Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
4 Research on Settings Formal instruction takes place in the classroom, but the classroom is not isolated from the rest of society. Classrooms are subject to school policies and explicit and implicit educational goals. Schools, in turn, are shaped by the political and social context in which they operate. Moreover, school is not the only acting for learning: children come to class shaped by their homes and by informal learning done outside school. RESEARCE ON CLASSROOM SETTINGS Repeatedly in the history of American education the prevalence of routine instruction has been criticized, attempts to change the situation have been undertaken. and those attempts have failed. Perhaps the problem is more fundamental than previously conceived. The roots of the problem may go deep into the ecology of everyday teaching practice. Descriptive studies suggest that the organization of American classrooms is largely based on teacher-led whole groups (Dunkin and Biddle, 1974). Teacher-led subgroups tend to occur primarily in early grades, particularly in reading instruction (Cazden, 1983). Small groups of students working together cooperatively are infrequent in today's school--limited to 2 percent of students, according to one study of 129 elementary school class- rooms (Sirotnik, 1981)--and are more common in social studies and science than in other subject areas (Stodolsky, 19841. Promising classroom strategies can take hold only where the wider school setting provides a favorable climate. Considerable work has been done on the relationship of alternative strategies of classroom 26
27 organization to student achievement through the production of active learning time. An extensive study describing the teaching and learning of mathematics and reading in grades 2 and 5 (Fisher et al., 1980) found that more substantive interaction between students and an instruc- tor is associated with higher levels of student engage- ments, and that class lessons depending on seat work do not permit the right kind of substantial interchanges crucial to effective learning time. Some strategies based on behaviorist principles have been developed for improving student engagement during seat-work lessons and appear to have been effective for learning simple skills. (See, for example, Rosenshine, 1979, and Gersten, 1984, for syntheses of findings from the models tested in the "Follow-Through" program.) The solution more generally adopted by teachers is to break the class into smaller groups for instruction. This solution, however, is fraught with problems. A large body of research on the organization of class- room activities focuses on the effects of different principles of grouping for instructional purposes. Group placement based on ability is common to reduce hetero- geneity and match instruction more effectively to learners' skills. Such placement has been shown to be stable over time (once in the low group, it is hard to get out) and to have differential impact on high- and low-group students. Studies consistently and robustly document that ability grouping has detrimental effects within classrooms on average and low-ability groups. Persell (1977:92), in a review of 217 studies, found that n there is a slight trend toward improving the achievement of high-ability groups but that is offset by substantial losses by the average and low groups.. Differences in instruction across high and low reading groups, with respect to both content and the quality of interaction, have been found to n sustain the poor performance of slower students and to increase the disparity between the two groups" (Good and Marshall, 1984:18). Inappropriate grouping may amplify relatively minor differences at the beginning of first grade into major differences in later grades (Hallinan and Sorensen, 1984). Both the literature on teacher-student ratio (e.g., Glass et al., 1982) and that on effective learning time (e.g., Denham and Lieberman, 1980) urge a reduction in the size of the instructional group representing the primary teaching unit. But neither of these research traditions motivates any particular principle of group-
28 ing. On the basis of a review of relevant studies of children organized to work in peer groups, Stodolsky (1984) concludes that Children working together produced problem solutions characterized by higher cognitive levels of response than individual children could produce.. There is also evidence that peer work groups need not be homogeneous, that cooperative, mixed-ability group processes seem genuinely to enhance learning and cognitive development In some circumstances (Sharan et al., 1984; Slavin, 1978.) Evidence about the proper mix of circumstances is sparse, but some principles have emerged. First, the school principal and school administration must support alternative classroom behavior that does not fit the stereotype of quiet children obviously controlled by the teacher. Second, the care with which tasks are designed and materials prepared is even more important in peer work groups than in teacher-led groups precisely because the teacher is only intermittently available. Third, the learning of curriculum content in a peer work group is positively related to the frequency of interaction within the group, and that in turn is correlated with social status in the classroom (Cohen, 1984). Thus, it appears that cooperative groups do not necessarily ensure equal exposure to learning. Differential treatment can come as readily from peers as from a teacher. However, some experimental evidence indicates that, by manipulating the activities of the individuals involved in activity groups, it is possible to change their perceived status evaluations and bring about positive educational outcomes (Kagan et al., in press; Cole, 1985). Constructing effective learning groups in the class- room holds great promise for increasing quality learning time devoted to higher-order skills. Therefore, the committee urges research on how to make student activity groups successful in multi-ethnic classrooms for a range of mathematics and science tasks, including improved understanding of the ideological and pragmatic reasons teachers group their students by ability and prefer teacher-led groups to cooperative student-led groups; investigating systemic factors relating to societal and institutional pressures on schools and teachers to arrange their classrooms and instruction so as to produce easily measurable performance results; and developing kinds of teacher training that facilitate widespread adoption of activity-centered curricula when this approach is appropriate.
