Improving Schooling for Language-Minority Children: A Research Agenda (1997)

Page 163

7—
Studies of School and
Classroom Effectiveness

Whereas Chapter 6 focuses on program evaluations, in which instructional language issues are paramount, this chapter focuses on empirical studies that attempt to identify school- and classroom-level factors related to effective schooling for English-language learners from early education programs through high school. Although instructional language issues are important in the research described in this chapter, they do not dominate. The research reviewed here is categorized according to four distinct methodologies: effective schools research, nominated schools research, prospective case studies, and quasi-experiments. Thus the chapter begins with a description of these methodologies. The chapter then summarizes the studies of each type reviewed and presents some observations on the features on these studies. Next is a detailed discussion of 13 attributes identified by the studies as being associated with effective schools and classrooms. The final section examines the methodological strengths and limitations of the four types of studies. Although different approaches to a review of this sort are possible, the present review is organized according to methodology because the committee believes the study findings should be viewed in light of the methodologies used to generate them. For example, some findings are richer in detail but less generalizable than others. Moreover, it is important to highlight the various methodologies because strengthening and integrating them is necessary if school and classroom research is to be improved.

State Of Knowledge

Beginning in the 1970s, and largely in response to findings by Coleman et al. (1966), Jencks et al. (1972), and others suggesting that differences in student

Page 164

outcomes were due largely to factors outside the control of schools, a group of studies appeared that challenged this conclusion by identifying effective schools and the characteristics that made them effective (e.g., Edmonds, 1979; Rutter et al., 1979; Weber, 1971; see especially Purkey and Smith, 1983). This research yielded what became (with some variations) a familiar list of "effective schools" characteristics, which included the following:

Despite early and ongoing criticism (e.g., Scott and Walberg, 1979; Stedman, 1985, 1987), effective schools research has evolved over the past two decades (Bliss et al., 1991), flourishing and even turning into a national movement. In terms of sheer numbers, it is now perhaps the most successful of the dozens of ideas informing school reform efforts nation-wide. According to Education Week (1995), more than 2,000 school districts—15 percent of the nation's 14,500—report using effective school research.1

In the 1990s, there has been a significant change in the way "effective" schools are identified, particularly in efforts to uncover effective schooling dimensions for English-language learners. Instead of designating schools as effective on the basis of measures of student learning or achievement, investigators now typically use a "nominated" schools design.2 As in the previous effective schools research, current investigators attempt to identify schools or programs that are ''exemplary." However, rather than being identified on the basis of outcome measures, schools are identified in accordance with the professional judgments of knowledgeable educators. Independent measures of student achievement are not in the data set reported by most of these investigators. In schools or classrooms with large numbers of English-language learners, this is often the case because investigators could not find adequate student achievement data to verify the validity of the nominations (Berman et al., 1992, 1995).3 However, in some instances, investigators have asked nominated schools to provide

1This figure might actually indicate diminished influence of the effective schools movement in the 1990s. A 1989 General Accounting Office survey estimated that 41 percent of U.S. school districts had programs based on effective schools research in 1988 (Bliss et al., 1991).

2However, there is still effective schools research that relies on student outcomes; moreover, not all research prior to the 1990s used student outcomes to determine school or classroom effectiveness.

3For example, Berman et al. (1992:6) found that "language proficiency results were of questionable validity, subject to sources of unreliability and not comparable; schools did not consistently assess LEP students on California assessment tests or had little accumulated data as a result of high transiency or poor attendance."

Page 165

information that corroborates their effectiveness4 or have attempted to verify the quality of nominees by examining "proxies" for student achievement in nominated classrooms, such as academic learning time (Tikunoff, 1983).

Prospective case studies and quasi-experiments represent a different approach to studying effective schooling. Instead of finding schools that are already "effective" or have been nominated as such, prospective studies attempt to document changes in school-wide programs or classrooms and the effects of these changes on student achievement. In the ideal situation, the changes are based on strong theory. In the discussion of program evaluation in Chapter 6, we note many problems with large-scale efforts that have provided very little "bang for the buck." Among the prescriptions suggested are small-scale evaluations of the implementation of theory-based programs. The prospective case study approach to studying school and classroom effectiveness comes close to this ideal.

An example of the prospective case study approach is the Case Studies in Bilingual Education project (Gold and Tempes, 1987), a collaborative effort during the mid-1980s between the California State Department of Education and five elementary schools serving large numbers of Spanish-speaking students. Applying theoretical models for the education of English-language learners and for the implementation of school-wide change, schools developed and carried out changes in curriculum, instruction, and organization to promote higher levels of academic achievement for Spanish-speaking English-language learners. Although the effects of the instructional program have been somewhat more mixed than is often reported (Samaniego and Eubank, 1991), the project appears to have been extremely successful overall and led to the development of a Title VII Academic Excellence Program, based on one of the California Case Study schools (Eastman), entitled Project MORE.

A quasi-experimental design (Cook and Campbell, 1979) is generally defined as a research design that approximates the control of randomized design. We define the term somewhat differently: a quasi-experimental design employs comparison schools or classrooms and measured outcomes. Thus our use of the term is meant to convey designs that afford stronger conclusions than those typically used in effective or nominated schools research or prospective case studies. Quasi-experimental studies also begin with a school(s) or classroom(s) that is not more effective and perhaps even less effective than one or more comparable schools or classrooms. Investigators and educators implement a

4 For example, to be considered for inclusion in the Descriptive Study of Significant Features of Exemplary Special Alternative Instructional Programs (Tikunoff, 1983), applicants had to describe their programs and provide evidence of exceptional student performance in some combination of the following areas across at least 2 successive years: relative gains in English-language proficiency, in academic performance, and in special language programs before students were exited, and extent to which grade promotion requirements were met while participating in or after exiting the program.

Page 166

change or intervention predicted to improve student outcomes. Student outcomes (pre and post) are measured at the target and comparison schools or classrooms to determine the effects of the change or intervention. Because of the existence of a comparison site, the quasi-experimental approach offers the strongest basis for claiming that what was done at a target site produced changes in student achievement.

Studies Reviewed

For this review of the literature on studies of school and classroom effectiveness, studies that met the following criteria were included: (1) the school population(s) or classroom(s) studied included substantial numbers of English-language learners, and (2) investigators made some attempt (or at least claim) to identify school- or classroom-level factors, including instruction, associated with positive outcomes (or good programs) for these students.5 Studies that examined the relationship between student knowledge/skills and task demands on literacy and learning are reviewed in Chapter 3. In general, these studies did not examine school and classroom factors that promote learning; rather, they focused on student and task attributes and their relationship to learning.

Other school-based efforts besides those examined here have aimed at improving outcomes for English-language learners, but there is insufficient information about them to permit an analytic review. For example, the Academic Excellence Program (funded by Title VII) identifies effective programs serving English-language learners. To be designated an Academic Excellence Program, a program must provide evidence that it has improved outcomes for these students and propose a plan for disseminating the program to other schools around the country. The Academic Excellence Programs represent a diverse array of curricular and instructional approaches that include, for example, content-based English as a second language (ESL), computer-assisted writing, two-way bilingual education, gifted and talented education, transitional bilingual education, programs for recent immigrants, and interactive computer technologies. Dissemination

5 A broad search was done to locate relevant articles and books for this review: The search focused initially on Education Resources Information Center (ERIC) documents dating back to 1985, using limiters relevant to the topics of interest. Very few studies were located, and as a result, indexes from the following journals were searched back to 1985 for relevant studies: Educational Researcher, TESOL Quarterly, Journal of Educational Issues of Language Minority Students, Journal of the National Association for Bilingual Education, Urban Education, Language and Education, Equity and Choice, American Educational Research Journal, Review of Educational Research, Educational Leadership, Harvard Educational Review, Applied Linguistics, Bilingual Research Journal, Read Perspectives, and Journal of Reading Behavior. All reference lists from useful retrieved documents were checked for additional sources, and an effort was made to obtain relevant studies. Finally, experts on the education of English-language learners were consulted regarding books and reports that might be of interest.

Page 167

efforts generally involve school-wide adoption (e.g., awareness, training, technical assistance, follow-up; see Wilson et al., 1994) or "restructuring" (Wilson et al., 1994). Thus, the programs as originally developed and their dissemination to new sites probably involve school- and classroom-level factors that would be relevant to this review. At the moment, however, such information is not readily available and therefore could not be incorporated here.

In addition, there are many national school reform networks. We reviewed the work of 13 of these projects6 and found that very few have provided empirical evidence on issues of successful schooling for English-language learners (see Chapter 10). In part this is because many of these projects remain unevaluated or do not specifically examine outcomes for these students. For example, Chasin and Levin (1995) provide a case study of an "accelerated school" (elementary level) where 13 different languages are spoken, but they do not report the English-learning status of English-language learners, address concerns that are specific to these students' educational experiences, or report changes in outcomes for these students.

As a result of our literature search, we identified reports of 33 studies for inclusion in this review; these studies and reports are identified in Annex Table 7-1. Most of the studies fall into one of four design categories (Annex Table 7-1, column 4): effective schools/classrooms design (6 studies); nominated schools/classrooms design (7 studies); prospective case study design (5 studies); and quasi-experimental or experimental design (13 studies). There are also 2 studies that do not fall into a design category.

Effective Schools Research

The basic design and logic of effective schools research still inform efforts to discover principles or processes that can be used to improve schooling opportunities and outcomes for "at-risk" students. However, the design has become more of a hybrid, relying on both student outcomes and nomination. More recent studies examine attributes of effective classrooms rather than schools. Some of these studies are reviewed in this section. Although most of these studies do not focus on English-language learners, our review found some studies involving such students: one pure effective schools study (Carter and Chatfield, 1986), three studies that rely on both nomination and student outcomes (Edelsky et al., 1983; Garcia, 1990a; Moll, 1988), and two studies (Mace-Matluck et al., 1989;

6These networks include the Accelerated Schools Project, Center for Educational Renewal, Coalition of Essential Schools, Core Knowledge Foundation, Effective Schools Networks, National Board for Professional Teaching Standards, National Paideia Center, New American Schools Development Corporation, New Standards Project, School Development Program ("Comer Schools"), and Success for All.

Page 168

Wong Fillmore et al., 1985) that examine the relationship between current schooling practices and the language and/or reading achievement of students who began their instruction in bilingual education classrooms. We include the latter studies in this category because their goal is to identify attributes of effective classroom instruction by examining correlations among student background information, instructional practices, and student outcomes.

Nominated Schools Research

Nominated schools studies included in this review are Lucas et al. (1990), Berman et al. (1992, 1995), and Tikunoff et al. (1991). Nominated classrooms studies include Tikunoff (1983), Pease-Alvarez et al. (1991), and Gersten (1996). A study that explores the replicability and stability of features identified in the initial study of nominated classrooms (Tikunoff, 1983) is also reviewed.

Prospective Case Studies

We reviewed five prospective case studies that document student change as a result of a theory-driven intervention. One is the Case Studies in Bilingual Education project (Gold and Tempes, 1987) mentioned earlier. A second explores the extent to which reciprocal teaching of question generation, summarizing, and predicting using students' primary language improves reading comprehension in that language; it also explores how these strategies are used in second-language reading (Hernandez, 1991). A third study examines the effects of a collaborative inquiry approach to science on learning by language-minority students (Rosebery et al., 1992). A fourth study examines strategies used by classroom teachers to facilitate students' comprehension of subject matter and improve their academic language skills (Short, 1994). A fifth (Cohen, 1984) examines the effect of status on peer interaction in activity centers structured to promote science and math learning; it also investigates the relationship between peer interaction and learning.

