R. (in press). Of rocks and soft places: Using qualitative methods to investigate the processes that result in disproportionality. In D. Losen & G. Orfield (Eds.), Minority issues in special education in the public schools. Cambridge, MA: Harvard Publishing Group.

??? Individuals with Disabilities Education Act, Pub. L. No. 104-17, 111 Stat. 37.

??? Kozol, J. (1991). Savage inequalities. New York: Crown.

??? Ladner, M., & Hammons, C. (2001). Special but unequal: Race and special education. In C. E. Finn, A. J. Rotherham, & C. R. Hokanson (Eds.), Rethinking special education for a new century (pp. 85-110). Washington, DC: Thomas B. Fordham Foundation.

??? Larry P. v. Riles, C-71-2270, FRP. Dis. Ct. (1979).

??? Lee, V. E., & Loeb, S. (2000). School size in Chicago elementary schools: Effects on teachers' attitudes and students' achievement. American Educational Research Journal, 37, 3-31.

??? Lee, V. E., Bryk, A. S., & Smith, J. B. (1993). The organization of effective secondary schools. Review of Research in Education, 19, 171-267.

??? MacMillan, D. L., & Reschly, D. J. (1998). Overrepresentation of minority students: The case of greater specificity or reconsideration of the variables examined. The Journal of Special Education, 34, 15-24.

??? Masten, A. S., & Coatsworth, J. D. (1998). The development of competence in favorable and unfavorable environments: Lessons from research on successful children. American Psychologist, 53, 205-220.

??? McLoyd, V. C. (1998). Socioeconomic disadvantage and child development. American Psychologist, 53, 185-204.

??? Mercer, J. (1973). Labeling the mentally retarded. Los Angeles: University of California Press.

??? Ortiz, A. A., & Wilkinson, C. Y. (1991). Aim for the best, assessment and intervention model for the exceptional bilingual student. Austin, TX: University of Texas, Austin.

??? Oswald, D. P., Coutinho, M. J., Best, A. M., & Singh, N. N. (1999). Ethnic representation in special education: The influence of school-related economic and demographic variables. The Journal of Special Education, 32, 194-206.

??? Parrish, T. B., Matsumoto, C. S., & Fowler, W. J. (1995). Disparities in public school spending, 1989-90, NCES 95-300. Washington, DC: National Center for Education Statistics.

??? Patton, J. M. (1998). The disproportionate representation of African Americans in special education: Looking behind the curtain for understanding and solutions. The Journal of Special Education, 32, 25-31.

??? Reschly, D. J. (1997). Disproportionate minority representation in general and special education: Patterns, issues, and alternatives. Des Moines, IA: Mountain Plains Regional Resource Center, Drake University (99 pages). ERIC Document No. 415 632.

??? Reschly, D. J. (1996). IQ and special education: History, current status, and alternatives. Washington, DC: National Academy of Sciences.

??? Rothstein, R. (2000). Equalizing education resources on behalf of disadvantaged children. In R. D. Kahlenberg (Ed.), A notion at risk: Preserving public education as an engine for social mobility (pp. 31-92). New York: The Century Foundation Press.

??? United States Census Bureau (2001). Poverty: 1999 Highlights [On-line]. Available: www.cenus.gov/hhes/poverty99/pov99h.html.

??? Ysseldyke, J. E., Vanderwood, M. L., & Shriner, J. (1997). Changes over the past decade in special education referral to placement probability: An incredibly reliable practice. Diagnostique, 23, 193-201.

PART III: CREATING MULTICULTURAL CLASSROOMS

Through the hotel window, the hubbub of Manhattan filters in with the late afternoon sun. In the hazy half light, the features of Chan Lor, Hmong refugee from Laos, are revealed. He is sitting in a stiff chair near the front of a small conference room, alongside his wife, Khou, and Faye Van Damme, a preschool teacher. Like the slow, whispering voice of fire, his words flow, gaining strength and surpassing the haunting sounds of the qeng (a bamboo wind instrument) that emanate from a tape player in the back of the room. He becomes silent, and Van Damme, teacher and researcher of family lives, begins to weave in her own story with a translation of the events reported by Chan, focusing on an eventful day in their respective lives: January 15, 1983. On that day, Faye celebrated her 15th birthday in Wisconsin, while a world away, Chan attempted the dangerous crossing of the Mekong River:

There is much laughter and frivolity as friends and family enter the balloon-and streamer-filled house. Games are played, gifts are opened, and cake is eaten; all in the name of fun.