29 Linguistic and cultural factors can serve as resources for changing contextual and motivational factors that promote educational excellence. For example, Erickson and Mohatt (1982), working among the Odawa in Canada, found that Odawa teachers were spontaneously and intu- itively adapting instruction in culturally responsive ways by using discourse modes prevalent in the children's community. The phenomenon that Erickson and Mohatt addressed was the apparent passivity and silence of Native American students in regular classrooms (Phillips, 1972). Erickson and Mohatt showed that it is possible to construct rules of participation in the classroom that are a functional blend of Anglo school curriculum and Native American discourse styles and that make the classroom run more effectively. Moreover, it appears that these patterns can be learned; an Anglo teacher was observed to change his rules for classroom participation over the course of the school year in the direction of the style of instruction used by the Odawa teachers. Classroom settings can be transformed for the better by taking students' language and culture into account. The best documented example is the decade-long research and development effort at the Ramehameha Early Education Program (BEEP) in Hawaii (Tharp, 1982). The teacher allows the children to discuss text ideas and therefore learn to read, using rules for speaking and turn-taking similar to those familiar to their culture, particularly overlapping speech and the cooperative production of narrative. In a series of related studies Moll and Diaz (1980, 1982, 1984) analyzed student and teacher language used in reading lessons in two bilingual classrooms. Mall and Diaz concluded that students' reading skills in their native language were seriously underestimated and were not being effectively taken advantage of in the second language setting because the teacher was mistakenly aiming the lessons at the students' oral skills and not their reading skills. They reorganized reading lessons so as to permit students to rely on and display reading skills acquired in their native language, at the same time acquiring advanced reading skills in their second language. Most of the research on language and culture is focused on instruction in literacy. Relationships of culture and language to instruction in science and mathematics are barely understood. Cazden (1979) has speculated about the benefits and liabilities associated with the use of students' first or second language in
30 science and mathematics. Science instruction that involves laboratory investigation may be a particularly good environment for learning in a second language because of the presence of concrete referents of objects and operations. However, as far as the role of culture is concerned, present knowledge does not even allow one to speculate. To further the goal of developing more effective mathematics and science instruction for all students, the committee urges research and development to explore the relationships among the cultures of various student subpopulations, the culture of the classroom, and the cultures of mathematics, science, and technology; and research to understand the role of language and culture in the teaching of science and mathematics. RESEARCH ON THE POLITICAL AND SOCIAL CONTEXT OF MATHEMATICS AND SCIENCE EDUCATION ffl e goals of mathematics, science, and technology education as stated by planning commissions and scien- tists consistently emphasize reasoning, thinking, and problem-solving skills (see, for example, National Council of Teachers of Mathematics, 1980; National Science Founda- tion, 1983; Task Force on Education for Economic Growth, 1983; American Chemical Society, 1984). Some of the research we cite shows that such an education is possible but that it is not typically achieved in contemporary American education. Apparently, an insufficient portion of the education experienced by most students is aimed toward developing reasoning skills. m is deficiency may be even more severe among minority and female students in that they are less frequently enrolled in higher-level mathematics and science courses. The question is why the recommendations and policies of special commissions, boards of education, and super- intendents about higher-order thinking skills appear to have so little effect on day-to-day classroom activities. What are the barriers that keep political and educational institutions from fully grappling with the achievement of stated goals in mathematics and science education? Are they to be found in the deployment of resources, the apportionment of responsibility among the several gov- ernance structures, local district policies and operating procedures, the decisions of individual teachers on what to teach?