Quasi-Experimental Research

Our review includes thirteen examples of quasi-experimental or experimental research on school- and classroom-level factors associated with schooling outcomes for English-language learners. Five of these studies are adaptations of Success for All for English-language learners. Success for All is an intervention program from the Johns Hopkins Center for Research on the Education of Students Placed at Risk that focuses on helping students attain high levels of reading proficiency in elementary school. The Success for All studies are reported in Dianda and Flaherty (1995), Slavin and Madden (1994, 1995), Slavin and

Page 169

Yampolsky (1992), and Calderon et al. (1996).7 Although four of the Success for All studies could have been reviewed here as one (as synthesized and reported in Slavin and Madden, 1995), they are in fact different studies replicated in different school contexts and settings. Two other quasi-experimental studies included here (Goldenberg and Gallimore, 1991; Goldenberg and Sullivan, 1994) are also fairly detailed case studies. Each describes the issues and dynamics of change at a single elementary school with a large Latino population, and each makes pre-post comparisons of student achievement with that in comparable schools in the district.

Finally, five quasi-experimental studies and an experimental study examine the effects of classroom interventions on English-language learners through use of comparison and control groups. The first quasi-experimental study explores the effect of metacognitive reading strategy training on the reading performance and student reading analysis strategies of third grade bilingual students (Muniz-Swicegood, 1994). The second examines the effect of thematically integrated mathematics instruction on achievement, attitudes, and motivation in mathematics among school students of Mexican descent, many of whom have limited English proficiency (Henderson and Landesman, 1992). The third explores the effect of curriculum content and explicit teaching of learning strategies on students' metacognitive awareness and their learning of content and knowledge (Chamot et al., 1992). The fourth investigates effective strategies for teaching literature to transition students (Saunders et al., 1996). The fifth investigates the effectiveness of a Spanish version of reading recovery entitled Descubriendo La Lectura (Escamilla, 1994). The experimental study investigates the effects of a "culturally appropriate" reading program on the reading performance of Hawaiian children (Tharp, 1982).

Other Studies

Two studies do not fall in any of the above categories. One (Minicucci and Olsen, 1992) is an exploratory study that examines 27 secondary schools; its purpose is descriptive. The other study (Fisher et al., 1983) explores the replicability and stability of features identified in the initial study of nominated classrooms (Tikunoff, 1983). It assesses replicability by studying a second sample of classrooms (89 at 8 sites) serving different ethnolinguistic groups, as well as by examining classrooms that have not been nominated as successful. It examines the stability of the instructional process by studying teachers and students for a second academic year in different settings.

7The Calderon et al. (1996) study encompasses only one aspect of Success for All—Bilingual Cooperative Integrated Reading and Composition (BCIRC).

Page 170

Observations on Studies of Effectiveness

Annex Table 7-1 identifies the important features of the 33 studies, including the school- and classroom-level attributes investigators claim are related to effective, or exemplary, schooling for English-language learners (as discussed in the next section). Several general observations can be made about this collection.

First, this is a heterogeneous group of studies representing levels of schooling from prekindergarten to high school, employing at least four different types of designs (as discussed above), and ranging from single-classroom and -school studies to a study of nine different "exemplary programs" in a total of 39 schools. By far the greatest number of schools has been studied within the nominated schools research design.

Second, school- and classroom-level factors associated with varying outcomes for English-language learners have received less attention than have other areas of research on these students. Clearly, the issue of language of instruction (whether English-language learners should be taught in their native language, and if so, to what extent) has dominated the research agenda (see Chapter 6). There have also been qualitative and ethnographic studies that have examined social context, language distribution, classroom interaction, and sociocultural enactments of classroom pedagogy (see Chapters 2, 3, and 4). Although these studies provide rich descriptions of educational environments, many do not relate practice to learning outcomes.

Third, although many non-English languages found in U.S. schools appear to be represented in these studies, by far the most commonly found is Spanish. This of course reflects the reality that approximately three-fourths of English-language learners are Spanish speaking. Most of the studies were conducted in schools that were predominately Latino. However, some sites within larger studies had substantial numbers of non-Spanish-speaking English-language learners. Only a few studies—Slavin and Yampolsky (1992) (Asian), Wong Fillmore et al. (1985) (Chinese), Rosebery et al. (1992) (Haitian-Creole), and Tharp (1982) (Hawaiian)—targeted non-Spanish-speaking English-language learners.

Fourth, as previously mentioned, by far the greatest number of schools and classrooms studied have been within the nominated schools design. These studies, as well as a few in the other categories, do not report student achievement data.8 The absence of outcome data does not mean that a study is uninformative.

8In their report on the California Case Studies, Gold and Tempes (1987:7) explicitly state that their project "was not designed as an experiment" and that they "carefully avoided efforts to set up premature or unreasonable comparisons." However, achievement data on the California Case Studies have been reported in various papers and publications (e.g., Krashen and Biber, 1988). Samaniego and Eubank (1991) conducted a more objective and rigorous secondary analysis of achievement data at four of the five sites. Three other studies included in this review (Lucas et al., 1990; Tikunoff, 1983; Tikunoff et al., 1991) report that some indicators of student outcomes informed the selection of the "effective" or ''exemplary" sites, but neither these data nor the criteria used by investigators are reported. Of the remaining studies, one was exclusively exploratory (Minicucci and Olsen, 1992) and makes no claim of trying to explain how effective programs came to be; the studies by Berman et al. (1992, 1995), Pease-Alvarez et al. (1991), and Gersten (1996) neither report outcome data nor apparently used student outcomes to inform the selection of nominated sites. With the exception of Short (1994), which is more of an exploratory study, the prospective and quasi-experimental studies report student outcome data.

Page 171

Indeed, these studies are filled with interesting and useful data about programs, staff, students, community, and, more generally, the very complex and challenging circumstances in which students and teachers must function. They also provide what in many cases are highly compelling accounts of dedicated educators working to create engaging, meaningful, and responsive settings for student learning. However, they do not link these settings to indicators of student outcomes, at least not in any explicit way.

Finally, as noted above, these studies report a wide range of school- and classroom-level attributes related to effectiveness (see columns 7 and 8 of Annex Table 7-1). These attributes, summarized in the following section, can be conceptualized and categorized in many different ways. It is important to keep in mind, however, that the attributes discussed here represent concepts refracted through at least two sets of lenses (the original investigators' and this committee's), that the empirical bases for making strong causal claims vary considerably and are sometimes unknown, and that there are caveats associated with some of the attributes. For example, different attributes may be more or less important for different age groups or different ethnic groups. Therefore, none of these individual attributes should be considered necessary or sufficient conditions for the schooling of English-language learners.

Attributes of Effective Schools and Classrooms

Based on the findings of the 33 studies reviewed, effective schools and classrooms have the following attributes9: a supportive school-wide climate, school leadership, a customized learning environment, articulation and coordination within and between schools, some use of native language and culture in the instruction of language-minority students, a balanced curriculum that incorporates both basic and higher-order skills, explicit skills instruction, opportunities for student-directed activities, use of instructional strategies that enhance understanding, opportunities for practice, systematic student assessment, staff development, and home and parent involvement. Each of these attributes is discussed in the following subsections.

9Note that not all studies include all attributes, but the general attributes appear in many of the studies.

Page 172

Supportive School-wide Climate

Teachers', students', and parents' beliefs, assumptions, and expectations for themselves and for each other probably exert a powerful influence on student learning opportunities and student outcomes (Rutter et al., 1979). It is not surprising, then, that a supportive school-wide climate, sometimes called school "ethos," is an attribute of effective schools for English-language learners. Such a climate is explicitly cited or can easily be inferred from almost all the studies reviewed.

Carter and Chatfield (1986), Moll (1988), Lucas et al. (1990), Tikunoff (1983), Tikunoff et al. (1991), Berman et al. (1992, 1995), and Minicucci and Olsen (1992) report that a positive school-wide climate was a feature of the effective or exemplary schools they studied. The schools varied in their particular manifestations of such a climate, but overall emphasized three things—value placed on the linguistic and cultural background of English-language learners, high expectations for their academic achievement, and their integral involvement in the overall school operation. The schools studied by Lucas et al. (1990:8) "celebrated diversity." For example, although they made English literacy a primary goal, they also encouraged students to enhance their native-language skills in classes for those students who spoke Spanish. Moreover, a number of teachers and counselors had made an effort to learn to speak Spanish. Moll (1988:467) notes that "in contrast to the assumption that working-class children cannot handle an academically rigorous curriculum, or in the case of limited English proficient students, that their lack of English fluency justifies an emphasis on low-level skills, the guiding assumption in the [effective] classrooms analyzed seemed to be the opposite: that the students were as smart as allowed by the curriculum." One school studied by Berman et al. (1995) had a house structure whereby each house was named for a California State University campus with which it forged a partnership—but the continuously reinforced message was that high levels of learning and achievement were expected of all students. Integral involvement of English-language learners also characterizes effective schools and classrooms. Berman et al. (1995) found that school restructuring enabled the exemplary schools to design and adapt programs that best suited the needs of English-language learners—and all students.

How does a school climate, or ethos, change from being "not conducive" to being "conducive" to high levels of achievement for English-language learners? Unfortunately, the studies do not offer much guidance here. Gold and Tempes (1987) report that teachers at their case study sites received training in Teacher Expectations and Student Achievement, a program designed to boost teachers' expectations for their students, which in turn is assumed to boost student achievement. Gold and Tempes also report that steps were taken to improve the perceived status of language-minority students—administrator and teacher support for use of Spanish at school, cross-cultural activities, and cooperative learning.

Page 173

Yet we do not know whether and how these attempts to influence school climate directly affected the climate, in turn affecting achievement. Only Goldenberg and Sullivan (1994) address this question directly and prospectively. They claim that changes in school climate were the result of a complex process aimed at improving student achievement, begun by identifying school-side goals and expectations for students, followed by consistent, visible, multiple, and long-term efforts to work toward those goals. Teachers responded positively to the more meaningful and substantive focus at the school.

Although the logic of attempting to change school climate through staff development and training to improve student achievement is supported by research on teacher expectations, an alternative hypothesis may merit attention: that school climate is at least as much a reflection of student achievement as an influence on it (Jussim, 1986). In other words, it may be that teachers hold high expectations when they have students who achieve, and conversely that they hold low expectations when students do not achieve. If this formulation is valid, it suggests that one important way to raise teacher expectations is to raise student achievement by creating structures at a school and helping teachers acquire skills and knowledge needed to be more successful with students, rather than by exhorting teachers to raise their expectations. Goldenberg and Gallimore (1991), for example, report that first grade reading expectations at the school they studied seemed to increase as a result of changes in first grade reading achievement, not as a result of training to raise expectations. Comer (1980) also describes how improved expectations followed the establishment of successful practices, which in turn raised expectations.