There is no laughter, only seriousness and dead quiet. It is a fifteen mile walk to the shores of safety. It is unknown who is going to die. Many lose their feet from walking on land mines. Shirttails are held, boats are loaded, and strangers are trusted; all in the name of freedom. (Van Damme, 1998, p. 2)

Faye Van Damme and Chan and Khou Lor are first-time visitors to New York, newcomers and new voices in the world of academic conferences. They bring with them a story of many border crossings, a story that is a weaving of two lives, a story about bridging the gap between schools and the homes of bilingual students and families.

In the following pages, I describe a narrative research project carried out by preservice and inservice teachers working with bilingual families in northeastern Wisconsin. The objectives of this project were to improve the understanding of preservice and inservice teachers of the lives of immigrant families and the issues they face and to facilitate a dialogue between the home and the school. I begin with an outline of the cultural roles of teachers, a theoretical framework that guided the development of this project: Teachers have the potential to be cultural storytellers, cultural healers, and cultural workers in their relations between schools and the homes of bilingual students. Next, I describe the context of this study within a semester-length course and with narrative inquiry as the chosen mode of research. Following this are excerpts from the family narratives, interpreted through the cultural roles framework. Finally, the implications of this project for educators, family participants, and researchers are developed.

In educating minority-language children and youth, teachers often serve as cultural border guards, inculcating the ways of the dominant culture while closing the door on students' own language and culture. How does this process work? First, there is often little contact between teachers and bilingual families (Valdes, 1996; Walker-Moffat, 1995). Problems being experienced by bilingual children often are blamed on "the home" (Delgado-Gaitan, 1996), and bilingual families are usually left out of conversations about second-language education policies in the schools (Soto, 1997). The ways in which bilingual families and their children are perceived by teachers and schools are still greatly influenced by theories of cultural deprivation as an explanation of school failure.

Theories of cultural deprivation and a "culture of poverty" that emerged in the 1950s and 1960s suggested that children from certain nondominant cultural groups grow up deficient in needed cultural attributes and are trapped in a cycle of failure (Bereiter & Englemann, 1966; Lewis, 1966; McCandless, 1952). The implicit assumption behind cultural deprivation theory is that the home culture (and language) is the problem and that to succeed, students from such homes must be taught to perform in traditional mainstream ways. Even federal legislation of the era, such as the Bilingual Education Act (1968), which ostensibly supported services to students in their home language, was compensatory in nature, focusing on the deficiencies of nonnative English speakers (Brisk, 1998).

Nevertheless, recent research suggests that bilingual families contribute in valuable ways to their children's learning. Moll, Velez-Ibañez, and Greenberg (1989) identified funds of knowledge, sources of skills and information for daily living, in their work within a working-class Mexican American community. In projects that include teachers as researchers of their communities, these funds of knowledge from students' homes become resources available for curricular innovation and pedagogical change. In a variation of this approach, Andrade (1998) engaged bilingual children as ethnographers of their communities and their classrooms. Ethnographic and narrative research has revealed rich portraits of the skills, strengths, and values present within bilingual and bicultural families (Carger, 1996; Hones & Cha, 1999; Valdes, 1996).

Because many teachers and schools lack a strong connection to the lives and learning of bilingual and bicultural families, there are important reasons to involve preservice and inservice teachers in family-based projects. In describing the success of a family studies program in New Mexico for preservice teachers, Grinberg and Goldfarb (1998) argued that teachers must be sensitized to realities of children's worlds and become skilled in bridging the worlds of the home and the school. McCaleb (1998) highlighted the importance of preservice teacher involvement with diverse families to promote literacy, whereas Olmedo (1997) described ways in which family oral histories gathered by teachers can be woven into the social studies and history curriculum. Research with linguistically and culturally diverse families, therefore, has implications for teacher education, academic achievement, school-community relationships, and the development of multicultural curriculum.