31 m e resources used for instruction in mathematics, science, and technology are provided by the combined contributions of federal, state, and local governments. Historically, the bulk of these resources and the rules by which they were to be allocated were determined locally. In recent decades, however, both the federal government and, even more importantly, state governments have begun to play an increasingly significant role--both in terms of financial resources and as sources for new initiatives and undertakings. Thus, between 1971 and 1980, the state contribution to revenue receipts of public schools rose from 38.3 to 47.4 percent, while the local share decreased from 52.8 to 43.3 percent; the f ederal share stayed relatively level (National Center for Education Statistics, 1983). It is often assumed that an increased role in funding brings with it increased authority for policy. However, little is known for certain about the significance of shifts in funding sources for instruction in mathematics and science. For education in general, four different views have been offered in the literature. The first regards federal and state initiatives as having little more than symbolic impact: the key decisions relating to the organization and morale of individual schools remain in the hands of local principals and superintendents who are selected by local school boards. The National Defense Education Act of 1958, for example, was the first major federal effort to raise the quality of public education. Yet it had had little direct impact on local school decision making, although it may have created a climate of opinion in which science and mathematics were thought to be important (Peterson, 1983; Sufrin, 1983). In the second view, federal and state initiatives present opportunities for upgrading and enhancing instruction for populations or subjects of special concern. Minimum standards can be set; states can attract higher-quality teachers through differentiated salary schedules; new, higher-quality curricula can be disseminated more easily in a more centralized system. At this time, quite clearly, much of the impetus for school reform is coming from governors, state legisla- tures, and state departments of education (Dougherty, 1983; Education Commission of the States, 1983). Local attentiveness to initiatives varies widely, however, and is dependent on local conditions (Elmore, 1983; Berman and McLaughlin, 1974-1975; Wimpelberg and Ginsberg, 1985) as well as on the vigor with which states implement policies and standards.
32 A third view (see, for example, Atkin, 1980) regards the greater involvement of federal and state governments with skepticism, if not outright suspicion. At best, compliance with federal and state requirements is seen as being accomplished through scrupulous attention to ritual, such as meeting special requirements for certifying mathematics teachers without further attention to what the mathematics teacher does once in the classroom (Meyer and Rowan, 19781. Worse, more regulations, guidelines, and controls are said to frustrate the creative teacher and impose operating procedures that may be entirely inappropriate in many local circumstances (Boyer, 1983; Sizer, 1984). In this view, many of the federal regula- tions of the 1960s governing the categorical programs for disadvantaged, handicapped, and other special groups of students were ineffectual, counterproductive, or had unfortunate secondary consequences. Still another view, perhaps a variant of the second, expects that current state and federal efforts for improving the quality of education will mainly be directed toward enhancing learning opportunities for the more able students, and that these efforts will be made at the expense of socially and educationally disadvantaged students. m e fear is expressed that the progress toward equality in educational opportunity made over the past three decades will be reversed. Surveys on the use of microcomputers in schools lend some substance to this concern (Center for Social Organization of Schools, 1983-1984; Lepper et al., 1984). The committee recommends a greater investment in research on the effects of the policy-making system on learning experiences in the classroom, particularly those related to the teaching and learning of higher-order skills, including (1) the effects of federal, state, and local district policies and procedures; (2) the under- standings that teachers and administrators have of goals proclaimed at the national and state levels; and (3) the decision-making processes of classroom teachers regarding the amount of time spent on and emphasis given to various aspects of the curriculum. Schools have often been described as both excessively resistant to change and excessively faddish in adopting change. Although some of the differences in description are undoubtedly due to differences in the prejudices toward specific changes and change in general on the part of observers, it is possible that both descriptions are partly correct. Change superimposed from above might be
33 adopted in outward form but then be turned into a marginal alteration compatible with established values, operating procedures, and investment of resources (Purkey and Smith, 1983; Zaltman et al., 1973). More general research on change in organizations suggests that both the mix of organizations and the character of individual organiza- tions change over time, but that those changes are not ordinarily attributable to the intention of organizational leaders (March and Simon, 1958; Cyert and March, 1963; Cohen and March, 1974). Rather, they reflect adaptations through differential birth and growth of different organizational forms (Greiner, 1972; Rimberley, 1980), incremental trial-by-trial learning from experience (Herriott et al., 1985), diffusion of ideas (Rogers, 1962; Rogers and Shoemaker, 1971), and serendipitous discovery of the organizational value of changes initiated for the local benefit of subgroups or individuals within the organization (March, 1981). Education organizations, specifically, tend to undergo changes because of the pressures exerted by interest groups and emerging societal issues rather than in accord