School Leadership

Consistent with findings of the effective schools research that began two decades ago, school-level leadership appears to be a critical dimension of effective schooling for English-language learners. At least half of the studies reviewed name leadership, often the principal's, as an important factor; the role of leadership can also be inferred from several of the other studies that do not explicitly cite it. A clear statement of the role of leadership comes from Tikunoff et al. (1991:10): "Without exception, exemplary SAIPs [Special Alternative Instructional Programs] came about because someone assumed leadership for planning, coordinating, and administering the programs." Both Carter and Chatfield (1986) and Lucas et al. (1990) name the principal's leadership as one of the elements that helps explain the effective or successful schools they studied. The principal is seen as playing a key role in many ways, for example, making the achievement of English-language learners a priority, providing ongoing direction and monitoring of curricular and instructional improvement, recruiting and keeping talented and dedicated staff, involving the entire staff in improvement efforts, and providing a good physical and social environment. Goldenberg and

Page 174

Sullivan's (1994) model identifies "leadership" as one of the four change elements; in this model, there are two crucial dimensions of leadership that help propel and maintain change—providing support and exerting pressure.

An important exception can be found in the Success for All studies, which do not name leadership as an important attribute. In fact, in another report (Slavin et al., 1995), Slavin and associates seem to suggest that, except for the operational role apparently played by the program facilitator, school-level leadership may not be particularly critical for success. Despite the fact that several schools discontinued use of the Success for All model following a change in principal, Slavin et al. say that many other schools "have survived changes of superintendents, principals, facilitator, and other key staff…" (p. 30). The implication is that personal leadership is far less critical for school reform and improvement than is having an effective program in place.

This position contrasts with that of Carter and Chatfield (1986). In discussing the implications of their effective schools study, they comment that school districts should reexamine policies requiring principals to move every few years. They give the example of an effective principal who was transferred because of such a policy. "If it is true (and we think it is) that three years is the minimum amount of time required to approach [school] effectiveness, then such policy should be questioned" (p. 230). The Success for All data would suggest that if an effective program is in place, principal changes (presumably within some limits) should not matter.

More generally, Success for All is an important exception to much of the literature on school change since its developers make strong claims about the exportability of the program, which has very specific materials, manuals, and structures, to a wide range of schools, students, staffs, and communities. Indeed, the program seems to stand independently of the personnel called upon to implement it. This stance is atypical of the school change literature as a whole, and some suggest that the Success for All program does not require a strong principal because leadership comes from a charismatic and dynamic individual outside the school (Robert Slavin, the developer of Success for All), teachers provide this leadership, and the program is very structured and limited to language arts and reading instruction. (We return to this issue later in the chapter.)

Customized Learning Environment

Staff in effective schools and classrooms design the learning environment to reflect school and community contextual factors and goals while meeting the diverse needs of their students. Many researchers have noted that there is no one right way to educate language-minority students; different approaches are necessary because of the great diversity of conditions faced by schools. They recommend that local staff and community members identify the conditions under which one or some combination of approaches is best suited and then adapt

Page 175

models to match their particular circumstances. Berman et al. (1992) state that they cannot identify which of the major approaches to educating English-language learners is most effective under all conditions and claim that different approaches are necessary because of the great diversity of conditions. Their 1995 study reports that "rather than using a single model for all LEP students, teachers adjusted curriculum, instruction, and use of primary language to meet the varying needs of students" (p. 13). Tikunoff et al. (1991) found that the form of an exemplary Structured Alternative Instructional Program and the nature of its success build upon and are influenced by its context. Moll (1988) observes that effective teachers hold similar views about teaching, but create their own instructional programs that are attuned to the needs of their students. Samaniego and Eubank (1991) found the same to be the case in verifying the California Case Studies. In their study of effective secondary schools, Lucas et al. (1990) found that language-minority students are more likely to achieve when a school's curriculum responds to their individual and differing needs by offering variety in three areas: the skills, abilities, and knowledge classes are designed to develop (i.e., native-language development, ESL, subject matter knowledge); the degrees of difficulty and sophistication among available classes (i.e., advanced as well as low-level classes); and the approaches to teaching content (i.e., native-language instruction, content ESL, and specially designed instruction in English). Berman et al. (1995) found that successful schools also plan for the needs of newcomers (newly arrived students who immigrated to this country after the early elementary grades) and include in the design of their programs strategies to meet their needs.

Articulation and Coordination Within and Between Schools

Effective schools are characterized by a smooth transition between levels of language development classes (e.g., between content-based ESL and sheltered instruction) and coordination and articulation between special second-language programs and other school programs, as well as between levels of schooling. Short (1994) found collaboration between language and content teachers that involved identifying the language and academic difficulties and demands of particular subjects for English-language learners, ensuring close articulation between program components, and integrating ESL and content area instruction. Slavin and Yampolsky (1992) describe the central concept underlying the Success for All program as all of the school's personnel working together to ensure the success of every child; this includes ESL teachers, who teach reading and closely integrate instruction in English with the requirements for success in the regular program, especially reading. Minicucci and Olsen (1992) document coordination and articulation between the ESL/bilingual education department and other departments and between different grade levels. Berman et al. (1995) found that effective transition from special language instruction to mainstream classes was gradual, carefully planned, and supported with activities designed to

Page 176

ensure students' success at mastering complex content in English. Saunders et al. (1996) found that their strongest evaluation results came from project schools where students were exposed to the programs' instructional components beginning in second grade and then participated in two years of transition activities to prepare them for English-only classrooms; this included explicit connections between learning English and students' prior learning and experiences in Spanish. Calderon et al. (1996) combined Cooperative Integrated Reading and Composition (CIRC) strategies (e.g., heterogeneous learning teams that work together to help each other learn academic material) with innovative transitional and ESL strategies as students began to transition from Spanish to English reading.

Use of Native Language and Culture

The advantages of native-language use are a prominent theme among these studies, either explicitly (e.g., Henderson and Landesman, 1992; Hernandez, 1991; Muniz-Swicegood, 1994; Lucas et al., 1990; Berman et al., 1995; Rosebery et al., 1992, Tikunoff, 1983; Pease-Alvarez et al., 1991; Calderon et al., 1996) or implicitly (Carter and Chatfield, 1986, and Goldenberg and Sullivan, 1994, both of which took place in school settings where there was a firm commitment to bilingual education). Even those studies that report on Special Alternative Instructional Programs, where most instruction takes place in English, cite teachers' use of students' native languages to clarify and elaborate on points they are making in English (Tikunoff et al., 1991). Moreover, findings from a study of nine Special Alternative Instructional Programs (Lucas and Katz, 1994:545) indicate that even in exemplary programs designed to provide instruction primarily in English, the classrooms were "multilingual environments in which students' native languages served a multitude of purposes and functions. Across sites, native language use emerged as a persistent and key instructional strategy realized in very site-specific ways."

Mace-Matluck et al. (1989:209) suggest that correlation patterns between English and Spanish reading measures indicate that "a child's knowledge and skills associated with decoding are related across the two languages, as are those associated with overall reading ability, but to a lesser degree." As a result, they conclude that "reading is a single process and that reading knowledge and skills gained in one language can be transferred, if the necessary conditions are met, to reading in another known language." Further, they state that "the practice of teaching children to read initially in their stronger language appears to be educationally sound." However, they cite Moll et al. (1981), who caution that learning is primarily situation specific; generalizability to other situations depends on whether the environment is organized to provide similar features that will facilitate its applicability to a different setting.

Nevertheless, several sites examined in these studies do not feature native-language programs. One of the Success for All sites, for example, has a largely

Page 177

Asian population, and instruction is all in English. In addition, while some of the Spanish-speaking students in the Success for All studies are in primary-language programs, some are in sheltered English programs. Success for All has significant and important effects on the achievement of English-language learners, regardless of whether they are in a primary-language or sheltered English program. However, Spanish-speaking students in the Spanish-language Success for All program do better when tested in Spanish than do Spanish-speaking students in the English-language Success for All program when tested in English (to the extent achievement across two languages can be directly compared). This is not surprising, since we would expect that in the short run, reading achievement in one's native language would be superior to reading achievement in a second language, holding constant the instructional program.

We do not yet know whether there will be long-term advantages or disadvantages to initial literacy instruction in the primary language versus English, given a very high-quality program of known effectiveness in both cases. The question will not likely be resolved by the Success for All studies since there were probably pre-existing differences between students in the primary-language and sheltered English programs. Although the Success for All studies are unlikely to contribute to the debate on language of instruction, however, they do show that this program model is highly effective in both primary-language and sheltered English/ESL contexts.

Similarly, most of the studies cited in this review can contribute little direct knowledge to important questions about adapting instructional programs to students' home culture (e.g., sociolinguistic patterns, cognitive styles). These studies take place in contexts where the students' home culture is valued and seen as a resource to build upon, rather than a liability to remediate. Most of the studies report some aspect of home culture validation, accommodation, or inclusion in their effective sites. For example, Wong Fillmore et al. (1985) found that instructional practices and settings work differently for different groups of children and that conditions and experiences must be tailored to the characteristics of each group. Tharp (1982) found that small-group reading lessons structured to capitalize on the pre-existing cognitive and linguistic abilities of Hawaiian children were successful in teaching the children to read. He attributes some of the success of the program to the fact that the reading lessons resembled a major speech event in Hawaiian culture—talk story (see Chapter 4).

Again, Success for All presents a challenging counterpoint. There is nothing in the Success for All literature indicating that cultural validation or cultural accommodation per se is an important element of the program or, indeed, that culture plays any direct role at all (aside from language). Success for All is an intensive, prescriptive, well-conceptualized program designed to help as many children as possible leave third grade reading at grade level. Except for the important language adaptation (including somewhat different strategies for teaching Spanish and English reading), the program is no different for African American

Page 178

students in Baltimore than for Latinos in bilingual education or sheltered English programs in California or for Cambodians in an ESL program in Philadelphia (Slavin et al., in press). Of course, it is possible that cultural adaptations were taking place in the Success for All schools studied (as a result of the programs or not), but this factor was not examined.

Thus, the studies reviewed do not answer a question that has dominated research and professional and public discourse about educating English-language learners: What role should home language and culture play in the education of these students? The studies reviewed here can, at best, make an oblique contribution to this debate, in part because there are no rigorous studies that have controlled for interactions among student background (e.g., prior schooling in the native language, age), ways in which the first and second languages are used, and other instructional variables (e.g., overall quality of schooling). To illustrate the complexity of this issue, we cite a study on reading and bilingual students (Mace-Matluck et al., 1989) in which enrollment in Spanish reading programs was generally found to be negatively associated with acquired English literacy skills, but much of the relationship was due to entry-level differences in oral language skills. However, there was some indication of relatively superior English literacy skills at fourth grade exit for those students with longer enrollments in such Spanish reading programs. See Chapters 5 and 6 for further discussion of assessment and methodological issues.

Balanced Curriculum

Some schools focus primarily on curriculum "beyond basic skills." For example, the 1995 Berman et al. study features schools (grades 4-8) that emphasize "meaning-centered thematic curriculum." In contrast with Carter and Chatfield's (1986) classic "effective schools" description, Berman et al. (1995) do not even mention basic skills, objectives, and testing. Pease-Alvarez et al. (1991) note that the effective kindergarten teachers they studied involved their students in a wide range of meaningful activities or experiences focused on a particular concept, rather than exclusively emphasizing basic skills.