Clearly, teachers have the potential to play a variety of cultural roles in the lives of linguistically diverse youth: the teacher can be a storyteller, a collector of stories, and an interpreter; the teacher can be a healer who can help students adjust to life in the dominant culture without stripping away their own cultural understandings; building on these two previous roles, the teacher can be a cultural worker who addresses issues of power in the classroom and in society.

In our English as a Second Language (ESL)--bilingual education licensure program at the University of Wisconsin, Oshkosh, we attempt to address the urgent need to build bridges between schools and various linguistic and cultural communities. We seek to prepare teachers who

• Value linguistic and cultural diversity, and understand their multiple cultural roles as second-language educators.

• Can assess themselves, their schools, and their society critically.

• Have the ability to engage in real educational problem solving with others.

• Can use participatory action and narrative research to reach out to minority-language communities and build authentic curriculum and instruction.

• Are producers of knowledge and creative members of professional communities.

Students in our program represent the growing demographic diversity of northeastern Wisconsin. Approximately 70% of the students are of European American heritage, 20% are Asian American (predominantly Hmong), and 10% are Latina (of diverse Latin American origins). Most have grown up in one of the nearby communities, although some spent their early years in Thai refugee camps, the Philippines, Puerto Rico, and other diverse localities. In general, our students are hard-working, they are interested in learning about other languages and cultures, and they have good academic preparation. What our students need, especially those coming from European American backgrounds, is a stronger connection to the lives of bilingual students and families in their homes and communities.

One of the courses in our licensure sequence, Principles of Bilingual/Bicultural Education, was chosen as a place to engage students in bilingual family research. This course introduces students to broader sociology, politics, and policies of language and culture; it explores ethnographic and narrative tools available for doing research within minority-language communities found in texts such as Valdes's (1996) Con Respeto: Bridging the Distances Between Culturally Diverse Families and Schools and Walker-Moffat's (1995) The Other Side of the Asian American Success Story; and it engages students in dialogues with members of bilingual families through a semester-length research project. The culmination of the bilingual family research projects are written family narratives from which students present excerpts as performance events. These narratives, coauthored and coedited wherever possible with family members, document the stories of the participants, framed by the historical, cultural, and sociological contexts of their lives as immigrants and refugees.

The objectives of this family research project are to improve the understanding of preservice and inservice teachers of the lives of immigrant families and the issues they face and to facilitate a dialogue between the home and the school. This project encourages participation of bilingual family members in the retelling of their own stories and voicing their ideas about issues of education inside and outside schools (Lincoln, 1993). Teachers, as cultural border crossers, are expected to gain a deeper understanding of the lives and concerns of bilingual families and to take a critical stance in support of these families and their counternarratives, or stories that provide a counterpoint to the dominant culture's depiction of minority groups (Giroux, 1997). There is also a need to inform the wider monolingual community about issues facing bilingual families and children.

Narrative research is the mode of inquiry used in this research project. Denzin's (1994) interpretive interactionism, involving the organization of life histories around "epiphanies" or life-shaping events, has been particularly influential for student researchers. This style "begins and ends with the biography and the self of the researcher," and encourages personal stories that are thickly contextualized, and "connected to larger institutional, group and cultural contexts" (Denzin, 1994, pp. 510-511). Moreover, the stories presented in the text "should be given in the language, feelings, emotions, and actions of those studied" (p. 511). As indicated in the work of Grumet (1991) and McBeth and Horne (1996), many ethical and methodological issues need to be addressed in the writing of bilingual/bicultural narratives, especially when not all student researchers share the same cultural background as their informants. The importance of addressing the informants' stories in a respectful manner is underscored, and family research participants are included in the process of analysis, interpretation, and, wherever possible, as coauthors of the final text.

