Equity Implications Based on the Conceptions of

Science Achievement in Major Reform Documents

Standards-based and systemic reform has an overarching goal: high academic standards for all students (McLaughlin, Shepard, & O'Day, 1995; Smith & O'Day, 1991). In science education, the goal of "science for all" is emphasized in defining and specifying science content standards by professional science education organizations in international and national contexts. Recent developments in science content standards are reflected in large-scale assessment efforts to align assessment with the new content standards.

The construct of science achievement-what K-12 students should know and be able to do in science-is central to science education reform. Major reform documents in science content standards and assessment frameworks define science achievement for K- 12 students. These documents impact policy decisions for curriculum, instruction, and assessment at state and district levels (Cohen, 1995; Elmore, 1995; Elmore & Fuhrman, 1994; McLaughlin, Shepard, & O'Day, 1995; Porter, Smithson, & Osthoff, 1994; Smith & O'Day, 1991). Policy guidelines, in turn, impact classroom practices through multiple venues (Knapp, 1997). Thus, conceptions of science achievement in major reform documents influence what students are expected to learn and achieve in science classrooms.

Standards-based and systemic reform emphasizes educational equity, along with high academic standards. The focus on equity addresses significant achievement gaps among diverse student groups in terms of ethnicity, language, gender, disabilities, and socioeconomic levels. There is concern that without careful considerations of equity, setting high academic standards may pose additional challenges and learning difficulties for these students (Kahle, 1997; Porter, 1995) and "further victimize students already harmed by gross inequities in the educational system" (McLaughlin, Shepard, & O'Day, 1995, p. 68).

This paper examines equity issues in science achievement and assessment, based on the conceptions of science achievement in major reform documents. The purpose is twofold: (a) to review and analyze conceptions of science achievement in major reform documents with a focus on an aggregated view of science achievement, and (b) to consider equity implications for science achievement and assessment. The paper critiques the perspective of equity in the reform documents and offers alternative perspectives of equity in science achievement and assessment. The discussion occurs in the context of standards-based and systemic reform in large education systems, including state and district levels. While state and district policies and guidelines impact classroom practices, the paper does not specifically address what teachers do to promote science achievement in their instruction and assessment.

The paper consists of four sections, each section reviewing a distinct body of literature. The first section reviews and analyzes conceptions of science achievement in major reform documents, including those on science content standards, performance standards, and assessment frameworks. This review provides the context for considering equity implications in science achievement and assessment in subsequent sections.

The second section describes a conceptual framework of equity that guides the discussion in this paper. Three theoretical perspectives on equity are described, including assimilation, cultural anthropology, and critical theory and postmodernism. After reviewing relevant literature on the assimilation and critical theory perspectives, the paper highlights the cultural anthropological perspective to promote equity in science achievement and assessment for diverse students.

In considering equity implications for science achievement, the third section addresses epistemological debates concerning the nature of science, views of science, and ways of knowing in science. Two issues are discussed: (a) what is the perspective of equity in science achievement in major reform documents? and (b) what are alternative perspectives of equity with diverse students? The discussion highlights the cultural anthropological perspective based on the multicultural and cross-cultural literature in science education.

In considering equity implications for science assessment, the fourth section addresses three issues with diverse students in large-scale assessments: (a) inclusion of all students and assessment accommodations (i.e., who is assessed?); (b) science knowledge and abilities in assessment (i.e., what is assessed?); and (c) different forms of assessment (i.e., how is science knowledge assessed?). Four areas of literature are reviewed, including science assessment, performance or alternative assessment, equity in assessment, and large-scale assessment. The discussion focuses on equity in large-scale assessment from the cultural anthropological perspective.

The paper concludes with implications for achieving equity in science achievement and assessment in large education systems at the state and district levels. It offers a balanced approach that integrates different, sometimes conflicting or opposing, perspectives in promoting science achievement for all students in standards-based and systemic reform.

The major reform documents examined are: (a) science content standards, including National Science Education Standards [NSES] (National Research Council [NRC], 1996) and Project 2061 (American Association for the Advancement of Science [AAAS], 1989, 1993); (b) performance standards in the New Standards Project (National Center on Education and the Economy [NCEE], 1997a, 1997b, 1997c; 1998); and (c) assessment frameworks, including 1996 National Assessment of Educational Progress [NAEP] (National Assessment Governing Board [NAGB], 1994, 1996) and Third International Mathematics and Science Study [TIMSS] (Martin & Kelly, 1996; McKnight, Schmidt, & Raizen, 1993; Robitallie et al., 1993; Schmidt, McKnight, & Raizen, 1997). Related documents and publications by these reform projects are also considered.

The reform documents are selected according to the following criteria: (a) they provide guidelines for standards-based and systemic reform in large education systems; (b) they cover science content for grades K-12; and (c) they are key documents representative of content standards, performance standards, and assessment frameworks in science education. Based on these criteria, the paper does not include some noteworthy efforts. For example, National Science Teachers Association documents (1992, 1995, 1996) are limited to the four science disciplines (biology, chemistry, earth and space science, and physics) traditionally studied in secondary school, grades 6-12. The Advanced Placement tests are administered to a small population of advanced high school students. The National Educational Longitudinal Study (NELS) focused only on 8th grade students before the current reform had been implemented.

In reviewing and analyzing the conceptions of science achievement in the reform documents, the paper addresses two issues: (a) a summary of the conceptions of science achievement and (b) an aggregated view of science achievement.

According to the analysis conducted by Project 2061, there is about 90% agreement in content standards between NSES and Project 2061 (AAAS, 1996, 1997). NSES also states that, "use of Benchmarks ... complies fully with the spirit of the content standards [in NSES]" (NRC, 1996, p. 15). The performance standards in New Standards are "built directly upon the consensus content standards" (NCEE, 1997a, p. 3), particularly NSES and Benchmarks for Science Literacy (p. 130). The assessment frameworks by 1996 NAEP and TIMSS also reflect the recent developments of science content standards. Thus, there is an overall agreement about conceptions of science achievement among the reform documents (for general descriptions about these documents, see Lee, 1998 and Raizen, 1998).

Table 1

The summary of the conceptions of science achievement at the categorical or topical level is presented in Table I (see Lee, 1998, for details). This table is organized using the categories of standards in NSES because this document represents an effort to establish a general agreement of the science and science education communities of over 150 associations in the nation (Collins, 1998). Ile sequence of the eight categories of NSES standards is slightly changed to fit with the other four sets of documents.

An aggregated view of science achievement in major reform documents is presented in Table 2. Five components of science content and four components of science process emerge in these documents (see Lee, 1998, for details). For each component, key indicators are identified, rendering the component more specific and concrete.