In much of the quasi-experimental research, however, classroom teachers combine basic and higher-order skills. In the Success for All schools, there is a balance between instruction in basic and higher-order skills at all grade levels. Story Telling and Retelling (STaR) is used in prekindergarten, in kindergarten, and early in first grade. Beyond the Basics is the name of the program in grades 2-6. The programs focus on comprehension, thinking skills, fluency, pleasure reading, and the use of increasingly complex material. Cooperative learning is used throughout the grades. Success for All's strong outcomes make the balance of these two levels of instruction very compelling. Both Goldenberg and Gallimore (1991) and Goldenberg and Sullivan (1994) report that the schools they worked with and studied included a "balanced" literacy program in which

Page 179

key skills and subjects such as phonics, word recognition, specific comprehension skills, and writing conventions were taught. However, they argue that early reading achievement improved at those schools partly because teachers incorporated language and meaning-based approaches into a system that had previously relied on basic decoding skills as the only avenue for learning to read.

Explicit Skills Instruction

The studies reviewed indicate that effective teachers for English-language learners use explicit skills instruction for certain tasks, mostly (though not always) to help students acquire basic skills.10 Wong Fillmore et al. (1985) suggest that quality explicit skills instruction is important for all students, but especially for Hispanic students. In the Significant Bilingual Instructional Features study (Tikunoff, 1983), a measure of effectiveness is active teaching, defined in part as "instruction, in which the teacher sets and articulates learning goals, actively assesses student progress, and frequently makes class presentations, illustrating how to do assigned work" (p. 4). Many of the studies that report actual student achievement (Carter and Chatfield's [1986] effective schools study; Goldenberg and Gallimore [1991]; Escamilla [1994]; and Goldenberg and Sullivan's [1994] and Slavin and Yampolsky's [1992] quasi-experiments) also report that the schools involved had substantial amounts of time available for explicit skills instruction.

Opportunities for Student-Directed Activities

The studies reviewed indicate that teachers supplement explicit skills instruction, characteristic of the initial effective schools research, with student-directed activities. Berman et al. (1995) found that effective teachers provide English-language learners with adequate opportunities to produce oral and written English, and emphasize an exchange of ideas in an intellectual conversation. Moll (1988:466-467) observes that teachers nominated as effective emphasize "the creation of classroom contexts in which children learn to use, try out, and

10The value of explicit skills instruction is corroborated by other researchers. According to Sternberg (1986), explicit skills instruction is highly effective for some tasks (i.e., teaching subject matter knowledge, knowledge of hierarchical relationships among bits of information, and knowledge of valid strategies in science, and enhancing beginning readers' ability to decode and use process strategies [e.g., summarize, clarify, question, predict] so that they better comprehend what they have read). Executive processes such as comprehension monitoring can also be taught through explicit skills instruction if developmentally appropriate for the student. Rosenshine and Stevens (1986) argue that explicit teaching is highly effective for well-structured skill and knowledge domains such as math computation, explicit reading comprehension strategies, map reading, and decoding.

Page 180

manipulate language in the service of making sense or creating meaning. The role of the teacher is to provide the necessary support and guidance so that children through their own efforts assume full control of the purposes and uses of oral language.'' Pease-Alvarez et al. (1991) found that effective kindergarten teachers encourage collaborative/cooperative interactions among students. Rosebery et al. (1992) and Henderson and Landesman (1992) also used a collaborative inquiry approach to successfully teach middle and high school language-minority students science and math, respectively. Cohen (1984) implemented a bilingual curriculum combined with complex instruction—a method of small-group learning featuring open-ended discovery on conceptual tasks.

Other prospective and quasi-experimental studies demonstrate the effectiveness of interventions that combine explicit skills instruction and student-directed work. Muniz-Swicegood (1994) used teacher modeling of comprehension strategies, teacher activation of student prior knowledge before reading, and a gradual shift of responsibility to students for carrying on the activities of the teacher to teach third grade Spanish-dominant students metacognitive reading strategies. Hernandez (1991) used a similar intervention to help Spanish-speaking seventh grade students improve their reading comprehension. In the Success for All studies, children were explicitly taught letter sounds, sound blending, word recognition skills, writing skills, and comprehension (metacognitive) strategies. However, they were also engaged in student-directed activity, such as cooperative learning and peer tutoring. CIRC and Bilingual CIRC (Calderon et al., 1996) combine explicit skills instruction in reading comprehension with cooperative learning, partner reading, and checking. The same is true for Saunders et al. (1996:30): "Like teachers described elsewhere (Gersten, 1996), teachers and advisors in our project saw the need to be comprehensive, to synthesize across rather than put in opposition various approaches to teaching and learning (directed lessons and instructional conversation, literature, and basals, writing projects and dictations)."

Instructional Strategies That Enhance Understanding

Effective teachers of English-language learners use specially tailored strategies to enhance understanding. This is important for students who are instructed in their second language. Dianda and Flaherty (1995:8) find that providing students with metacognitive skills they can use to "think about and prepare for a task, monitor themselves as they complete the task, and evaluate the outcomes helps language-minority students deal with context-reduced tasks." The Success for All reading and language arts program teaches students why, when, and how to use metacognitive strategies. Muniz-Swicegood (1994) employs instruction in self-generated questioning strategies; Hernandez (1991) teaches and models comprehension strategies, such as question generating, summarizing, and predicting, to Spanish-speaking students, thus significantly improving their Spanish reading

Page 181

comprehension; Chamot et al. (1992) use explicit instruction in learning strategies.

The use of routines to minimize the dependence on language is also helpful. Edelsky et al. (1983) note that the effective teacher they observed used routines for such purposes as writing projects, literature study, conferencing procedures, and science experiments. Sometimes she used written cues to make part of a routine or process explicit. Calderon et al. (1996) postulate that because students learned the CIRC process, protocols, and reading and learning strategies in Spanish first, these were easily transferable to the ESL content.

The studies reviewed also note strategies to help make instruction comprehensible to English-language learners: adjusting the level of English vocabulary and structure so it is appropriate for the students given their current level of proficiency in English; using explicit discourse markers such as "first" and "next"; calling attention to the language in the course of using it; using the language in ways that reveal its structure; providing explicit discussion of vocabulary and structure; explaining and in some cases demonstrating what students will be doing or experiencing; providing students with appropriate background knowledge; building on students' previous knowledge and understanding to establish a connection between personal experience and the subject matter they are learning; and using manipulatives, pictures, objects, and film related to the subject matter (Wong Fillmore et al., 1985; Gersten, 1996; Mace-Matluck et al., 1989; Saunders et al., 1996; Short, 1994).

Opportunities for Practice

This attribute entails building redundancy into activities, giving English-language learners opportunities to interact with fluent English-speaking peers, and providing opportunities for extended dialogue. Saunders et al. (1996:15) help students "work the text," which means studying it carefully—reading it, rereading it, discussing it, writing about it, and listening to what others have written about it. CIRC (Calderon et al., 1996) uses a set of activities that take place before, during, and after reading to ensure that students understand the text profoundly. These activities include, for example, the building of background knowledge and vocabulary, the making of predictions, teacher and then student reading of the same selection, discussion of answers to key questions, story mapping, story retelling, and story-related writing.

Through interactions with native speakers, second-language learners gain access to language that is unavailable in traditional teacher-directed classroom settings. Berman et al. (1995) report that exemplary schools provided opportunities for contact between monolingual English speakers and English-language learners during instruction in core content, in electives, or in alternative activities such as projects. Wong Fillmore et al. (1985) note that all learners profit from opportunities to interact with peers who speak the target languages, but Hispanic

Page 182

(rather than Chinese) students profit in particular; Chinese learners profit from opportunities for interaction with peers after they have reached intermediate levels of English proficiency. CIRC (Calderon et al., 1996) provides opportunities for English-language learners to interact with peers, which helps students develop fluency in and comfort with English.

Effective teachers create opportunities for extended dialogue to enhance English acquisition and learning.11 In the Special Alternative Instructional Programs reviewed by Tikunoff et al. (1991), teachers structured activities that provided English-language learners with opportunities for frequent, meaningful interactions among themselves and the teacher. In addition, they encouraged contributions by these students by focusing on the content of their responses, rather than on grammatical correctness, during content area instruction. Garcia (1990a) observes that in effective kindergarten and third and fifth grade classrooms, teachers allowed for a great deal of student-to-student interaction in the child reply component of instructional discourse segments. Gersten (1996) notes that effective teachers use questions that press students to clarify or expand on initial statements, as well as encourage students to participate in conversations.

Recently, a good deal of attention has been paid to instructional conversations—discussion-based lessons that focus on an idea or concept that has both educational value and meaning and relevance for students (see Chapter 4). The teacher encourages students to express their ideas either orally or in writing and guides them to increasingly sophisticated levels of understanding. This is a particular instance of the opportunity for extended dialogue proposed as a feature of effective instruction. Saunders et al. (1996) found that students who have opportunities for using language to elaborate and develop ideas in writing and discussion outperform their peers who do not. In a forthcoming study (Saunders and Goldenberg, in press), the authors report that fourth grade English-language learners who participated in an instructional conversation outperformed comparable students who participated in a more conventional recitation or basal-like recitation lesson.

Systematic Student Assessment

Many studies have found that effective schools use systematic student assessment—a feature identified in the effective and nominated schools research—to inform ongoing efforts to improve achievement (see also Chapter 5). For example, Carter and Chatfield (1986) report ongoing assessment of student outcomes to monitor program effectiveness. Four studies that report actual student achievement (Carter and Chatfield's [1986] effective schools study, and

11Much of the discussion here is based on research by Tharp and Gallimore (1991), Goldenberg (1992), and Rueda et al. (1992).

Page 183

Goldenberg and Sullivan's [1994], Slavin and Yampolsky's [1992], and Slavin and Madden's [1994] quasi-experiments) note the systematic assessment of student achievement by their study sites. In the Success for All programs, students are assessed every 8 weeks to determine who needs tutoring and whether groups should be changed, and to identify students who might need some other type of assistance. In the study documented by Goldenberg and Sullivan (1994), the school had developed a mechanism for assessing and discussing student progress, at least in the aggregate, on a regular basis. Although systematic student assessment is not identified as a feature in a fifth study that reports student achievement (Goldenberg and Gallimore, 1991), the principal and instructional specialist met with teachers quarterly for individual "pacing conferences," where they reviewed the progress and achievement of students in each teacher's classroom.

Staff Development

Staff training and development are important components of effective schools for English-language learners not identified in the original effective schools research. As previously mentioned, one important way to raise teacher expectations is to raise student achievement by helping teachers acquire skills and knowledge needed to be more successful with students, rather than exhorting teachers to raise their expectations. Often the training identified in the studies reviewed here is specific to teachers of these students, such as English-language development and use of sheltered instruction (Lucas et al., 1990). In other instances (e.g., Slavin and Yampolsky, 1992; Slavin and Madden, 1994), the training is in instructional strategies that are specific to the implemented program, such as use of thematic units, vocabulary development, classroom management, instructional pace, and cooperative learning, but not targeted at English-language learners per se.