In the opening weeks of class, students write autobiographically of their own connections to language, culture, family, and education. They are asked to keep their personal stories in mind as they embark on narrative, ethnographic research with volunteering minority-language families over a 10-week period. Individuals and small groups of students are paired with each family, learning about their lives and initiating dialogues about the education of their children. Thus, data sources for this research include, with permission from students and their informants, students' written accounts of their own lives, transcriptions of audiotaped interviews with research participants, field notes, videotapes and photographs of family life, and library research.

One of the products of the bilingual family research project are written narratives of family lives. These narratives highlight the multiple voices of participants involved in the research, including the voices and autobiographical experiences of the researchers themselves. Moreover, these narratives served to illustrate the cultural roles as storytellers, healers, and workers that preservice and inservice teachers fulfilled as part of this project. Excerpts from these narratives illustrate the journey undertaken by these students of bilingual family lives and what they have learned about how to bridge the two worlds of school and home.

Students' awareness of the importance of their biographer role increased after writing their own cultural autobiographies. One of our students, a third-generation Polish American, writes of her search for information about her family cultural heritage:

When I decided to get to know my own roots, there weren't very many people left who knew the stories behind the immigration of my great-grandparents. It seemed as if Judga and Busia never shared stories about their homeland, their village, their reason for leaving all that was familiar to them behind. Instead, I had to find my stories in courthouses, archives, and family history centers. (Kiedrowski, 1998, p. 2)

With their appreciation of their own cultural past comes a new interest in creating opportunities for their minority-language students to retell their life experiences with language and culture. Several of our students develop curricular units that integrate language skills and content areas by using storytelling as a theme. For example, one of our inservice teachers, with the support of a Hmong community activist, prepared her elementary ESL students to present stories of the Hmong people dramatically. The children helped choose stories to enact; developed and practiced their individual characters, dialogues, and scenes; and, with the help of the teacher, parents, and the community activist, organized their stage costumes, set, and props. The final dramatic presentation of these stories, in English and Hmong, was given to their entire elementary school and invited guests.

Our students often see a clear connection between their roles as storytellers and healers. While engaging in ethnographic interviewing with a Hmong parent, one of our students also played the role of the therapist by acknowledging both her informant's traumatic cultural struggle as well as her culture's strength:

Mee said, "The most difficult thing about my life was when my mom, my brothers and my own two children died and my husband leaving me. I was left alone not knowing how to do anything. ... I am waiting to die."

My heart hurt for her as I watched her cry. ... I asked her if she wanted me to stop interviewing her and shut off the tape recorder. She shook her head. ... Mee has great inner strength to have gone this far. ... I truly believe that what has made Mee strong mentally has been the strong Hmong tradition of self-reliance and independence. Her children and other Hmong people living close to her have given her support and other opportunities to help her survival in the United States. (Thompson, 1998, pp. 11-12)

Children of immigrants often grow apart from their parents as they adapt more quickly to the new language and culture and as they lose the ability to really communicate across generations. One of the goals of the family narrative project (described subsequently) was to provide a forum in which the stories of parents and other elders could be explicitly valued by teacher-researchers and to initiate a process of cultural therapy (Spindler & Spindler, 1994) between generations.

Many of the narratives reflect the emerging dialogue between the student-researchers and their informants. Through family interviews, our students became aware of the tremendous importance that minority parents place on education. Often, these parents have limited English skills and little formal education, yet they find ways to support their children's learning, as in the case of Xia, a Hmong mother:

The only formal education Xia had were the two years of adult school when she arrived in America. Because Xia lacks skills in English, her husband, who understands English better, attends the teacher-parent conferences. However, in supporting the children in other school related functions such as school awards night, movie nights, etc., it is Xia who attends with the children. Even though Xia is unable to carry on a full conversation in English, she is courageous and unintimidated by unknown situations. (Vang, Robinson, & Smith, 1998, p. 6)

Students reflected on our earlier discussion of the ethnocentric attitudes and values often present in parent education programs (Valdes, 1996) and were able to use these family narrative studies to begin to identify culturally sensitive ways to encourage families to support the educational experiences of their children.