with plans and initiatives of governing boards or adminis- trators (Or eeben, 1976; Cusick, 1983). School systems often respond to such pressures through changes in organizational form such as decentralizing the ad~inis- tration of large urban districts or the creation of new forms such as regional vocational centers. Sometimes changes that are initiated for the benefit of particular subgroups--Title I (now Chapter I) providing for compensa- tory education for disadvantaged children, special pro- grams for handicapped children, the creation of special- ized science and mathematics schools and of magnet schools--result in improved staffing patterns for other students as well and greater attention to curriculum. Schools also change through the diffusion of ideas, methods, and curricula, albeit through a slow process of adaptation (Berman and McLaughlin, 1974-1975; Kiesler and Turner, 1977). Evidence from recent studies of school processes and practices (Goodlad, 1984; Lightfoot, 1983) indicates that an important differentiating characteristic between schools is their receptivity to change, or their being a "renewing school" (Sarason, 19851. Although systems for planned change in education have been designed (Havelock, 1969) and occasionally tried out (Raizen, 1979), how change processes operate in different types of schools and school systems to Facilitate or hinder educational improvement is not well understood
34 (Mann, 1976). The accelerating introduction of computers into schools and their various uses in different settings (Center for Social Organization of Schools, 1983-1984) sharpen the need for better understanding of the change processes involved when innovations are adopted (Education Turnkey Systems, 1985). The committee recommends more focused research on the extent to which the conditions for specific changes exist in educational institutions, where the loci for change are and how they vary in different schools, and how cur- ricular and instructional changes are related to specific conditions. Because of the potential of computers and information technology, the committee recommends special attention to the processes of change involved in their introduction and use in schools. RESEARCH ON THE HOME AS A SETTING FOR EDUCATION A sizable body of research shows that the home setting influences educational outcomes, although little of the work specifically addresses outcomes in mathematics, science, and technology. Most of the research concerns four characteristics of home settings: the composition of the household, the socioeconomic status of the family, parental attitudes and behavior toward education, and resources in the home. Research on household size and mathematics achievement shows a rather consistent correlation between the number of persons in the household and achievement: larger household size is associated with lower achievement. For example, this finding appears in a study of fifth- and sixth-graders drawn from 700 schools in two geographic regions (Hanushek, 1972). An analysis based an s~xth- graders in a large, eastern city also confirms a cor- relation between family size and mathematics achievement (Michelson, 1970). Whether this result or one relating high scores in mathematics to schools in which most students live with both parents (Mayeske et al., 1972) stems from a causal link between household composition and performance is not established. m e attribute used most commonly in research on home and learning is socioeconomic status (SES), usually measured by parental education, income, and occupation. SES, based on these three indicators, bears a strong relationship to outcomes in mathematics and science. A study of school districts in Colorado found that the
35 greater the number of persons in the district who had completed high school, the higher the mathematics achieve- ment test scores for high school students (Bidwell and Kasarda, 1975). Likewise, the Hanushek tl972) analysis cited earlier shows a positive correlation for individual students between father's education and performance on mathematics tests. m e correlations, however, may be misleading. Rakow (1984) included parental education in a model of science achievement that used four other predictors : student ability, student motivation, quantity of science ~nstruc- tion, and quality of instruction (as measured by size of the budget for teaching science). Student ability emerged as the most significant predictor; parental education was not nearly as important in the model. Moreover, home environment was even less predictive for the sample of nonwhite students than for the white students. In a study of determinants of student achievement for seniors who participated in the High School and Beyond Study in 1982, family SES seemed to be less important than parental encouragement and support, and the effect of SES on achievement by black students was even smaller than for other groups (Rock et al., 1985). Gemmill et al. (1982) studied attitudes toward and performance in mathematics for Mexican-American students and a matched group of Anglo students and found that parental behavior made a difference for both groups. A review of research on gender and mathematics tFox, 1977) shows that support and encouragement from parents is crucial to participa- tion in mathematics, but that parents give less encour- agement to their daughters than to their sons. Parental involvement also appears to have a positive effect on incidental learning from television (Walling, 1976) and other informal learning situations outside school (Rock et al., 1985). Parental encouragement is a major explanatory factor cited in research on the high performance of Japanese students in science and mathematics. Troost (in press) found that Japanese parents have a high participation rate in schools and that parents and schools are con- sistent in placing high demands on students. Japanese students spend three to four times as many hours on homework as do U.S. students (Fetters et al., 1983; Walberg et al., no date), commonly attend after-school schools (jukus), and spend more time discussing school work with their parents than do U.S. students (Fetters et al., 1983; Stevenson, 1983).