Staff development for all teachers in the school, not just language specialists, was an important component of many of these programs. Although the programs provided ongoing staff development directly related to resolving new instructional issues for ESL and bilingual education teachers, they also recruited excellent content area teachers and trained them in English-language development strategies. Carter and Chatfield (1986), Lucas et al. (1990), Minicucci and Olsen (1992), and Berman et al. (1995) also document staff development explicitly designed to prepare all teachers to work with English-language learners. This contrasts with prior policies whereby ESL teachers were expected to teach content matter to these students, and mainstream teachers had no training in how to instruct them.

In preparing teachers, Moll and his colleagues (Moll et al., 1992) have avoided one pitfall often associated with culturally responsive pedagogy (defined as teaching practices attuned to the cultural background of students)—the tendency to base instructional practices on teachers' assumptions and stereotypical

Page 184

beliefs about groups of students. Drawing on the principle that "the students' community represents a resource of enormous importance for educational change and improvement," teachers and researchers involved in this work interview parents and other community members to identify the information and skills or "funds of knowledge" that are available to Mexican households through their social networks (see also Chapter 4). Teacher-researchers participating in the project then organize their curriculum accordingly. In addition, they call upon the expertise of community members and incorporate community-based knowledge sources into their curriculum. As a result, their culturally responsive curriculum is based on empirical findings about the community, rather than stereotypes.

Ongoing professional staff development is not prominently featured—if it is mentioned at all—in the original effective schools literature. Yet it is now a universally agreed-upon component of any effective school and certainly of any effort to change and improve a school. A real question that remains is what sort of training is most relevant for improving school processes, as well as teacher knowledge and skills. It is also important to validate the effectiveness of this training through assessments of student outcomes (see also Chapter 8).

Home and Parent Involvement

Home and parent involvement—an attribute that, like staff development, was not a part of the original effective schools conceptualization—plays an important role in enhancing outcomes for English-language learners. Moll (1988), Garcia (1990b), and Carter and Chatfield (1986) all note that in the effective schools they document, an ongoing community/school process is an important contributor to the school's success. Garcia observes that in effective schools, teachers have a strong commitment to school-home communication, and parents are involved in formal parent support activities. In the school documented by Carter and Chatfield, volunteers from the community worked in the school, both directly with children and in helping teachers prepare classroom materials. The school's staff also tried to enhance home-school connections even further by improving the quality of homework teachers assigned.

Lucas et al.'s (1990) effective high schools encouraged parents to become involved in their children's education. This was accomplished in different ways at the various schools, through means such as parent advisory committees, newsletters, monthly parent nights, evening student performances, teacher-parent meetings, student-of-the-month breakfasts, honors assemblies, and community liaisons. It is important to note that home-school connections apparently played a role even in high school, despite the fact that parents' direct involvement in children's schooling declines as children get older (Stevenson and Baker, 1987). Lucas et al. do report, however, that parent participation was the "least developed component" in the high schools they visited (p. 334).

Page 185

Different types of home-school connections probably have different types of effects (Epstein, 1992). For example, newsletters can make parents knowledgeable about what is happening at school, but we should not expect them to improve student achievement unless they contain information parents can actually use to influence, for example, TV viewing, time spent reading or doing homework, or trips to the library or museums. Similarly, parent participation in school governance is likely to help parents feel empowered, but we should not expect it to make parents more knowledgeable about disciplining or motivating children or helping them achieve more academic success unless parents are provided with pertinent information or training in the course of their governance activities.

Neither the studies reviewed here nor any other existing studies can answer the question of what type of home or parent involvement is most effective. Extrapolating from the observations in these studies, however, two hypotheses seem reasonable. First, cognitive or academic effects are most likely to be the result of home-school connections that focus specifically on cognitive or academic learning at home, that is, increasing and improving home learning opportunities through the use of homework or other organized activities designed to promote learning. Second, schools with comprehensive home involvement programs encompassing various types of home-school connections probably help families and children in a number of important ways. The more types of productive connections homes and schools can forge, the more positive and powerful the effects on children, families, and schools will be. At least in U.S. settings, these hypotheses are probably valid regardless of students' cultural or language background (Goldenberg, 1993). (See Chapter 4 for further elaboration on this theme.)

Methodological Strengths and Limitations of the Studies

Each of the major types of studies reviewed here has its methodological strengths and limitations. These are examined in the subsections that follow.

Effective and Nominated Schools and Classroom Designs

The nominated schools and classroom designs have introduced a valuable element to the literature—rich and highly detailed descriptions, some quantitative and some qualitative, of schools and classrooms. School and community contexts, relevant histories with specific populations of students, and the perceptions of key players—students, teachers, administrators, and parents—are prominently featured in articles and technical reports, as are detailed and sophisticated studies of classroom learning and discourse environments. Although detailed accounts of effective schools are not completely absent from the earlier effective schools literature (e.g., Weber, 1971), researchers have become increasingly sophisticated in the range and depth of data they collect. The resulting rich portraits

Page 186

of supposedly effective programs and practices are especially welcome given the growing diversity of the U.S. school population.

As exploratory strategies, both the effective and nominated schools designs make a great deal of intuitive and logical sense. But there are also limitations to what they can tell us. First, and most fundamental, neither design directly or empirically addresses the issue of how a school or classroom came to be effective, except for possible retrospective accounts and inferences. Carter and Chatfield (1986:204), who provide one example of the effective schools paradigm applied to English-language learner schooling issues, explicitly recognize this limitation: "This paper makes no claim of providing solutions to the profound problems associated with educational change, knowledge utilization, and innovation acceptance." Carter and Chatfield also recognize the danger inherent in attempting to implement willy-nilly the results of effective schools research. Such attempts almost certainly lead to what they call an "implementation of attributes" approach, involving the issuance of top-down mandates requiring local schools to "implement'' strong leadership, high expectations, a safe and businesslike school climate, and so on. "Serious, objective research is required to analyze school-improvement strategies and ultimately to develop strategies appropriate to the complexity of effective schools" (p. 203).

A second limitation of these designs, related to the first, is the difficulty of separating cause from effect: Do the characteristics of schools cause them to be effective, or does effectiveness lead to these characteristics? This is a particular problem with effective and nominated schools designs, both of which make strong claims about causes (e.g., high expectations, positive school climate) and effects (e.g., high student achievement). However, since the conditions that presumably make a school effective are gauged at the same time as (or even after) achievement data are collected, such claims are problematic. For example, in the effective schools studies, we do not know whether high expectations preceded or followed increased effectiveness. The cause-effect problem is particularly acute in the effective and nominated schools studies reviewed here since these studies do not include comparison sites, which would at least permit the studies to claim (empirically) a correlation between school factors that contribute to effectiveness on the one hand and desirable outcomes for students on the other.

A third limitation is that the nominated schools design now in favor reports no data whatsoever on student outcomes, although some gauge of student outcomes may have been used in the selection process. Exemplary schools are selected because they satisfy criteria shared by nominators and investigators regarding what effective schooling for English-language learners should look like. In many cases these schools are chosen regardless of whether it can be shown that the characteristics they display lead to desired outcomes (social, cognitive, or affective), gauged independently of the processes considered "effective," "exemplary," or "desirable." Thus, these studies by definition remain inconclusive on the question of effectiveness. Moreover, there is a major risk of

Page 187

circularity: the exemplary schools do things that are "exemplary" or, as some authors note, are consistent with what researchers and leading practitioners say should be done. Berman et al. (1992:Vol.1, 26) readily acknowledge the issue:

It is impossible to know how many programs were included or excluded [in their study of well-implemented programs for English-language learners] by nominators on the basis of their own philosophies about what constitutes a good LEP program. …

Problems with the methods used to investigate instructional processes and environments compound the issue. Some studies describe only one school, which makes it difficult to determine which factors are important. In some cases, very little information is provided on data collection methods. For those studies that report information, other problems are apparent, including poor or absent reporting of interrater reliability, limited depth and duration of observations, lack of systematic methods for observing classroom events, or combining of information across sites.

Prospective Case Studies

Prospective case studies have the advantage, in principle, of collecting data contemporaneously with change efforts, permitting observation and analysis of the actual change process, participants' views and perspectives, and the apparent ongoing results of the changes undertaken. Under ideal circumstances, they would be true cases of the implementation of theories regarding effective schooling.

Our systematic review uncovered very few studies beyond those described in the previous section. Even in the California Case Studies project, funding was terminated prematurely before important issues, such as the processes and dynamics of change and school-level factors related to change and improvement, could be documented, analyzed, and reported (N. Gold, personal communication, December 1995; Crawford, 1995).

An inherent characteristic of prospective studies in schools is the close relationship between the educator who implements the program and the researcher. In the studies reviewed here, the school changes that were made and studied were instigated by the researcher/author of the report in collaboration with educators at their respective sites. Thus, this set of studies constitutes "action research," where the line separating efforts to improve schooling for English-language learners and the conduct of academic research has been blurred or removed. An advantage of these studies, from a purely pragmatic standpoint, is that if the changes are effective and actually work, the students and teachers at the intervention site will have benefited. However, methodological problems, possibly related to the close collaboration of researchers and educators, can compromise study findings. For example, in one study, investigators who analyzed the interview

Page 188

protocols for changes in student knowledge knew which protocols were pretest and which were post-test.

Many of the issues noted above regarding effective and nominated schools research are classic threats to validity. How do we separate cause from effect? How do we establish causality at all? How do we know whether something has an "effect"? How do we gauge that effect? Similar questions can be asked about the prospective case study design that focuses exclusively on the intervention site: How do we know that shifts in achievement patterns following implementation of the changes are due to those changes? What about extraneous factors, general upward drift in scores, or regression effects?

Quasi-Experiments

Traditionally, threats to validity have been addressed within an experimental framework or, when dealing with social phenomena where random assignment is impossible, a quasi-experimental framework. From a design standpoint, the quasi-experimental design obviously offers a stronger basis for claiming that changes in student achievement resulted from something that happened at a target site. In the absence of a comparison site with students who are comparable in features such as demographics and transience, changes in student outcomes at a particular school can be due to any number of extraneous factors or artifacts. Quasi-experiments also permit stronger causal inferences about school processes, dynamics, and structures on the one hand and improvements in student outcomes on the other.

However, school changes are so complex and involve so many dimensions that it is usually very difficult to draw tight linkages between specific processes or program components and student outcomes. For example, although one study (Henderson and Landesman, 1992) indicated that students in thematic math instruction outperformed those in a regular math program, the cause of this effect is unclear. The experience of the treatment group differed from that of the control group on more dimensions than thematic instruction, including student access to bilingual instruction, extended periods of time with the same teacher and classmates, and instruction in omitted content between themes and at the conclusion of the school year. Many of the quasi-experimental studies also have the limitation of providing very little information about the intervention received by the control group, thus making it difficult to gauge the actual merits of the intervention.

Quasi-experimental designs are really just parallel case studies and do not preclude in-depth study and subtle analysis of school and instructional organization features. On the contrary, richer descriptions of the processes and dynamics of school change would permit clearer interpretations or hypotheses about what explains changes in student outcomes—or the failure to effect such changes. Quasi-experimental designs do require that investigators either take an active hand in helping to bring about changes at a school or be present when a school, on

Page 189

its own, decides to try to instigate changes, so that appropriate measures in the "before" state can be taken. In either case, investigators must then gauge the effects of those changes on student outcomes, using appropriate measures and comparable schools as controls.