In addition to the categories of "content standards" stated in NSES, the table includes "process standards." The distinction between "content" and "process" standards is based on mathematics standards, even though the authors did not state this distinction (National Council of Teachers of Mathematics, 1989; also see AAAS, 1993, p. 209; Romberg, 1998). New Standards, NAEP, and T1MSS identify science process standards. Although NSES and Project 2061 do not identify them as such, the documents emphasize process standards throughout the texts. Content standards generally indicate what students should know (i.e., knowledge and understanding), and process standards indicate what students should be able to do (i.e., abilities and skills). The separation of content and process is a classification decision at one level and an epistemological one at another. This separation emphasizes the importance of doing, the process of knowing through doing, and representation of what is known in the doing (Roth & McGinn, 1998).

Equity in the goal of "science for all" is critically important with increasingly diverse student populations. Traditionally, some groups have not performed well in science and have been underrepresented in science-related careers. The groups generally include students from diverse languages and cultures, students with disabilities, students from low socioeconomic backgrounds, and female students. Patterns of achievement gaps between these groups and mainstream, male students are "alarmingly congruent over time and across studies" based on large-scale databases (Rodrfguez, 1998a; also see National Center for Education Statistics, 1992; National Science Foundation, 1994; O'Sullivan, Reese, & Mazzeo, 1997).

To consider equity implications of the conceptions of science achievement in major reform documents, it is necessary to establish a conceptual framework of equity. Equity is defined in many different ways, and these definitions are often inconsistent and even contradictory (Lynch, in press; Secada, 1994b). In the discussion here, equity is distinguished from equality.

Equity is defined in general terms as * "the quality of being fair or impartial ... that which is fair and just" (Webster's Encyclopedic Unabridged Dictionary of the English Language, 1989). Equity in terms of justice goes beyond the letter of a law to unwritten and evolving notions of justice, as social, political, and economic climates change in a society. According to Rawls (197 1), a just institution is one that equitably distributes social goods, such as rights, liberties, and access to power, among its participants. Based on the notion of equity as social justice, Secada (1994a) states, "equity in education refers to the scrutiny of social arrangements that undergird schooling to judge whether or not those arrangements are consistent with standards of justice" (p. 22).

Table 2: Science Achievement: An Aggregated View from Major Reform Documents

Equity is associated with fairness and justice, whereas equality is associated with sameness or an absence of differences (Grant, 1989; Secada, 1989, 1994a). The distinction between equity and equality, although related, is important in considering science achievement with diverse students (Gallard, Viggiano, Graham, Stewart, & Vigliano, 1998). For example, when a group in power determines what knowledge should count as science and should be taught in school science, without recognition of students' experiences in home and community cultures, it is a rejection of the students themselves and a disregard for their cultures. In the absence of justice or fairness for these students, equal opportunities to learn the way science is defined lead to unjust outcomes for these students. Only after justice is assured, does access to equal opportunities become meaningful. Educational systems often focus on equal opportunities without adequate consideration of social justice to diverse student groups. Secada (1989) emphasizes equity issues in school curriculum for diverse students as follows:

The point is this: equality of education explores quantitative differences between groups. It cannot address those qualitative issues related to the curriculum against which that equality is assessed. It is precisely such issues that must be addressed as part of any effort to ensure educational equity. If we fail to ask whether or not the curriculum is just in what it legitimates as knowledge, we may well achieve equality of education, but it seems highly unlikely that that equality will represent a just distribution of knowledge. (p. 75)

Theoretical Perspectives of Equity in Science Achievement and Assessment

The epistemology or the nature of knowledge in school curriculum is a key issue in considering equity in academic achievement and assessment (Banks, 1993a, 1993b, 1995; Secada, 1989). Because science and mathematics have been regarded as highly objective and universal knowledge, they have been slow in considering issues of equity with diverse students (Banks, 1993a; Parker & Rennie, 1998; Stanley & Brickhouse, 1994; Taylor, 1998). Recently, epistemological debates concerning the nature of science, views of science, and ways of knowing in science have become important in science education (Brickhouse, 1998; Brickhouse & Stanley, 1995; Good, 1993, 1995; Loving, 1995; Matthews, 1998; Nicholls, Gilmer, Thompson, & Davis, 1998; Stanley & Brickhouse, 1994; Taylor, 1998). Stanley and Brickhouse (1994) state, "the definition of what counts as science is at the heart of the curriculum reform debates," especially as it pertains to equity (p. 389).

Two epistemological issues about equity in science achievement and assessment are raised. First, what counts as science and what should be taught in school science? Is Western science the only or the proper domain of science, or are alternative views of science also recognized? Second, is there a way(s) of knowing in science? Is the scientific way of knowing the only valid way, or do alternative ways of knowing also hold potential value?

Three theoretical views provide explanations for equity in science achievement and assessment with diverse students: (a) assimilation; (b) cultural anthropology; and (c) critical theory and postmodernism (e.g., Keller, 1982; Rennie, 1998; Willis, 1996). Each is discussed next.

The assimilationist perspective indicates when individuals from diverse backgrounds adopt the mainstream culture while ignoring or rejecting their cultural backgrounds (Portes & Hao, 1998). According to this perspective, science learning and achievement occurs when students learn the scientific way of knowing, sometimes to the exclusion of alternative views or ways of knowing from their own backgrounds (Good, 1993, 1995; Matthews, 1994; Williams, 1994). This conservative orientation does not consider diversity of students' language, culture, gender, and socioeconomic backgrounds as important because science is represented as universal knowledge. Educational efforts focus on providing students with equal access and opportunities for positive science experiences already available to mainstream males. The goal is to enable students to become members of the science community without changing existing science systems. When disparities abound between Western science and alternative views, however, it would be unjust or unfair if Western science is forced upon students who do not share its values, meanings, or practices. Facing such conflicts, some students may avoid or resist learning science, whereas others may abandon or marginalize their cultural ways of knowing. This concern becomes more serious as the demographics of the student populations in the nation becomes more diverse (Hodgkinson, 1985; U.S. Department of Commerce, 1993).

According to critical theory, postmodernism, and radical feminism, science learning and achievement is a political process that must be addressed to promote equity. As students at the margins gain access to science, they learn to appropriate the language and discourse of science and use it for their own intentions. They transform the nature of science and establish more equitable power structures than the existing systems of hegemony, domination, and oppression (Calabrese Barton, 1997, 1998a, 1998b; Calabrese Barton & Osborne, 1998; Eisenhart, Finkel, & Marion, 1996; Howes, 1998; Keller, 1982; Mayberry, 1998; Rodrfguez, 1997). This radical orientation shifts the focus from a traditional view where science lies at the center to be reached by students at the margins to inclusion where students' identities remain at the center (Calabrese Barton, 1998a, 1998b). Although it is important to recognize and value the lived experiences of diverse students, it would be unjust or unfair if the students do not gain access or opportunities to learn Western science which holds "high status knowledge" in the mainstream culture and the science community. This high status knowledge is important for all students, as society becomes more scientifically and technologically oriented and individuals need to have such knowledge to participate meaningfully in the economy and the workforce ("socially enlightened self-interest" in Secada, 1994a).