Personal stories shared by participating families about life and death on the U.S.-Mexico border stand in stark contrast to the political furor over illegal immigration that has garnered so much media and policy attention. One of the Mexican mothers interviewed had to cross the border at Nogales separated from one of her children. From a McDonald's restaurant window, she watched her youngest child taken back across the border three times by la migra, the border patrol:

Four times she had to cross. She had the figure of a little saint in her hand. It gave me goose bumps. I felt very bad because she told me later, "Look, mommy! This helped me cross." I felt better with my older girls, but I also felt bad because I saw that they were so strong. They didn't cry. They didn't yell, nothing. I was happy that they didn't cry, but sometimes I feel bad because they were so young. (Lupe, interviewed in Serrano, Dryer, Fink, & Cortes, 1998, p. 24)

Although they are supportive of the cultural worker roles, some students, especially those who work with Latinos, take issue with the border guard descriptor and have suggested the term gatekeeper instead. Nevertheless, it becomes clear to our students that whichever term was chosen, there is the implication that one culture, one language, and one grand or metanarrative are seen as important: Students either assimilate to these, or they are shut out. On the other hand, the border crosser seeks to prepare students with the tools necessary to be heard in the dominant language and culture and, moreover, to use these tools to develop counternarratives--stories of their lives and communities that offer critical perspectives to those of the dominant culture. Students came to appreciate their own need to become cultural border crossers by listening to the real border-crossing experiences of families.

Students, sometimes accompanied by their participant coauthors, have arranged to present this family narrative research at inservices in their school districts, at community service organizations, and at professional conferences. The written narratives are disseminated through journals and edited volumes, as well. This narrative research project provides a forum for bilingual family members to share their stories and ideas and for preservice and inservice teachers to learn from the families they serve and to begin making contributions to the field of second-language practice and research.

With each day in New York, with each new session attended at the international Teachers of English to Speakers of Other Languages conference, Chan and Khou Lor, refugees from Laos, became more animated as they discussed ideas for the future. He wants to attend the conference next year, in Vancouver; he wants me to come back with him to Laos to help document some of the stories of the forested highlands; he wants to go to the university to get a degree in educational counseling, and his wife, Khou, appears ready to pursue licensure as a bilingual teacher. Along with Faye Van Damme, they also see themselves as writers of lives whose work will be of interest to others. After meeting with editors one evening, we return to find that firefighters have converged on our cheap, midtown Manhattan hotel to answer a false alarm. Khou makes sure to get some photographs "for the book."

Preservice and inservice teacher-researchers and their family participants appear to benefit greatly from the bilingual family research project. To improve the research process and provide guidance for other researchers, family and student participants have identified several questions to consider:

Legend for Chart:

A - Cultural Storyteller
B - Cultural Healer
C - Cultural Worker

A
                B
                C

Biographer: Uses her own autobiography
as a reflective tool for practice;
encourages students to tell and learn
from their own cultural life stories
(Ayers, 1989; Goodson, 1992; Paley,
1995).
      Therapist: Seeks ways to address
      multiple cultural conflicts faced by
      minority students/families; eases the
      transition into the dominant culture
      without sacrificing meaningful
      aspects of students' own culture
      (Spindler and Spindler, 1994).

      Border Guard: Transmits the
      metanarrative of the dominant culture
      (in the dominant language); seeks to
      prepare students with the cultural
      literacy they need to function in
      American society (Bloom, 1987;
      Hirsch,1988; Schlesinger, 1992).

Ethnographer: Collects, interprets, values,
and utilizes as part of the curriculum
stories/knowledge from variety of student
cultures (Delgado-Gaitan &
Trueba, 1991; Diaz Soto, 1997; Moll
and Greenberg, 1990).
      Trainer: Prepares students with literacy
      tools to be fluent in the language/culture
      of power, to adapt without necessarily
      assimilating (Delpit, 1987; Gee,
      1990; Gibson, 1988).