Some of the studies reviewed here—particularly those that examine student outcomes and relate them to changes in school-wide and classroom functioning and organization—suggest processes by which schools and classrooms can reorganize themselves to promote higher levels of achievement for students. One framing of this complex issue has been articulated by Slavin and Madden (1995) in their most recent summary of the research on Success for All: Can a school become effective by successfully adopting an effective, externally developed program, or is a certain amount of "reinventing the wheel" required, school by school? Although Slavin and Madden's results provide a strong basis for concluding that some well-defined effective programs can be exported successfully to other schools in other communities with different students and staffs, their position (and, apparently, their data) runs counter not only to much of the accepted wisdom in the school reform literature, but also to previous efforts to disseminate and replicate effective programs (e.g., Anderson et al., 1978).

A Note About Process

In the conduct of the research recommended below, collaborative efforts between practitioners and other groups (e.g., state departments of education, universities, research laboratories) are especially important, given the inordinate difficulty of simultaneously bringing about substantive school change and conducting research on it. The California Case Studies in bilingual education, a partnership between the California State Department of Education and five public elementary schools, provides a model for such collaboration between a public agency and the schools (Gold and Tempes, 1987). School-university partnerships (e.g., the early Success for All studies; see Goldenberg and Sullivan, 1994) are also recommended, complex and fraught with difficulties though they may be (Trubowitz et al., 1984).

Research Needs

Identification of Effective Classrooms and Schools

7-1. Researchers should make explicit their principles for selecting effective schools and classrooms. These principles should be based on some combination of indicators of process (e.g., curriculum, leadership, school climate, instructional strategies) and outcomes (e.g., standardized and performance-based achievement measures). The definition should be influenced by local priorities and contexts.

Page 190

Given the variation in the way effectiveness is defined across studies, research needs to address what we mean by "effective." Research on effective schools could benefit greatly from the development of some principles of effectiveness for English-language learners that would still leave room for variations based on local priorities and contexts. These principles might incorporate issues of equity and access for all students (which would address the issue of separation of English-language learners from native-English speakers), theoretical foundations for programs and practices, evidence from student achievement, and evidence from student behavior and engagement in school (e.g., attendance, suspensions, graduation rates). The development of principles of effectiveness is especially important in the field of education research, a methodologically pluralistic field in which interpretive and postmoderist approaches thrive alongside traditional approaches that pursue objectivity and positivism. In some cases, effectiveness appears to be tied to a particular theory of teaching and learning. Other studies take a different approach and define effectiveness in terms of measurable student achievement outcomes: a school or teachers are relatively effective if students are achieving at some criterion level, or at least significantly better than students in comparable schools.

Research on effectiveness should clearly state how effective schools and classrooms are recognized. Nominated schools studies are useful in the provision of exploratory data and in their ability to integrate a wide array of complex information, some of which is difficult to capture through objective methods. However, we urge researchers to identify the nature of the complex decisions made by nominators. In addition, we must look for concrete and documented evidence that programs and practices claiming to be "exemplary" also help produce desirable student outcomes.

Extent of Variability in the Definition of Effective Schools and
Classrooms for English-Language Learners

7-2. Research should investigate how definitions of effectiveness interact with local site characteristics and student characteristics.

The following sorts of questions might be pursued. Does effective leadership have one set of attributes in a school with certain cultural and demographic characteristics and a different (perhaps related) set of attributes in a school with a different cultural make-up? Do effective home-school connections have different characteristics depending on the cultural group involved? In general, do attributes of effective schooling for language-minority students vary by students' linguistic, cultural, or national-origin group; socioeconomic level (including transiency); degree of exposure to English outside of school; generational status (immigrant, first or second generation); and/or schooling level (early or late elementary, middle, or high school)? The hypotheses and possible interactions

Page 191

are nearly limitless. But if we are to advocate differentiated programming for students, this should be done on the basis of evidence that such programming has practical and meaningful benefits to students, their families, and their communities.

A key issue is whether effective teacher practices for students generally are sufficient to help English-language learners succeed in school, or whether knowledge and skills specific to the latter are needed. Lucas et al. (1990:329) take an unequivocal stand: "Teachers who are expert in the instruction of mainstream students are not necessarily effective instructors of language-minority students." They cite examples from their nominated high schools of policies that encourage teachers to get ESL and bilingual certification and training. Carter and Chatfield (1986:228), in contrast, are more cautious, suggesting the question is an empirical one:

The present popularity of the immersion approach for LEP children implies that bilingual education is unnecessary and that LEP children can achieve without it. …[S]uch situations should be carefully studied; the independent influence of school effectiveness with minority-language children can only be isolated by the careful analysis of any such situations.

7-3. Research should examine the extent to which "generic" reform efforts incorporate English-language learners. Moreover, this research should explore whether these reform efforts are beneficial, and if not how they can be adapted to this group of students.

Several questions in this area need further study: To what extent do school reform efforts include English-language learners? Are the reforms in curriculum, instruction, assessment, school organization and governance, and community engagement beneficial for English-language learners? Are some more beneficial than others? Are adaptations needed to make them beneficial, or are they effective for all students without major adaptations? If adaptations are needed, what might they be?

This is a variation on recommendation 7-2, addressing whether teachers need special skills and knowledge to teach English-language learners effectively. There are those who believe that good teaching is good teaching, no matter who the students are. Research on how current reforms in curriculum, assessment, and school organization and the particulars of classroom instruction are and are not appropriate for and effective with English-language learners would provide empirical evidence to support or refute such beliefs.

Developmental Issues

7-4. There is a clear need for research to examine the effects of instructional interventions and social environments on the linguistic, social, and cognitive development of young children.

Page 192

At all grade levels, special challenges exist for English-language learners and their educators. As a result, findings based on elementary school studies may not be fully appropriate for early education or secondary contexts, and some questions may not have been dealt with at all. At the primary level, such challenges include the fact that young children have not yet developed many concepts in their native language. The few studies at the early childhood level focus on program effects on the English-language development of young children (see Chapter 2). Given the increasing numbers of children who attend school earlier, this is an especially important area of research. Other research areas that merit attention include the effects of various kinds of programming on students' first-language development, as well as their social and cognitive development. An important aspect of this research is the need for appropriate assessment (see Chapter 5).

7-5. More studies are needed to identify attributes of effective middle and secondary schools and classrooms serving English-language learners.

Although there are more studies that focus on middle and secondary school than on very young English-language learners, more prospective and quasi-experimental research is needed at these levels, as is research examining issues that adversely affect students at these levels. Issues related to sheltered instruction are particularly important for secondary school students, who are expected to master more complex content through their second language. Other challenges include the relative lack of time to master language and content needed for graduation and post-secondary opportunities, great variability among students in prior preparation for this level of schooling, and noninstructional features such as transiency and family/work responsibilities. Specific research questions include the following: How is such content best made accessible? What practices are effective for students with limited prior formal schooling? What factors contribute to students' persistent classification as English-language learners after a number of years in U.S. schools, and how can their needs be met? How can secondary schools integrate English-language learners into the life of the school while meeting their specific needs? What programs and practices lead to best access to post-secondary opportunities? Assessment issues are also extremely important and difficult in connection with these questions (see Chapter 5).

7-6. Research is needed to assess the effectiveness of newcomer programs, either in relationship to each other or compared with doing nothing at all. Further research is needed to determine whether there is a well-realized and -packaged program that could be readily exported.

Late enrollment, especially of non-English-speaking students entering the upper grades, presents a challenge; the constant influx of new students is clearly disruptive to teachers, support staff, and students themselves. A program that is effective under fairly stable conditions might be taxed to the breaking point with

Page 193

a continual stream of new enrollments. Many new enrollments, particularly among English-language learners, are also new to schooling in the United States. Some schools and districts have opted for "newcomer centers" where late-enrolling students, particularly those who come from other countries, attend school for some transition period (Friedlander, 1991). The curriculum and services offered address the needs of newcomers specifically by providing intensive language and cultural orientation, as well as basic skills and content remediation as appropriate. A number of models and programs are currently in operation, and anecdotal data suggest that newcomer programs are very successful and popular with students (Chang, 1990; Friedlander, 1991; Mace-Matluck et al., in press). Studies have described some of the programs (Chang, 1990; Friedlander, 1991; McDonnell and Hill, 1993; Olsen and Dowell, 1989) or included such programs in their reviews of promising school practices (Mace-Matluck et al., in press; Romo, 1993), but no comprehensive research has been conducted to compare student achievement in newcomer programs with that in other program options (e.g., transitional bilingual education, sheltered instruction).

Special Issues

7-7. Once learning goals have been set by the community, research is needed to determine the linguistic and cultural adaptations that will help English-language learners meet these goals. What methods work best to give English-language learners access to the academic and social opportunities that native English speakers have while they are learning English? Such methods include both school-wide adaptations, such as the way sequences of classes are organized to give English-language learners optimal access to subject matter knowledge and English proficiency, and classroom adaptations, such the use of particular teaching strategies and classroom composition.

An obvious characteristic that distinguishes English-language learners from others with regard to effective schooling is that their native language is not English, the primary medium of instruction in the United States. This fact raises two major issues that remain unresolved in the research literature: How should language of instruction (native language and English) be arranged, and when instruction is in the second language, what adjustments, if any, are needed?

The need to determine the appropriate instructional role for the native language and English is a recurring theme in our review. Because of the complexity involved (e.g., the need to consider demographic and contextual factors) and the dearth of rigorous studies addressing the question, the evidence in the literature is contradictory. For example, there is conflicting evidence regarding whether the native language and English should be used concurrently or kept separate. Wong Fillmore et al. (1985) found that when two languages are used concurrently, students listen only to their native language and ignore the second, thus hampering

Page 194

English learning. However, other investigators have found otherwise (Tikunoff, 1983; Gersten, 1996). For example, Tikunoff (1983) found that in successful classrooms, the students' native languages were used substantially by instructors to develop the lesson content. When a student was not comprehending what was required or needed feedback to complete a task, teachers frequently switched to the student's native language and used it to clarify the instruction.

There are also many open questions related to how to adapt English-language materials and instruction so they are both comprehensible and conceptually appropriate for English-language learners. For example, some experts recommend reducing nonessential details and simplifying grammatical structures in text (Short, 1994). Others, however, maintain that simplifying "surface linguistic features" will not necessarily make the text easier to comprehend, and that reducing nonessential details may delete important background information crucial to the interpretation of meaning when knowledge of language forms is limited (Saville-Troike, 1991). Moreover, Berman et al. (1992:5) found that sheltered English classes (in which both content instruction and ESL instruction are provided in a self-contained classroom, and teachers use a simplified form of English and modify their teaching techniques to make instruction comprehensible to English-language learners) result in instruction "prone to low-expectations and overly simplified curriculum."

Concerns about effectiveness must go beyond language as well. As noted elsewhere in this report, not only are English-language learners language-minority students; they are also usually from cultural or ethnic groups whose sociolinguistic, cognitive, or motivational attributes may not coincide or be congruent with attributes conducive to success in U.S. schools. There is a wealth of writing and research bearing on precisely this issue (e.g., Tharp, 1989). However, there is surprisingly little direct empirical confirmation that culturally accommodated instruction (or school-wide organization) actually produces higher levels of academic achievement. We need research that will examine in depth and in detail how instruction, assessment, curriculum, school organization, school leadership, and professional development can be designed and adapted to be "effective" with students of widely varying backgrounds and experiences—including English-language learners of different backgrounds.