This paper considers equity from the cultural anthropological perspective. According to this perspective, science learning and achievement occurs when students successfully participate in Western science, while also engaged in alternative views and ways of knowing in their everyday worlds (Aikenhead, 1996; Cobern & Aikenhead, 1998; Costa, 1995; Gallard et al., 1998; Maddock, 1981; Phelan, Davidson, & Cao, 1991; Pomeroy, 1994). This moderate orientation considers the contributions and strengths of both Western science and alternative views (Lee & Fradd, 1998; Loving, 1997; Stanley & Brickhouse, 1994). Students have access and opportunities to learn the high status knowledge of Western science as it is practiced in the science community and taught in school science. At the same time, alternative views of science and ways of knowing in diverse backgrounds are recognized and valued. By using the language and discourse at home and in the community, diverse students construct the formal language and discourse 6f science.

This moderate orientation is in line with the notions of "biliteracy" and "biculturalism." Research on second language acquisition indicates linguistic and cognitive advantages of biliteracy, in that literacy and proficiency in one language promotes cognitive and metacognitive abilities as well as the acquisition of additional languages (August & Hakuta, 1997; Cummins, 1984, 1986). In a similar manner, the moderate orientation of science and science learning enables students to use the language of science as well as the language of their diverse backgrounds, to understand the culture of science and of their backgrounds, and to behave competently across settings (Fradd et al., 1997; Lee & Fradd, 1998; McKinley, Waiti, & Bell, 1992).

Based on the conceptions of equity discussed in the previous section, two issues concerning equity in science achievement are discussed here: (a) what is the perspective of equity in science achievement in major reform documents? and (b) what are alternative perspectives of equity, particularly according to the cultural anthropological perspective?

NSES and Project 2061 emphasize equity along with excellence as a dual goal of science education reform. NSES (NRC, 1996) underscores equity as the first of its four guiding principles: "Science is for all students. This principle is one of equity and excellence" (p. 20, original emphasis). The premise of Project 2061 is equity, as reflected in the title of the document, Science for All Americans.

In terms of what counts as science and what should be taught in school science, both NSES and Project 2061 define Western science as the proper domain of science. The view of Western science is reflected across the categories of science standards in the reform documents (see Table 2). Because the Western view is particularly pronounced with regard to the history and nature of science, the discussion here focuses on these two areas. In describing the history of science, Science for All Americans (AAAS, 1989) states:

The recommendations in this chapter focus on the development of science, mathematics, and technology in Western cultures, but not on how that development drew on ideas from earlier Egyptian, Chinese, Greek, and Arabic cultures. The sciences accounted for in this book are largely part of a tradition of thought that happened to develop in Europe during the last 500 years-a tradition to which people from all cultures contribute today. (p. 136)

In describing the contributions of diverse cultures to science and technology, NSES recognizes such contributions in terms of technological inventions to solve human problems and needs, but not in terms of the tradition of thought to understand and explain natural phenomena. NSES (NRC, 1996) describes historical contexts in science as follows:

Modem science began to evolve rapidly in Europe several hundred years ago. During the past two centuries it has contributed significantly to the industrialization of Western and non-Western cultures. However, other, non-European cultures have developed scientific ideas and solved human problems through technology. (p. 201)

NSES and Project 2061 also define the nature of science in the tradition of Western science. The documents stress the scientific world view based on the Western tradition, as opposed to alternative world views. According to these documents, science is a way of knowing that distinguishes itself from other ways of knowing and from other bodies of knowledge (AAAS, 1989, pp. 3-5; NRC, 19969 p. 201). Project 2061 states, "there are many matters that cannot usefully be examined in a scientific way. There are, for instance, beliefs that-by their very nature-cannot be proved or disproved (such as the existence of supernatural powers and beings, or the true purposes of life)" (p. 4). NSES also states, "[E]explanation's on how the natural world changes based on myths, personal beliefs, religious values, mystical inspiration, superstition, or authority may be personally useful and socially relevant, but they are not scientific" (p. 201).

Thus, the reform documents define the history and nature of science largely in terms of the Western science tradition, while disregarding alternative views of science. NSES and Project 2061 briefly describe the contributions of non-Western cultures to science and technology. New Standards, 1996 NAEP, and TIMSS do not mention these contributions.

Defining Western science as the proper domain of science, NSES and Project 2061 assume that all students will learn science when provided with opportunities. However, the documents do not offer a coherent conception or a systematic approach to provide such opportunities for all students (Lee, 1997). Instead, the documents describe diverse students' needs and educational strategies sporadically throughout the texts.

Scholars have critiqued reform documents, particularly NSES and the Project 2061, for equity considerations (Calabrese Barton, 1998b; Eisenhart et al., 1996; Rodrfguez, 1997). Rodrfguez (1997) argues that despite an emphasis on equity as the key principle, equity issues are almost invisible in NSES: "The invisibility discourse dangerously compromises the well-intended goals of the NRC by not directly addressing the ethnic, socioeconomic, gender, and theoretical issues which influence the teaching and learning of science in today's schools" (p. 19). In their analyses of major reform efforts, including NSES and Project 2061, Eisenhart, Finkel, and Marion (1996) point out that although the vision of science for all is commendable, the documents do not address obstacles and barriers in implementation:

We applaud this vision of a scientifically literate citizenry in which many and diverse people act in socially compassionate and democratically responsible ways. However, we are concerned that the means being used to promote this vision are too narrowly focused ... These limitations of the current implementation plans will, we think, make achievement of "science for all Americans" difficult. (p. 266)

In summary, the reform documents present the assimilationist perspective of equity. There is little consideration of justice or fairness in recognizing historical contributions or alternative views of science from diverse language, culture, gender, and socioeconomic backgrounds. The documents also fail to offer guidelines in providing equal opportunities for all students. With the notion of "one size fits all," the documents do not consider issues of diversity. Gallard et al. (1998) state, "[T]he reform documents do not address the fundamental problem of including and valuing the experiences, languages, and cultures of all minority students who must learn science" (p. 951). This assimilationist perspective may lead to "epistemological marginalization" of female and non-Western orientations (Nicholls et al., 1998, p. 976).

Alternative perspectives of science and science achievement have been emerging in multiculturalism, cultural anthropology, feminism, critical theory, and postmodernism. Scholars in these areas have raised concerns about power relations and the alienation and marginalization of females and nonwestern Students. They also challenge the basic notion of science and science achievement as traditionally defined and argue for more inclusive notions. Two issues are addressed here: (a) alternative perspectives of what counts as science and (b) alternative perspectives of ways of knowing by diverse students. Based on the cultural anthropological perspective, the discussion draws from the multicultural and cross-cultural literature in science education.