      Border Crosser:Critically engages with
      multiple cultural/linguistic communities;
      helps students develop their
      counternarratives to the dominant culture's
      metanarrative; actively works to
      create diverse democratic communities
      inside and outside the classroom
      (Giroux, 1997).

(a) The boundaries between these role descriptions are hazy at
best. For example, the ethnographer can also be seen as a
cultural healer and/or a cultural worker, and the border guard is
also a cultural storyteller.

??? Andrade, R. (1998). Life in elementary school: Children's ethnographic reflections. In A. Egan-Robertson & D. Bloome (Eds.), Students as researchers of culture and language in their own communities (pp. 93-114). Cresskill, NJ: Hampton.

??? Ayers, W. (1989). The good preschool teacher: Six teachers reflect on their lives. New York: Teachers College Press.

??? Bereiter, C., & Englemann, S. (1966). Teaching disadvantaged childen in preschool. Englewood Cliffs, NJ: Prentice Hall.

??? Bilingual Education Act (Title VII of Elementary and Secondary Education Act of 1968).

??? Bloom, A. (1987). The closing of the American mind. New York: Simon & Schuster.

??? Brisk, M. E. (1998). Bilingual education: From compensatory to quality schooling. Mahwah, NJ: Lawrence Erlbaum Associates, Inc.

??? Carger, C. (1996). Of borders and dreams: A Mexican-American experience of urban education. New York: Teachers College Press.

??? Delgado-Gaitan, C. (1996). Protean literacy: Extending the discourse on empowerment. New York: Falmer.

??? Delgado-Gaitan, C., & Trueba, H. (1991). Crossing cultural borders: Education for immigrant families in America. New York: Falmer.

??? Delpit, L. (1988). Power and pedagogy in educating other people's children. Harvard Educational Review, 58 (3), 280-298.

??? Denzin, N. (1994). The art and politics of interpretation. In N. Denzin & Y. Lincoln (Eds.), Handbook of qualitative research (pp. 500-515). Thousand Oaks, CA: Sage.

??? Diaz Soto, L. (1997). Language, culture, and power: Bilingual families and the struggle for quality education. Albany, NY: SUNY Press.

??? Gee, J. (1990). Social linguistics and literacies: Ideology in discourses. New York: Falmer.

??? Gibson, M. (1988). Adaptation without assimilation: Sikh immigrants in an American high school. Ithaca, NY: Cornell University Press.

??? Giroux, H. (1997). Pedagogy and the politics of hope: Theory, culture and schooling. Boulder, CO: Westview.

??? Goodson, I. (1992). Studying teachers' lives: An emergent field of inquiry. In I. Goodson (Ed.), Studying teachers' lives (pp. 1-17). New York: Teachers College Press.

??? Grinberg, J., & Goldfarb, K. P. (1998). Moving teacher education in/to the community. Theory into Practice, 37, 131-139.

??? Grumet, M. (1991). The politics of personal knowledge. In C. Witherell & N. Noddings (Eds.), Stories lives tell: Narrative and dialogue in education (pp. 67-77). New York: Teachers College Press.

??? Hirsch, E. (1988). Cultural literacy: What every American needs to know. New York: Vintage.

??? Hones, D., & Cha, C. S. (1999). Educating new Americans: Immigrant lives and learning. Mahwah, NJ: Lawrence Erlbaum Associates, Inc.

??? Kiedrowski, S. (1998). A Hmong family narrative. Unpublished manuscript, University of Wisconsin, Oshkosh, College of Education and Human Services.

??? Lewis, O. (1966). The culture of poverty. Scientific American, 215, 19-25.

??? Lincoln, Y. (1993). I and thou: Method, voice and roles in research with the silenced. In D. McLaughlin & W. Tierney (Eds.), Naming silenced lives: Personal narratives and processes of educational change (pp. 29-47). Boston: Routledge & Kegan Paul.

??? McBeth, S., & Horne, E. (1996). "I know who I am": The collaborative life history of a Shoshone Indian woman. In G. Etter-Lewis & M. Foster (Eds.), Unrelated kin: Race and gender in women's personal narratives (pp. 111-130). Boston: Routledge & Kegan Paul.