36 Several analyses suggest that the presence of educa- tional resources in the home facilitates learning (e.g., Walberg et al., 1981; Rakow, 1984). But the resources matter only if they are used. The National Assessment of Educational Progress survey on science (Bueftle et al., 1983) found that females were less likely to participate in saience-related activities at home than were males. Females watched fewer science programs on television, read fewer books on science, and were less likely to work on science projects or hobbies. The research literature on the home in relationship to mathematics, science, and technology education is limited in several respects. First, it is underdeveloped theoretically. There is no overarching perspective that indicates the kinds of variables that should make a difference and the reasons they deserve attention. Consequently, the variables selected for analysis and the measures of these variables vary considerably among research studies. Second, most of the studies on achievement use tests of basic skills, not tests on reasoning. It is not obvious that the home influences that appear to affect the acquisition of basic skills have similar effects on other types of skills. Third, many of the studies fail to look simultaneously at both home and other influences. They tend to be more descrip- tive and conventional than causally sophisticated. Fourth, the research seldom examines how effects differ for various segments of students. To remedy these limitations, the committee recommends research on factors associated with the home that bear on mathematics, science, and technology education, including (1) identification of critical variables and development of a theoretical framework that relates them to different types of learning outcomes; (2) disaggregating effects for different segments of the student population, e.g., by age, ability, ethnic group, and type of school dis- trict; and (3) studies that distinguish factors associated with the home from those in the wider community (e.g., influences of peers, neighborhoods, mass media) but examine their interactions and joint effects on learning. RESEARCH ON Of - FC~SR=M SETTINGS In science, technology and, to a lesser extent, mathe- matics, educational experiences are increasingly available to individuals outside the traditional school settings
37 (Bryant and Anderson, 1983). Museums, science centers, newspapers, and hobby groups and other clubs all have potential for influencing large Hers of people. Some of this instruction is intentional: it happens on television, through such programs as 3-2-1 Contact, Wewton's Apple, Mr. Wizard, New Tech Times, Nova, Discover, Science & Technology Week, Voyage of the Mimi, , and Search for Solutions; in the community, through - public facilities such as science museums, planetaria, marine aquaria, and libraries; in print, through publications such as Ranger Rick, Odyssey, and Highlights; and through such associations as JETS (Junior Engineering Technical Society) and model rocket groups (Sneider et al., 1984). There is also some tentative evidence (Krugman and Hartley, 1970; Hawkins, 1973; Gaffney, 1980; Comstock and Tully, 1981) to suppose that some unintentional instruction occurs, again, on television, through such programs as Quincy, The Whiz Kids, and Otherworld on film, through motion pictures , such as Ice Man, Quest for Fire, The Swamp Thing, E.T., War Games, and Splash; and also in print through such comic books as Superman, Spiderman, and The Incredible Hulk. Some information has been accumulated on the goals, methods, effectiveness, and relation to school curricula of educational programs provided by the more popular of these media, like television and museum programs. With respect to goals, such programs as the Nova series (Ambrosino and Burns, 1977) and 3-2-1 Contact (Thomson, 1980) must constantly take into account the goals of television itself as well as their own educational goals Regarding methods, there are pedagogical strengths and weaknesses endemic to every informal educational method, including the use of museum exhibits (Danilov, 1982), television animation (Dusewicz, 1981), and computer simulation (Library of Congress, 1971; Stevens and Roberts, 1983; White 1984). Research on the effective- ness of news programs (Berry, 1983) suggests that redundant pictures and words enhance learning from television, while redundant printed information does not (Reese, 1984). Other research has addressed the efficacy of the timing of pictorials in mathematics learning (Brody and Legenze, 1980), of high-impact production features like action and music (Calvert and Watkins, 1979) and color (Chute, 1980), and of the participation of parents (Gaffney, 1981). An interesting question concerns the associations, possible and actual, between .
38 out-of-school education and in-school experiences (Passow, 1985). Some museums, for example, sponsor precollege science education programs for school classes (Goldman' 1970; Screven, 1984; Pittman-Gelles, 1985), and there are other ongoing efforts to integrate educational programming using mass media into the school curriculum (University of Iowa, 1978). Concerning the intentional educational programs, there is some evidence for the pedagogical success of such individual efforts as the children's television programs Sesame Street and the Electric comDanU (Lesser, 1975; Harvey et al., 1976; Pearl, 1982) and various museum exhibits. However, general understanding of intentional learning efforts is not sufficient to account for their effects--why some succeed and others fail--or predict their impact. Much more needs to be known before the potential of the nonschool media for providing quality learning time in informal settings can be adequately exploited. Therefore, the committee recommends research on the effects of various nonschool instructors on children's knowledge and perceptions of mathematics, science, and technology, including both the effects of intentionally educational programs provided outside school and uninten- tional learning or mislearning acquired through science fiction and other entertainment programming through the mass media, especially television, film, and print. We also recommend research to determine how the effects of instruction that children receive in the school are influenced by the informal instruction they receive in the larger world. Developing an understanding of this relationship might eventually lead to making school instruction more effective by taking account of children's learning from intended and unintended out-of-school instruction.