The literature on multicultural science challenges the epistemology or the nature of science as traditionally defined. Ogawa (1995) states, "[T]he science in the slogan 'science for all' is still Western modem science, and such a slogan forces everyone to learn Western modem science alone . . . . Why should we teach Western modem science alone and not other sciences?" (p. 584). The issue is clearly presented in the debate on universalism versus multiculturalism epistemology of science (Brickhouse, 1994; Brickhouse & Stanley, 1995; Good, 1993, 1995; Hodson, 1993; Loving, 1995; Matthews, 1994, 1998; Ogawa, 1995; Siegel, 1997; Stanley & Brickhouse, 1994; Williams, 1994). Stanley and Brickhouse (1994) state, "science education has remained immune to the multiculturalist critique by appealing to a universalist epistemology; that the culture, gender, race, ethnicity, or sexual orientation of the knower is irrelevant to scientific knowledge" (p. 388). Recently, the multicultural science literature questions the dominance of the Western science and, instead, advocates for inclusion of female and nonwestern oriented sciences. The effort centers on making science accessible, meaningful, and relevant for diverse students by connecting their home and community cultures to Western science. Multiculturalist responses to the history and nature of science as traditionally defined and as presented in reform documents are discussed next.

Multiculturalists emphasize the contributions made by nonwestern cultures in the history of science. Needham (1981) claims that the Chinese inventions, including paper making, gunpowder, and the navigational compass, are more than mere technologies. He provides evidence that these inventions were "theoretically driven (although its philosophical basis was radically different from that of western science) and involved observation and careful experimentation" (cited in Hodson, 1993, p. 699). Others have written about scientific accomplishments in Islamic (Sardar, 1989), Indian (Kumar & Kenealy, 1992), and African cultures (Bernal, 1987). From a pedagogical stance, recognizing the contributions of other cultures in science and technology not only motivates diverse students to participate in these areas (Rodrfguez, 1997), but it also provides a broader view of what science is and represents.

The nature of science from the multicultural science perspective has gained increasing attention. Based on Cobem's (1991) comprehensive framework on world views, scholars have identified alternative views of the world in diverse language, culture, and gender groups that are often incompatible with the scientific world view. Western science focuses on explaining, predicting, and controlling nature, and Western culture emphasizes individuality and independence. In contrast, diverse cultures tend to value close, harmonious relationships between humans and nature as well as among' individuals or groups (Hampton, 1991; Hewson, 1988; Pomeroy, 1992; Robbins, 1983). In addition, diverse cultures tend to believe in supernatural forces, spirits, and myths more strongly than the Western culture. These alternative worldviews are in conflict with the scientific way of knowing or scientific worldview. Research results on worldviews are documented across cultures within the US and around the world, including Native Americans in the US (Allen & Crawley, 1998; Kawagley, Norris-Tull, & Norris-Tull, 1998), African Americans and Hispanics in the US (Atwater, 1994; Lee, in press), Aboriginal people in Australia (Christie, 1991), Africans (Jegede & Okebukola, 1991, 1992; Lawrenz & Gray, 1995), and Asians (Kawasaki, 1996; Ogawa, 1995).

According to the multicultural science perspective, narrow definitions of science based on the Western science tradition is misleading and myopic. Scholars argue that alternative perspectives should also be part of science based on both epistemological validity (Hodson, 1993; Ogawa, 1995; Smolicz & Nunan, 1975) and a moral imperative for a just society (Hodson, 1993; Siegel, 1997). Representing a balanced perspective of science (Brickhouse, 1994; Brickhouse & Stanley, 1995; Lee & Fradd, 1998; Loving, 1997), Stanley and Brickhouse (1994) state:

the modem science framework is quite powerful when applied in certain situations. But, Western scientific frameworks cannot provide a vantage point beyond other frameworks whereby we could judge, once and for all, what we can know . . . . Feminists and other cultural critics have been much more useful in showing how the perspectives that have been most frequently excluded are those belonging to marginalized groups in our society. Bringing these kinds of perspectives into science is essential. (p. 395)

All students bring into the science classroom their ways of looking at the world that are formed by their environments and personal experiences (Driver, Asoko, Leach, Mortimer, & Scott, 1994). In considering equity in science achievement, it is important to examine the extent to which the nature of science is compatible or incompatible with the knowledge and experiences of students from diverse backgrounds.

The literature on multicultural science education challenges the way of knowing or learning opportunities in science as traditionally defined. Based on a theory of social justice for multicultural education (Banks, 1993a; Grant & Sleeter, 1986), Rodrfguez (1998b) states, "[T]he basic premise of multiculturalism is that all learners at any grade level must be provided with equitable opportunities for success" (p. 591). Atwater (1996) defines multicultural science education as "a field of inquiry with constructs, methodologies, and processes aimed at proving equitable opportunities for all students to learn quality science" (p. 822, original emphases; also see Atwater & Riley, 1993).

The emerging body of research in multicultural science education indicates that students from diverse language, culture, and gender backgrounds display ways of knowing that are incompatible with the nature of science or the way science is taught in school (Atwater, 1994; Baker, 1998; Baker & Leary, 1995; Barba, 1993; Barba & Reynolds, 1998; Brickhouse, 1998; Calabrese Barton, 1998a; Howes, 1998; Kahle, 1990; Lee & Fradd, 1998; Matthews & Smith, 1994; Rakow & Bermudez, 1993; Rennie, 1998; Rosebery, Warren, & Conant, 1992; Warren, Rosebery, & Conant, 1989). Several -examples below illustrate such differences.

For example, the emphasis on "scientific inquiry into authentic questions generated from student experiences" (NRC, 1996, p. 31) may pose challenges to students from cultures that respect teachers' authority of telling and directing students, rather than promote students' exploration or alternative solutions (Atwater, 1994; Hodson, 1993; Lee, Fradd, & Sutman, 1995; Ogunniyi, 1988; Prophet & Rowell, 1993). Because inquiry is not part of their cultural experiences, the students need to be explicitly taught how to engage in the inquiry process (Delpit, 1988, 1995; Gee, 1997, in press) as they learn to ask their own questions and find answers. As the students develop abilities to investigate and explore in the science classroom, they may recognize conflicts between their school experiences and cultural expectations for respecting authority.

Discourse patterns among diverse groups often differ from the scientific modes of discourse. Rather than developing arguments based on evidence and logic, some groups freely incorporate emotion and personal beliefs to seal an argument (Estrin, 1993; Kochman, 1989). The separation of affect and emotion from cognition is also incongruent with female (Brickhouse, 1994) and cultural groups (Anderson, 1988). Thus, for some groups, shared social and emotional networks may play an equally important role as empirical validation and reasoning.