??? McCaleb, S. P. (1998). Connecting preservice teacher education to diverse communities: A focus on family literacy. Theory Into Practice, 37, 148-154.

??? McCandless, B. R. (1952). Environment and intelligence. American Journal of Mental Deficiency, 56, 674-691.

??? Moll, L. (1992). Bilingual classroom studies and community analysis: Some recent trends. Educational Researcher, 3 (3), 20-24.

??? Moll, L., & Greenberg, J. (1990). Creating zones of possibilities: Combining social contexts for instruction. In L. Moll (Ed.), Vygotsky and education: Instructional implications for applications of sociohistorical psychology (pp. 319-348). New York: Cambridge University Press.

??? Moll, L., Velez-Ibañez, C., & Greenberg, J. (1989). Year one progress report: Community knowledge and classroom practice: Combining resources for literacy instruction (IARP Subcontract No. L-10). Washington, DC: Development Associates.

??? Olmedo, I. M. (1997). Family oral histories for multicultural curriculum perspectives. Urban Education, 32, 45-62.

??? Paley, V. (1995). Kwaanza and me. Cambridge, MA: Harvard University Press.

??? Phelan, P., & Davidson, A. (1994). Looking across borders: Students' investigations of family, peer, and school worlds as cultural therapy. In G. Spindler & L. Spindler (Eds.), Pathways to cultural awareness: Cultural therapy with teachers and students (pp. 35-59). Thousand Oaks, CA: Corwin.

??? Rabinow, P., & Sullivan, W. (1987). The interpretive turn: A second look. In P. Rabinow & W. Sullivan (Eds.), Interpretive social science: A second look (pp. 1-30). Berkeley, CA: University of California Press.

??? Richardson, L. (1992). The consequences of poetic representation: Writing the other, rewriting the self. In C. Ellis & M. Flaherty (Eds.), Investigating subjectivity: Research on Lived experience (pp. 125-140). Newbury Park, CA: Sage.

??? Schlesinger, A. (1992). The disuniting of America. New York: Norton.

??? Serrano, L., Dryer, L., Fink, A., & Cortes, M. G. (1998). The Cortes family: An ethnographic study. Unpublished manuscript, University of Wisconsin, Oshkosh, College of Education and Human Services.

??? Soto, L. (1997). Language, culture, and power: Bilingual families and the struggle for quality education. Albany: State University of New York Press.

??? Spindler, G., & Spindler, L. (Eds.). (1994). Pathways to cultural awareness: Cultural therapy with teachers and students. Newbury Park, CA: Corwin.

??? Tedlock, D. (1983). The spoken word and the work of interpretation. Philadelphia: University of Pennsylvania Press.

??? Thompson, A. (1998). A better place for them: A Hmong family narrative. Unpublished manuscript, University of Wisconsin, Oshkosh, College of Education and Human Services.

??? Trueba, H. (1994). Foreword. In G. Spindler & L. Spindler (Eds.), Pathways to cultural awareness: Cultural therapy with teachers and students (pp. vii-xi). Thousand Oaks, CA: Corwin.

??? Valdes, G. (1996). Con respeto: Bridging the distances between culturally diverse families and schools. New York: Teachers College Press.

??? Van Damme, F. (1998). The story of a Hmong refugee. Unpublished manuscript, University of Wisconsin, Oshkosh, College of Education and Human Services.

??? Vang, C., Robinson, M., & Smith, E. (1998). A narrative study of the Thao family. Unpublished manuscript, University of Wisconsin, Oshkosh, College of Education and Human Services.

??? Villanueva, I. (1996). Change in the educational life of Chicano families across three generations. Education and Urban Society, 29, 13-34.

??? Walker-Moffat, W. (1995). The other side of the Asian American success story. San Francisco: Jossey-Bass.

??? Zeichner, K. (1993). Educating teachers for cultural diversity (NCRTL Special Report). East Lansing: National Center for Research on Teacher Learning, Michigan State University.