7-8. Involving families of English-language learners and engaging community resources on their behalf poses special challenges for schools. More focused research is needed to provide information about the challenges to such involvement and engagement, the potential benefits, and successful approaches.

Several important questions need to be addressed in this area: What approaches are most effective in getting families of English-language learners of different backgrounds involved in school activities? How can school involvement be made more inviting to parents? How can school staff be made more receptive to parent and community involvement? What types of family involvement

Page 195

are most effective in influencing the academic achievement of English-language learners? Other entities besides schools and families play a significant role in the lives of many English-language learners. Research needs to reach beyond the classroom and school to examine how external agencies can work with schools to increase educational effectiveness for English-language learners.

7-9. Research is needed to determine the resources required for effective instruction of English-language learners in different contexts.

Many authors attribute poor implementation of program models to lack of sufficient resources. Berman et al. (1992) found, for example, that schools face severe resource limitations, and thus have problems implementing their programs, regardless of the model they adopt. Thus outcomes for most schools may be due to implementation factors that have nothing to do with the validity of the approach underlying a program.

Prospective Research

7-10. Prospective research that examines the school change process is needed, beginning from the point before a school undertakes change, to document the processes and outcomes on a sound theoretical and programmatic basis. Prospective studies should document the problems, possibilities, dynamics, difficulties, successes, and outcomes of school and program change. An important focus would be on how schools and teachers maintain effective components once in place. Research should also determine which kinds of improvement strategies are exportable and which aspects may be influenced by local context.

In addition, future research should examine the benefits and shortcomings of different improvement strategies, again using models and programs already in existence. A component of this research would be to examine whether educators and policymakers find empirical research or rich cases more compelling in prompting them to change their current practices. Some prospective case studies of sites on the verge of reform could help answer these important policy implementation questions.

To what extent and under what conditions are successful models transportable? How much adaptation, even reinvention, is needed? Clearly, such research should build on lessons learned from previous studies of implementation. For example, a 4-year two-phase study conducted in the 1970s (Berman and McLaughlin, 1978) examined the initiation and implementation of federally funded local projects, including Title VII. The study findings indicate the importance of "mutual adaptation, the process by which a project is adapted to the reality of its institutional setting, while at the same time, teachers and school officials adapt their practices in response to the project. Effective strategies

Page 196

provide each teacher with necessary and timely feedback, allow project-level choices to be made to correct errors, and encourage commitment to the project'' (p.viii). In their evaluation of Academic Excellence Program dissemination efforts, Wilson et al. (1994:85) conclude that more attention should be paid to finding out and communicating what works best to disseminate effective bilingual education practice. … [There need to be] on-site case studies of various projects followed by larger scale surveys." For example, are generalizable findings more or less convincing than a well-told story of a school that worked? Does evidential preference interact with one's stance toward the innovation; that is, are advocates more likely to rely on the compelling case, whereas opponents are more likely to cite quantitative findings? Here again, we are likely to learn more about how to help English-language learners, indeed all children, if we have an understanding of the processes leading to successful change at particular sites, including the successful implementation and adaptation of effective models.

The effect of very locally specific factors is an important issue. The integration of various research-based responses in a single site may be impossible, or some responses may be incompatible with local factors. Samaniego and Eubanks' (1991) secondary analysis of the California Case Studies data suggests that the language characteristics of the community in which a bilingual program exists have a strong "effect on the development of second language proficiency [suggesting] that no bilingual education approach, however sensible and theoretically well-supported, can be applied in a uniform way with equal success in substantially different learning environments." Slavin and Madden (1994) are undoubtedly correct when they warn of the dangers of reinventing the wheel, school by school. Yet local and school realities also seem to suggest the need to consider site- and community-specific dimensions in the creation and evolution of effective schools for English-language learners. A certain amount of invention, perhaps even reinvention, may always be necessary. It is undoubtedly also true, however, that schools and students clearly stand to gain by learning about and even importing successful practices, even if not entire programs, developed elsewhere.

Page 197

Annex: Table 7-1
Studies Of School And Classroom Effectiveness

Page 198

TABLE 7-1 Studies of School and Classroom Effectiveness

There was a problem loading page 199.

Page 200

TABLE 7-1 Continued

There was a problem loading page 201.

Page 202

TABLE 7-1 Continued

There was a problem loading page 203.

Page 204

TABLE 7-1 Continued

There was a problem loading page 205.

Page 206

TABLE 7-1 Continued

There was a problem loading page 207.

Page 208

TABLE 7-1 Continued

There was a problem loading page 209.

Page 210

TABLE 7-1 Continued

There was a problem loading page 211.

Page 212

TABLE 7-1 Continued

There was a problem loading page 213.

Page 214

TABLE 7-1 Continued

There was a problem loading page 215.

Page 216

TABLE 7-1 Continued

There was a problem loading page 217.

Page 218

TABLE 7-1 Continued

There was a problem loading page 219.

Page 220

TABLE 7-1 Continued

There was a problem loading page 221.

Page 222

TABLE 7-1 Continued

There was a problem loading page 223.

Page 224

TABLE 7-1 Continued

There was a problem loading page 225.

Page 226

TABLE 7-1 Continued

There was a problem loading page 227.

Page 228

TABLE 7-1 Continued

There was a problem loading page 229.

Page 230

TABLE 7-1 Continued

There was a problem loading page 231.

Page 232

TABLE 7-1 Continued

There was a problem loading page 233.

Page 234

TABLE 7-1 Continued

There was a problem loading page 235.

Page 236

TABLE 7-1 Continued

There was a problem loading page 237.

Page 238

TABLE 7-1 Continued

There was a problem loading page 239.

Page 240

TABLE 7-1 Continued

There was a problem loading page 241.

Page 242

TABLE 7-1 Continued

1Study design is "effective schools," "nominated schools," "prospective case study," or ''quasi-experimental." An effective schools design is one in which one or more effective schools are selected on the basis of test data showing that students at the school(s) achieve either at grade level or at least at higher levels than the school's sociodemographic characteristics would predict. A nominated schools design is one in which schools are chosen on the basis of nominations from professionals who consider the school "good," "effective," "exemplary," etc. In both effective schools and nominated schools designs, researchers work essentially retrospectively, attempting to determine features of the school's organization or operation that help explain its effectiveness. Neither design directly or empirically addresses the issue of how a school came to be effective, except for retrospective accounts or inferences. A prospective case study begins with a school that is no better or perhaps even worse (in terms of effectiveness) than other comparable schools. It then examines changes in the school over time and tries to explain how the school(s) went from less to more effective and what the effects of these changes have been. There is no direct, concurrent comparison with other comparable schools. A quasi-experimental study is also a prospective study and begins with a school (or schools) that is no more effective and perhaps less effective than other comparable schools. However, it examines the effects of specific interventions on student outcomes and, most critically, concurrently in comparison with student outcomes at a comparable school (or groups of schools) not participating in the intervention. The quasi-experimental design offers the strongest basis for making causal inferences about school processes, dynamics, and structures on the one hand and improvements in student outcomes on the other.

There was a problem loading page 243.

Page 244

References

Anderson, R., R. St. Pierre, E. Proper, and L. Stebbins
1978 Pardon us, but what was the question again? A response to the critique of the follow-through evaluation. Harvard Educational Review 48:161-170.

Berman, P., and M.W. McLaughlin
1978 Federal Programs Supporting Educational Change, Vol. VII: Implementing and Sustaining Innovations. R-1589/8-HEW. Santa Monica, CA: Rand.

Berman, P., J. Chambers, P. Gandara, B. McLaughlin, C. Minicucci, B. Nelson, L. Olsen, and T. Parrish
1992 Meeting the challenge of language diversity: An evaluation of programs for pupils with limited proficiency in English. Vol. 1 [R-119/1: Executive Summary; Vol. 2 [R-119/2]: Findings and Conclusions; Vol. 3 [R-119/3]: Case Study Appendix. Berkeley, CA: BW Associates.

Berman, P., B. McLaughlin, B. McLeod, C. Minicucci, B. Nelson, and K. Woodworth
1995 School Reform and Student Diversity: Case Studies of Exemplary Practices for LEP Students (Draft Report). National Center for Research on Cultural Diversity and Second Language Learning and BW Associates. Berkeley, CA.

Bliss, J., W. Firestone, and C. Richards, eds.
1991 Rethinking Effective Schools: Research and Practice. Englewood Cliffs, NJ: Prentice Hall.

Calderon, M., R. Hertz-Lazarowitz, and R. Slavin
1996 Effects of Bilingual Cooperative Integrated Reading and Composition on Students Transitioning from Spanish to English Reading. Unpublished paper for the Office of Educational Research and Improvement, U.S. Department of Education, Washington, DC.

Carter, T., and M. Chatfield
1986 Effective bilingual schools: Implications for policy and practice. American Journal of Education 95:200-232.

Chamot, A.U., M. Dale, J.M. O'Malley, and G. Spanos.
1992 Learning and problem solving strategies of ESL students. Bilingual Research Journal 16(3-4):1-33.

Chang, H.
1990 Newcomer Programs: Innovative Efforts to Meet the Educational Challenges of Immigrant Students. San Francisco, CA: California Tomorrow.

Chasin, G., and H. Levin
1995 Thomas Edison accelerated elementary school. In J. Oakes and K. H. Quartz, eds., Creating New Educational Communities. 94th Yearbook of the National Society for the Study of Education, Part 1. Chicago, IL: University of Chicago Press.

Cohen, E.
1984 Talking and working together: Status, interaction, and learning. Pp. 171-187, Chapter 10 in P.L. Peterson et al., eds., The Social Context of Instruction: Group Organization and Group Processes. Orlando, FL: Academic Press.

Coleman, J., E.Q. Campbell, C.J. Hobson, J. McPartland, A.M. Mood, F.D. Weinfeld, and R.L. York
1966 Equality of Educational Opportunity. Office of Education, U.S. Department of Health, Education, and Welfare. Washington, DC: U.S. Government Printing Office.

Comer, J.
1980 School Power: Implications of an Intervention Project. New York: Free Press.

Cook, T.D., and D.T. Campbell
1979 Quasi-experimentation. Chicago, IL: Rand McNally.

Page 245

Crawford, J.
1995 Bilingual Education: History Politics Theory and Practice. Los Angeles: Bilingual Educational Services.

Dianda, M., and J. Flaherty
1995 Effects of Success for All on the Reading Achievement of First Graders in California Bilingual Programs. Los Alamitos, CA: The Southwest Regional Educational Laboratory.

Edelsky, C., K. Draper, and K. Smith
1983 Hookin' 'em in at the start of school in a 'whole language' classroom. Anthropology & Education Quarterly 14:257-281.

Edmonds, R.
1979 Effective schools for the urban poor. Educational Leadership 37(1):15-24.

Education Week
1995 Next generation of effective schools looks to districts for lasting change. Education Week (April 12):8-9.

Epstein, J.
1992 School and family partnerships. Pp. 1139-1152 in M. Alkin, ed., Encyclopedia of Educational Research. 6th ed. New York: MacMillan.