Cultivation of scientific habits of mind may pose difficulties to diverse students. Although some scientific values and attitudes are found in most cultures, others are more characteristic of Western science that promotes a "critical and questioning stance" (Williams, 1994, p. 517), such as being skeptical, making arguments, openly criticizing, thinking independently, and tolerating ambiguity or uncertainty (AAAS, 1989). These values and attitudes may be incongruent with the norms of diverse cultures that favor cooperation, social and emotional support, consensus building, and respect for authority (Atwater, 1994; Hodson, 1993; Lee & Fradd, 1996; McKinley, Waiti, & Bell, 1992). The students may have difficulty developing scientific values and attitudes, while retaining their cultural norms (Aikenhead, 1996; Cobern & Aikenhead, 1998; Costa, 1995; O'Loughlin, 1992; Phelan, Davidson, & Cao, 1991).

Although the distinction between the scientific world view and alternative views may be relatively straightforward to educated adults, children's world views involve a complex interaction of personal and supernatural beliefs with scientific understanding (Cobern, 1991; Hewson, 1988; Lee, in press; Ross & Shuell, 1993). In addition, students from diverse backgrounds tend to express alternative world views more strongly than their mainstream counterparts (Allen & Crawley, 1998; Hewson, 1988; Jegede & Okebukola, 1991, 1992; Kawagley, Norris-Tull, & Noffis-Tull, 1998; Lawrenz & Gray, 1995; Lee, in press). For example, after personal experiences of a major natural disaster, African American and Hispanic elementary students attributed the cause of the disaster to societal problems (e.g., race, crime, violence) and spiritual and supernatural forces (e.g., god, devil, or evil spirits) more strongly than their Anglo counterparts who tended to give explanations as natural phenomena (Lee, in press).

While ethnicity and gender have received attention in the literature, social class and poverty have been largely ignored. Because ethnicity is often correlated with social class, social class issues are implied in discussions of ethnicity. However, social class and poverty present serious challenges in making science meaningful and relevant. Not only do these students have alternative views of science and ways of knowing, they often do not have access or opportunities to learn Western science because of poverty (Calabrese Barton, 1998a, 1998b). Even when these students have access, -the discourse and culture of science may be so foreign to them to render learning extremely difficult or almost impossible. Thus, school knowledge may contribute to the cycle of the reproduction of unequal social class structures (Anyon, 1981).

As the examples above illustrate, the way of knowing in science and alternative ways of knowing in diverse cultures may sometimes be incompatible. At times, students may find themselves caught in conflicts between what is expected in science and school science and what they experience at home and in the community. Science educators propose a balanced approach in enabling the students to acculturate to mainstream Western science while at the same time to incorporate the unique views of their home and community cultures (Lee & Fradd, 1998; Loving, 1997; Stanley & Brickhouse, 1994). This balanced approach is in line with the cultural anthropological perspective. According to this perspective, science learning and achievement occurs when students: (a) acquire the norms, values, practices, and discourse of the culture of science, (b) recognize and value the ways of knowing in their home and community cultures, and (c) learn to make transitions between the culture of science and their everyday worlds (Aikenhead, 1996; Cobern & Aikenhead, 1998; Costa, 1995; Gallard et al., 1998; Giroux, 1992; Maddock, 1981; Phelan et al., 199 1; Pomeroy, 1994).

In summary, what counts as science raises a serious question about equity in science achievement. The Western science tradition as currently practiced in the science community and taught in school science presents "high status knowledge" to which every student should have access. At the same time, students from diverse backgrounds bring to the learning process alternative views of science and ways of knowing. This presents a challenge. On the one hand, an emphasis on the high status knowledge without consideration of alternative views makes science less accessible, relevant, or meaningful for these students who have generally been bypassed in science education. On the other hand, an emphasis on alternative views that are culturally and socially significant but may be unimportant science topics in the science community and in school science does not promote equitable outcomes. A balanced approach incorporates alternative views in defining what counts as science and what should be taught in school science.

At the heart of standards-based reform is the alignment of assessment with content standards (McLaughlin et al., 1995; Smith & O'Day, 1991; Webb, 1997a, 1997b). The National Academy of Education panel report states, "The intention of standards-based reform is to set higher standards for all students ... New kinds of assessment reflecting these new standards are seen as instrumental in effecting the reform" (McLaughlin et al., 1995, p. 52). While expecting high academic standards for all students, equitable and fair means for all students to demonstrate their achievement also need to be provided (Webb, 1997a, 1997b). In a broad sense, equitable assessment suggests the ideal of 'Just' measures in educational and social conditions (Madaus, 1994).

This paper considers equity implications of reform documents from the cultural anthropological perspective. According to this perspective, assessments need to consider the knowledge and abilities of diverse students in their home and community cultures, while measuring the science standards expected of all students. Assessments also need to allow diverse students to demonstrate their knowledge and abilities in ways compatible with their backgrounds, as they also acquire the ways expected by the mainstream culture and the science community.

Three issues are discussed with regard to equity in large-scale assessments. First, with the goal of achieving high academic standards for all students, large scale assessments include diverse students and provide accommodations (i.e., who is assessed?). Second, a basic issue of equity in assessment involves the epistemology or the nature of science knowledge and abilities (i.e., what is assessed?). Third, a recent discussion has focused on forms of assessment, particularly performance assessment (i.e., how is science knowledge assessed?). With regard to each of these issues, equity implications of major reform documents are addressed, including New Standards, 1996 NAEP, and TIMSS. The discussion here indicates that empirical research is very limited to show whether new assessment technologies and innovations either diminish or magnify existing inequities (August & Hakuta, 1997; Madaus, 1994; McDonnell, McLaughlin, & Morison, 1997).

Until very recently, many students with special needs, including English language learners and students with disabilities, have been excluded from most large-scale assessments. Over the past several years, large-scale assessments increasingly include diverse students and provide accommodations (August & Hakuta, 1997; McDonnell et al., 1997). Inclusion and accommodations have major policy implications for equity, while also creating technical problems.

Rodrfguez (1998a) points out that beyond overall achievement results by ethnicity or gender, there is little information about disaggregation of results for gender-by-ethnic groups (e.g., African American male and female students) or subgroups within an ethnic group (e.g., Mexican Americans, Chicanos/as, Puerto Ricans, and students from various Latin American countries within the generic category of "Hispanics"). He argues that this lack of information fails to provide important insights about achievement gaps by specific groups. He also argues that generic ethnic categories may create or reinforce stereotypes about a certain group without considering differences among subgroups or individuals. For example, the model minority stereotype for Asian American students, particularly in mathematics and science, masks great disparities and challenges facing many students, including Southeast Asian refugees with little schooling or limited literacy development in their home countries (Lee, 1996; Tobin & McRobbie, 1996), In contrast, high-achieving Hispanic students may be at disadvantage because of lower expectations of teachers and school personnel (Rodrfguez, 1998a).

'Mere is also inadequate information about specific populations, such as students with disabilities and English language learners. These students are often exempted from large-scale assessments at the state and district levels (August & Hakuta, 1997; McDonnell et al., 1997; McLaughlin et al., 1995). The consequence, in terms of equity, is that this practice "literally creates a kind of systemic 'ignorance' about the educational progress" of these students and "leaves the school, district, or system utterly unable to account for the learning of these students" (Lacelle-Peterson & Rivera, 1994, p. 70).

Recently, large-scale assessments tend to include more students with disabilities and English language learners (George & Van Home, 1996; Glaser & Linn, 1997). The impact of the inclusion of students who have been pushed aside and ignored in the accountability systems in the past should not be underestimated (McLaughlin et al., 1995). As a policy, inclusion indicates that these students can claim their rights to be full beneficiaries of education reform efforts and that the educational system is accountable for the academic progress and achievement of all students (Shepard, Taylor, & Betebenner, 1998). The inclusion also ensures the representativeness of the outcomes of the educational system.

When students with special needs participate in assessments, accommodations are made to ensure that the students can demonstrate their knowledge and abilities accurately. The 1996 NAEP in science included students with disabilities and English language learners on the basis of a set of inclusion criteria to assess "the achievement of all students at a given grade or age" (O'Sullivan et al., 1997, p. 55, original emphasis). The 1996 NAEP offered various assessment accommodations for both student groups (O'Sullivan et al., 1997). These accommodations included one-on-one testing, small group testing, extended time, oral reading of directions, signing of directions, enlarged versions of test booklets, use of magnifying equipment, use of an individual to record answers, a Spanish/English glossary of science terms, and bilingual dictionaries. A report on the efficacy of accommodations on assessment results is planned (O'Sullivan et al., 1997).

Although English language learners are more likely to be assessed than students with disabilities in large-scale assessments, English language learners are less likely to be given accommodations (Butler & Stevens, 1997; National Center for Research on Evaluation, Standards, and Testing, 1997). Assessments for English language learners need to distinguish their science achievement, English language proficiency, and general literacy. To measure science achievement, assessments need to be done in the language of instruction with special assistance in their first language (August & Hakuta, 1997; Fradd & Larrinaga McGee, 1994). Even using bilingual assessments, it is difficult to distinguish the students' science achievement from their English proficiency and general literacy (Ruiz-Primo & Shavelson, 1996; Shaw, 1997).

There is little empirical research on the uses and effects of accommodations in large-scale assessments for students with disabilities (McDonnell et al., 1997) and English language learners (August & Hakuta, 1997; Butler & Stevens, 1997). The limited available research indicates that students who received accommodations performed better than those without accommodations (Koretz, 1997; Shepard et al., 1998). Overall, both student groups were not as far below general education students on performance assessment. More detailed analysis, however, reveals that the scores were implausibly high with inappropriate uses of accommodations (Koretz, 1997; Shepard et al., 1998). The results suggest validity issues-accommodations may fail to correct unfair disadvantage, or overcompensate to give unfair advantage. In both cases, accommodations reduce the validity of assessment results.

In summary, the exclusion of diverse students in large-scale assessments in the past raises skepticism about assessment results on which claims about science achievement are made. Recent policies for inclusion and accommodations have major significance for all students. Educational systems are accountable for the academic progress and achievement of all students. While research on inclusion and accommodations involving English language learners and students with disabilities is just beginning, little consideration is given to social class issues in large-scale assessments. From the equity perspective, all students should count in assessments and need be provided with accommodations to demonstrate their knowledge and abilities accurately.

As presented in Tables I and 2, there is a significant level of alignment between the content standards in NSES and Project 2061 and the assessment frameworks for New Standards, 1996 NAEP, and TIMSS. Consistent with how science is defined in the standards documents, the knowledge and abilities in the assessment frameworks are defined in terms of Western science with little consideration of alternative views of science or ways of knowing.

Within the Western science tradition, science content standards in the reform documents include a broad range of knowledge and abilities that are personally and socially relevant and meaningful (see Table 2). Assessment based on these standards (i.e., standards-based assessment) presents both promises and fears with diverse students (Champagne & Newell, 1992; Garcfa & Pearson, 1994; Lynch, in press; McLaughlin, Shepard, & O'Day, 1995; Ruiz-Primo & Shavelson, 1996). Proponents argue that assessments based on meaning and relevance, rather than discrete knowledge from textbooks, may narrow achievement gaps among ethnic, socioeconomic, and gender groups. Also, authentic tasks drawn from students' real-life situations may motivate and enhance their performance. In contrast, skeptics claim that standards-based assessment may widen the achievement gaps because open-ended tasks and application of knowledge to novel situations may differentially favor students with many opportunities to participate in science-rich environments of home and community over those lacking such opportunities. Assessment tasks based on the content and experiences within the classroom are fairer than those requiring knowledge and abilities in novel situations.

Although empirical evidence supporting either position above is very limited, Shavelson, Baxter, and Pine (1992) and Hamilton (1998) offer important insights for standards-based assessment with diverse students. Shavelson et al. (1992) report that assessment tasks that were closely related to instructional experiences seemed most sensitive to equity concerns. In the Hamilton (1998) study, gender differences were most prominent on short essay items that had not been taught directly in school, whereas the differences were least prominent or non-existent on items based on in-school experiences. Thus, regardless of using performance tasks (Shavelson et al., 1992) or traditional multiple-choice and short essay items (Hamilton, 1998), achievement gaps among ethnic, socioeconomic, and gender groups tend to be larger on items that call on outside-of school knowledge and experiences.

One way to promote equitable assessments, based on the results in Shavelson et al. (1992) and Hamilton (1998), is to include tasks and items closely tied to school science experiences. This, of course, requires that all students have access and opportunities for quality science instruction. The importance of school based tasks and items in assessment raises concerns in measuring applications and problem-solving in novel situations with diverse students because they generally have limited science experiences and opportunities in their home and community environments. The more assessment tasks are removed from school experiences, the wider achievement gaps may be between students from science-rich environments and those with limited opportunities. This has major significance to students from low social class or poverty, because they often have limited access and resources both at home and in school.

Another way to promote equitable assessments is to incorporate tasks and activities that are based on the knowledge and experiences of diverse students in their home and community environments. It is possible to select such tasks and activities for students who share a common background. However, in large scale assessments involving students from "a wide diversity of backgrounds, it is increasingly difficult for test-makers to find contexts which are equally meaningful to all students" (Parker & Rennie, 1998, p. 905). Despite this difficulty, efforts need to be made to include equitable representations of tasks and activities across diverse backgrounds.