Escamilla, K.
1994 Descubriendo la lectura: An early intervention literacy program in Spanish. Literacy, Teaching and Learning 1(1):57-70.

Fisher, C.W., L.F. Guthrie, and E.B. Mandinach
1983 Verification of Bilingual Instructional Features. San Francisco, CA: Far West Laboratory for Educational Research and Development.

Friedlander, M.
1991 The Newcomer Program: Helping Immigrant Students Succeed in U.S. Schools. Program information guide series, No. 8. Washington, DC: National Clearinghouse for Bilingual Education.

Garcia, E.E.
1990a Instructional discourse in 'effective' Hispanic classrooms. Pp. 104-117 in Rodolfo Jacobson and Christian Faltis, eds., Language Distribution Issues in Bilingual Schooling. Bristol, PA: Multilingual Matters.
1990b Education of Linguistically and Culturally Diverse Students: Effective Instructional Practices. The National Center for Research on Cultural Diversity and Second Language Learning. Educational Practice Report, No. 1. Center for Applied Linguistics, Washington, D.C.

Gersten, Russell
1996 Literacy instruction for language-minority students: The transition years. The Elementary School Journal 96(3):228-244.

Gold, N., and F. Tempes
1987 A State Agency Partnership with Schools to Improve Bilingual Education. Paper presented at the annual meeting of the American Educational Research Association, Washington, DC. California State Department of Education.

Goldenberg, C.
1992 Instructional Conversations and Their Classroom Application. The National Center for Research on Cultural Diversity and Second Language Learning. Educational Practice Report, No. 2. Center for Applied Linguistics, Washington, D.C.
1993 The home-school connection in bilingual education. Pp. 225-250 in B. Arias and U. Casanova, eds., Ninety-second Yearbook of the National Society for the Study of Education. Bilingual education: Politics, Research, and Practice. Chicago, IL: University of Chicago Press.

Page 246

Goldenberg, C., and R. Gallimore
1991 Local knowledge, research knowledge, and educational change: A case study of first-grade Spanish reading improvement. Educational Researcher 20(8):2-14.

Goldenberg, C., and J. Sullivan
1994 Making Change Happen in a Language-minority School: A Search for Coherence. EPR #13. Washington, DC: Center for Applied Linguistics.

Henderson, R.W., and E.M. Landesman
1992 Mathematics and Middle School Students of Mexican Descent: The Effects of Thematically Integrated Instruction. Research Report: 5. National Center for Research on Cultural Diversity and Second Language Learning, Santa Cruz: University of California.

Hernandez, J.S.
1991 Assisted performance in reading comprehension strategies with non-English proficient students. The Journal of Educational Issues of Language Minority Students 8:91-112.

Jencks, C., M. Smith, H. Acland, M.J. Bane, D. Cohen, H. Gintis, B. Heyns, and S. Michelson
1972 Inequality: A Reassessment of the Effect of Family and Schooling in America. New York: Harper.

Jussim, L.
1986 Self-fulfilling prophecies: A theoretical and integrative review. Psychological Review 93:429-445.

Krashen, S., and D. Biber
1988 On Course: Bilingual Education's Success in California. Sacramento, CA: California Association for Bilingual Education.

Lucas, T., and A. Katz
1994 Reframing the debate: The roles of native languages in English-only programs for language minority students. TESOL Quarterly 28(3):537-561.

Lucas, T., R. Henze, and R. Donato
1990 Promoting the success of Latino language-minority students: An exploratory study of six high schools. Harvard Educational Review 60:315-340.

Mace-Matluck, B. J., R. Alexander-Kasparik, and R. Queen
in press Toward an Effective Educational Delivery System for Low-schooled Immigrant Adolescents. Washington, DC: Center for Applied Linguistics.

Mace-Matluck, B.J., W.A. Hoover, and R.C. Calfee
1989 Teaching reading to bilingual children: A longitudinal study of teaching and learning in the early grades. NABE Journal 13:3.

McDonnell, L., and P. Hill
1993 Newcomers in American Schools: Meeting the Educational Needs of Immigrant Youth. Santa Monica, CA: Rand.

Minicucci, C., and L. Olsen
1992 An exploratory study of secondary LEP programs. R-119/5; Vol. V of Meeting the Challenge of Language Diversity: An Evaluation of Programs for Pupils with Limited Proficiency in English. Berkeley, CA: BW Associates.

Moll, Luis C.
1988 Some key issues in teaching Latino students. Language Arts 65(5):465-472.

Moll, L.C., C. Amanti, D. Neff, and N. Gonzalez
1992 Funds of knowledge for teaching: Using a qualitative approach to connect homes and classrooms. Theory into Practice 31(2):132-141.

Moll, L., E. Diaz, E. Estrada, and L. Lopes
1981 The Construction of Learning Environments in Two Languages. San Diego, CA: Laboratory of Comparative Human Cognition.

Page 247

Muniz-Swicegood, M.
1994 The effects of metacognitive reading strategy training on the reading performance and student reading analysis strategies of third-grade bilingual students. Bilingual Research Journal 18(1&2):83-97.

Olsen, L., and C. Dowell
1989 Bridges: Promising Programs for the Education of Immigrant Children. San Francisco, CA: California Tomorrow.

Pease-Alvarez, L., E. E. Garcia, and P. Espinosa
1991 Effective instruction for language-minority students: An early childhood case study. Early Childhood Research Quarterly 6:347-361.

Purkey, S., and M. Smith
1983 Research on effective schools: A review. Elementary School Journal 83:427-452.

Romo, H.
1993 Mexican Immigrants in High Schools: Meeting Their Needs. ERIC Digest. Charleston, WV: ERIC Clearinghouse on Rural Education and Small Schools.

Rosebery, A. S., B. Warren, and F. R. Conant
1992 Appropriating scientific discourse: Findings from language minority classrooms. The Journal of the Learning Sciences 2(1):61-94.

Rosenshine, B., and R. Stevens
1986 Teaching Functions. Pp. 376-391 in M. Wittrock. ed., Handbook of Research on Teaching. 3rd ed. New York: Macmillan.

Rueda, R., C. Goldenberg, and P. Gallimore
1992 Rating Instructional Conversations: A Guide. Educational Practice Report, No 4. Santa Cruz, CA: National Center for Research on Cultural Diversity and Second Language Acquisition.

Rutter, M., B. Maughan, P. Mortimore, and J. Ouston
1979 Fifteen Thousand Hours: Secondary Schools and Their Effects on Children. Cambridge, MA: Harvard University Press.

Samaniego, F., and L. Eubank
1991 A Statistical Analysis of California's Case Study Project in Bilingual Education (TR#208). Davis, CA: Intercollegiate Division of Statistics, University of California, Davis.

Saunders, W., and C. Godenberg
in press Can you engage students in high-level talk about text and support literal comprehension too? The effects of instructional conversation on transition students' concepts of friendship and story comprehension. In R. Horowitz, ed., Talk About Text: Developing Understanding of the World Through Talk and Text. Newark, DE: International Reading Association.

Saunders, W., G. O'Brien, D. Lennon, and J. McLean
1996 Making the transition to English literacy successful: Effective strategies for studying literature with transition students. In R. Gersten and R. Jimenez, eds, Effective Strategies for Teaching Language Minority Students. Monterey, CA: Brooks Cole.

Saville-Troike, M.
1991 Teaching and Testing for Academic Achievement: The Role of Language Development. Focus, Occasional Papers in Bilingual Education, No. 4. Washington, DC: National Clearinghouse for Bilingual Education.

Scott, R., and H. Walberg
1979 Schools alone are insufficient: A response to Edmonds. Educational Leadership 37(1):24-27.

Short, D.J.
1994 Expanding middle school horizons: Integrating language, culture, and social studies. TESOL Quarterly 28(3):581-608.

Page 248

Slavin, R., and N. Madden
1994 Lee Conmigo: Effects of Success for All in Bilingual First Grades. Paper presented at the annual meeting of the American Educational Research Association, New Orleans, April. Center for Children Placed at Risk of School Failure, Johns Hopkins University, Baltimore, Maryland.
1995 Effects of Success for All on the Achievement of English Language Learners. Paper presented at the annual meeting of the American Educational Research Association, San Francisco, April. Center for Children Placed at Risk of School Failure, Johns Hopkins University, Baltimore, Maryland.

Slavin, R., and R. Yampolsky
1992 Success for All. Effects on Students with Limited English Proficiency: A Three-year Evaluation. Report No. 29. Baltimore, MD: Center for Research on Effective Schooling for Disadvantaged Students, The Johns Hopkins University.

Slavin, R., N. Madden, L. Dolan, and B. Wasik
1995 Success for All: A Summary of the Research. Paper presented at the annual meeting of the American Educational Research Association, San Francisco, April. Center for Children Placed at Risk of School Failure, Johns Hopkins University, Baltimore, Maryland.

Slavin, R., N. Madden, L. Dolan, and B. Wasik
in press Every Child, Every Schoo: Success for All. Thousand Oaks, CA: Corwin Press.

Stedman, L.
1985 A new look at the effective schools literature. Urban Education 20:295-326.
1987 It's time we changed the effective schools formula. Phi Delta Kappan 69: 215-224.

Sternberg, R.J.
1986 Cognition and instruction: Why the marriage sometimes ends in divorce. Pp 375-382 in R.F. Dillon and R.J. Sternberg, eds., Cognition and Instruction. Orlando, FL: Academic Press.

Stevenson, D., and D. Baker
1987 The family-school relation and the child's school performance. Child Development 58:1348-1357.

Tharp, R.
1989 Psychocultural variables and constants: Effects on teaching and learning in schools. American Psychologist 44:349-359.

Tharp, R., and R. Gallimore
1991 The Instructional Conversation: Teaching and Learning in Social Activity. The National Center for Research on Cultural Diversity and Second Language Learning. Research Report No. 2. Washington, DC: The Center for Applied Linguistics.

Tharp, R.G.
1982 The effective instruction of comprehension: Results and description of the Kamehameha Early Education Program. Reading Research Quarterly 17(4):503-527.

Tikunoff, W.J.
1983 An Emerging Description of Successful Bilingual Instruction: Executive Summary of Part I of the SBIF Study. San Francisco, CA: Far West Laboratory for Educational Research and Development.

Tikunoff, W.J., B.A. Ward, L.D. van Broekhuizen, M. Romero, L.V. Castaneda, T. Lucas, and A. Katz
1991 A Descriptive Study of Significant Features of Exemplary Special Alternative Instructional Programs. Final Report and Vol. 2: Report for Practitioners. Los Alamitos, CA: The Southwest Regional Educational Laboratory.

Trubowitz, S., J. Duncan, P. Longo, and S. Sarason
1984 When a College Works With a Public School: A Case Study of School-College Collaboration. Boston: Institute for Responsive Education.

Page 249

Weber, G.
1971 Inner-City Children Can be Taught to Read: Four Successful Schools. Washington, DC: Council for Basic Education.

Wilson, C.L., P.M. Shields, and C. Marder
1994 The Title VII Academic Excellence Program: Disseminating Effective Programs and Practices in Bilingual Education. Menlo Park, CA: SRI International.

Wong Fillmore, L., P. Ammon, B. McLaughlin, and M. Ammon
1985 Learning English Through Bilingual Instruction. Final Report. Berkeley: University of California.

Page 250

There was a problem loading page 250.