Some scholars consider equity issues of science knowledge and abilities in assessment within the larger structure of power in society. Madaus (1994) argues that although assessment technology is not by nature socially unjust, it is intertwined with the distribution of wealth, ethnicity/race, and gender relations. As more advanced technologies in assessment are introduced and developed, elite groups determine what is to be assessed and how it is assessed. This technical power leads to political power by allowing those in power to control information sources, while keeping the "uninformed" public away from the decision-making process. Thus, despite potential benefits of new technologies, all technologies create problems in unanticipated ways. Negative impacts are most deleterious with groups that are not from the culture of power. Madaus (1994) cautions that "testing as a technology has the potential to perpetuate current social and educational inequalities" (p. 76).

Rodrfguez (1998a) addresses achievement gaps among ethnic and gender groups in terms of social injustice in the educational system at large. Contrary to the notion of meritocracy-all students who work hard get proper rewards the educational system is structured in ways to benefit those in power. Unfortunately, "the students most adversely affected by the meritocracy myth come from the fastest growing ethnic groups" (Rodrfguez, 1998a). He claims that to promote participation and achievement of minorities and female students in science, the meritocracy myth needs to be exposed and dealt with.

In summary, the science knowledge and abilities to be assessed have major significance with diverse students. From the cultural anthropological perspective, assessments need to measure the science content standards expected of all students, while also recognizing alternative views of science and ways of knowing from diverse backgrounds. Defining science in the Western science tradition, the assessment frameworks for New Standards, 1996 NAEP, and TIMSS fail to consider the knowledge and abilities of diverse students in their home and community environments. In fact, large-scale assessments face a dilemma - in order to determine what students know and can do with science in local settings, it becomes difficult to include culturally relevant examples and contexts that are equally meaningful and relevant to all students in large-scale assessments. Ways to promote equitable representations of science knowledge and abilities across diverse backgrounds need to be established.

Much of the efforts in assessment centers on techniques and procedures. Traditional multiple-choice tests have been widely criticized because they fail to measure the types of knowledge and abilities called for in current reform and have had an overall negative impact on teaching and learning. A major development in assessment involves alternative forms of assessment. Although a variety of terms are used and meaning * s vary among the terms (Garcfa & Pearson, 1994; Newmann, Brandt, & Wiggins, 1998; Terwilliger, 1997, 1998), the term "performance assessment" is most commonly used as relatively value-neutral.

Recently, performance assessment has been used beyond the classroom and applied to large-scale implementation. In addition, performance assessment has been heralded for diverse students who generally have performed poorly in science. These advances raise challenges and dilemmas, to be discussed next.

The forms of assessment in large-scale assessments, such as NAEP and TIMSS, have changed significantly in recent years. Traditionally, NAEP science assessments used mostly multiple-choice items with some open-ended items. The Second International Science Study [SISS] also used multiple-choice items exclusively (International Association for the Evaluation of Educational Achievement, 1988). In response to the current emphasis on scientific understanding, 1996 NAEP and TIMSS included open-ended, free-response items (including both short-answer and extended-response items) as well as multiple-choice items. In addition, along with the emphasis on scientific investigation and communication, 1996 NAEP for the first time included performance exercises (also called hands-on tasks) (O'Sullivan et al., 1997, p. 42). TIMSS also used performance tasks (also called hands-on activities) with a sub-sample of students (Martin & Kelly, 1996).

In performance assessment in 1996 NAEP and TIMSS, students manipulate materials, take measurements, conduct investigations, and communicate their observations and results. Such assessment comes closer to problem-solving in real world situations than seeing pictures of an experimental design and answering questions on paper. Despite its critical importance in standards-based assessment, performance measures in large-scale assessments pose limitations. First, because of the need for standardization, performance tasks are provided for students along with the materials to use, procedures to follow, graphs or tables to report the data, and questions to answer, This standard procedure does not allow students to ask their own questions, design and conduct investigations, and communicate observations or results in their own ways. Second, performance tasks in 1996 NAEP and TIMSS are administered in controlled settings. This constraint does not allow scientific investigations of natural events as an on-going process. These limitations are inherent in external, large-scale assessments.

New Standards focuses on performance assessments that teachers can use in the classroom, and that states and urban school districts can use in systemic reform. As part of science instruction in the classroom, New Standards does not have the constraints of NAEP or TIMSS. The samples of student work in the New Standards documents (NCEE, 1997a, 1997b, 1997c, 1998) indicate that the assessment system effectively measures many components of science achievement. These components include scientific investigation and communication, as well as understanding key concepts in three fields of science and unifying themes across the fields. In contrast, the assessment system rarely addresses technology, science in personal and social perspectives, history and nature of science, and scientific habits of mind. Although New Standards offers important insights for large-scale assessments, the more flexible and varied assessment activities and settings are, the more difficult it becomes to develop standard procedures within the constraints of large-scale implementation (e.g., Lomask, Baron, & Greig, 1998).

Because performance assessment is a recent movement, little research has been conducted on its efficacy or impact on any student population, let alone students from diverse language, culture, gender, and socioeconomic backgrounds (Champagne & Newell, 1992; Darling-Hammond, 1994; Garcia & Pearson, 1994; Lacelle-Peterson & Rivera, 1994; Lynch, in press; Madaus, 1994; McLaughlin et al., 1995; Ruiz-Primo & Shavelson, 1996; Shaw, 1997). Discussions of advantages and disadvantages of performance assessment with diverse students, compared to traditional multiple-choice tests, are largely based on inferences and insights from related research endeavors (Champagne & Newell, 1992; Darling-Hammond, 1994; Garcia & Pearson, 1994; Klein, 1997; Lacelle-Peterson & Rivera, 1994).

Advocates of performance assessment charge that traditional multiple-choice tests are biased in terms of ethnicity, gender, and socioeconomic levels (Darling-Hammond, 1994; Garcia & Pearson, 1994; Lacelle-Peterson & Rivera, 1994). They argue that these tests generally reflect the mainstream culture, contain content bias, incorporate linguistic and cultural bias, and fail to adequately include diverse student groups in the norming process. Instead, they advocate that performance assessment provides diverse students with flexible and multiple types of assessment settings, allows the students to participate in assessment activities consistent with their cultural preferences, and enables the students to communicate ideas in multiple ways that may not occur in a particular standard format. Multicultural education literature supports the claim that diverse students perform differently in assessment settings (Champagne & Newell, 1992; Garcia & Pearson, 1994; Parker & Rennie, 1998; Shaw, 1997). For example, students from diverse cultures and girls perform better in situations promoting cooperation, whereas mainstream boys perform better in individual, competitive situations.

Some educators, including advocates of performance assessment, also point to difficulties and limitations. For example, performance assessment tends to rely heavily on students' ability to read and write standard English, confounding literacy skills with content knowledge (Simmons & Resnick, 1993). This is particularly problematic for students learning English as a new language as well as those who come from non-standard English backgrounds (Abedi, Lord, & Plummer, 1997; Garcia & Pearson, 1994; Lacelle-Peterson & Rivera, 1994). Others express a concern that "because new formats do not allow [diverse] students to get by with rote memorization, they will reveal weaknesses in ability to think or communicate" (Champagne & Newell, 1992, p. 853). Thus, performance and other alternative assessments have the potential either to be fairer or to magnify existing inequities (Champagne & Newell, 1992; Monty Neill & Medina, 1989).

The limited available research does not support the contention that performance assessment is more equitable with diverse students than traditional multiple-choice tests. Generally, patterns of achievement gaps among ethnic and socioeconomic groups tend to be the same on both traditional multiple-choice tests and performance assessment in science (Lynch, in press; Stecher & Klein, 1997).

In summary, forms of assessment are important in enabling all students to demonstrate their knowledge and abilities accurately. From the cultural anthropological perspective, assessments allow diverse students to demonstrate what they know and can do with science in ways compatible with their backgrounds. Noting the weaknesses inherent in traditional multiple-choice tests, proponents of performance assessment highlight potential advantages for diverse students. The limited available research, however, suggests that achievement gaps among ethnic and socioeconomic groups tend to be the same on both forms of assessment.

Standards-based and systemic reform aims at developing a unifying vision of high academic standards for all students. The reform documents are as much political statements as they are educational goals and road maps (Collins, 1998; Kirst & Bird, 1997; Massell, 1994). Science standards documents, such as NSES and Project 2061, influence science curriculum frameworks in almost all 50 states and numerous school districts (American Federation of Teachers, 1997; Council of Chief State School Officers, 1997).

Tensions and dilemmas abound in conceptualizing and promoting science achievement for all students. For example, NSES emphasizes scientific inquiry as central to science, whereas Project 2061 underscores scientific understanding (Lee, 1998; Raizen, 1998). Although the current reform emphasizes a small number of key ideas in greater depth-the principle of "less is more"-the task given to schools and teachers with 855 benchmarks in Project 2061 or 77 sections of learning goals in NSES still poses a daunting challenge. Thus, even within the traditional science education community, difficult decisions need be made about the scope and priorities of science standards (e.g., see the California case in Olson, 1998).

Efforts to promote equity in science achievement present serious tensions. Reform documents define science and science achievement according to the assimilationist perspective. Based on moral and ideological as well as epistemological grounds, educators argue for reconceptualization of science and science achievement. This argument would force the science community to consider alternative perspectives and change the fundamental fabric of extant science curricula. Even if a balanced approach could be adopted, it might require a substantial redefinition of what science is and pose difficulties in combining traditional and alternative perspectives of science (Brickhouse & Stanley, 1995; Lee & Fradd, 1998; Loving, 1997; Stanley & Brickhouse, 1994). It would also require redistribution of opportunities and resources in the educational system (Clune, 1998; Donmoyer, 1995).

There is a clear indication that large-scale assessments, such as NAEP and TIMSS, influence state- and district-level assessments (George & Van Home, 1996; Glaser & Linn, 1997; National Center for Research on Evaluation, Standards, and Student Testing, 1997). NAEP science assessments have been used for state-level results since 1990 (Allen, Swinton, Isham, & Zelenak, 1998; Jones, 1996; Jones, Mullis, Raizen, Weiss, & Weston, 1992). States also have expressed an interest in creating linkages that allow comparisons of state assessments with national and state-level NAEP (Glaser & Linn, 1997). Some states incorporate released items from NAEP and TIMSS in their assessment programs, compare their achievement results with those of other states or countries, and analyze their science curricular and teaching practices (Champagne, 1997). As systemic reform continues and is likely to intensify (American Federation of Teachers, 1997; Council of Chief State School Officers, 1997; McLaughlin et al., 1995), the central role of assessment in evaluating the impact of standards based and systemic reform on student achievement will increase at state and district levels.

Equity issues in standards-based assessment present tensions. There is limited empirical evidence about whether standards-based assessment will remedy or exacerbate inequities with diverse -students. Even when epistemological and conceptual issues in assessment are resolved, there is a practical concern that performance assessments are more complex to implement and interpret than traditional multiple-choice tests. Performance assessments are also considerably more expensive than traditional tests (Klein, 1997).

Given such tensions and dilemmas, what do educators do? In reality, educators are being required to integrate different, sometimes conflicting or opposing, perspectives into a workable model within the reality of the educational system, the school, and the classroom. For example, because the traditional perspective of science indicates high status knowledge, all students should have access and opportunities to learn. This goal, however, may not be possible for some students, particularly those who have traditionally been under-served, unless science is made meaningful and relevant to their lived experiences based on culture, language, gender, and socioeconomic backgrounds. The cultural anthropological perspective provides a more inclusive and broader view of science achievement and assessment for diverse students.

It is difficult enough to develop reasonably agreed-upon conceptions of equity in science achievement and assessment. It is even more difficult to develop and implement policies and plans that incorporate different perspectives of equity in science achievement and assessment. Beyond ideological and political issues, planning and implementation require practical matters of opportunities and resources. At every step, difficult choices and trade-offs have to be made. Despite such tensions and difficulties, it is imperative that educators emphasize equity at the center of standards-based and systemic reform. Without concerted efforts to ensure social justice as well as equal opportunities for all students, the vision of standards-based and systemic reform will remain as rhetoric and not become reality.

This work was supported by a cooperative agreement between the National Science Foundation and the University of Wisconsin-Madison (Cooperative Agreement No. RED 9452971). At UW-Madison, the National Institute for Science Education is housed in the Wisconsin Center for Educational Research and is a collaborative effort of the College of Agricultural and Life Sciences, the School of Education, the College of Engineering, and the College of Letters and Science. The Cooperative effort is also joined by the National Center for Improving Science Education, Washington, DC. Any opinions, findings, or conclusions are my own and do not necessarily reflect the views of the supporting agencies.

This paper is part of the author's fellowship at the National Institute for Science Education. The author acknowledges the generous support of Norman Webb, team director for the Strategies for Evaluating Systemic Reform project, and Andrew Porter, Director, both at the National Institute for Science Education. The author also appreciates the comments of Sandra H. Fradd at the University of Miami, Kenneth Tobin at the University of Pennsylvania, and Sharon Lynch at George Washington University. Many others have been helpful throughout the preparation of the monograph. Finally, the author wishes to thank the reviewers for their valuable comments and suggestions.

References must be obtained from the original