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.
Research Needs in Science and Mathematics Education Improving the quality of teaching and leaving in mathematics, science, and technology is a problem with many dimensions. In the last 20 years progress has been made in developing a basis for better education in science, mathematics, and technology. Experimentation with curriculum content and instructional strategies has yielded valuable insights. Cognitive studies are pro~rid- ing more understanding of the nature of learning and effec~cive instruction. Organizational studies have identified barriers and opportunities in implementing change in schools. Advances in the basic disciplines of the sciences and mathematics and in information engineer ing provide new resources for curricula and instruction. In this chapter the needs of otathematies and science education are considered, together with a set of research goals to meet those needs. The discussion is organized around four areas: the mathematics and science curric- ulum, the knowledge and skills- of teachers, settings for learning, and change in schools. Though some of the research goals pertinent to these areas are well served through work within specific disciplines, others require collaboration among several disciplines and between re- search and practice. Examples of needed interdisciplin- ary research are outlined, drawn in part from the com- mittee's preceding report (Committee, 1985), to illus- trate their potential contribution to progress in science and mathematics education. THE MATHEMATICS AND SCIENCE CURRICULUM Current curricula and instructional practices have evolved through a slow process of accretion of goals, 13 -
14 educational experience, institutional mechanisms, and organizational behaviors. For example, biology as a lOth-grade course was institutionalized in the 1920s (Moyer and Mayer, 1985~. Schools consider changing long-established practices only if al~cernati~res can be demonstrated to be superior. Patience sad persistence is reseeded for the development of such alternatives 9 for demonstrating their effectiveness through exemplary progrPrnn in a variety of school settings, and for the process of adoption asked adaptation that enables their installation in enough appropriate sites to mak@ a difference in education The commit~cee identified three curriculum- related areas that are likely to profit from interdisciplinary research and development. (1) improving he teaching of reasoning and understanding and the context of instruc- tion; (2) introducing new content into science and me the matics curricula; and (3) developing linkages between curriculum and testing. Improving Instruction Teaching Reasoning and Understanding o A critical problem with the o~athematice curriculum in grades K-l2 asked the science curricula in grades 7-12 is too much emphasis on facts and too little emphasis on basic concepts and method of mathes~a~cical and scientific reasoning O As for the science eurricul~ in elementary school, it hardly exists, and where it does9 it is more often a reading-progr~ about science or an eclectic selection of ~scier~ce projects rather than an organized science program. The problem of too much rote learning and too little teaching for understanding also exists at the level of introductory college courses (Arons, 1981, 1983) - ~ the only science courses most teachers take to prepare for leaching O And teachers teach as they are tauthe. They also use the textbook as their central instructional tool (Stake and Easley9 1978~. Analysis of current science textbooks has documented that the learning of special or technical vocabulary, ices, rote memorization, is a central feature of these ~cox~cs (Yager, 1983) . It should not be surprising ~cha~c recent studies on the overall outcomes of schooling show gains in elementary knowledge and skills by younger students--the ~basics. the schools
15 have been stressing--but that higher-level processes are being acquired less well Champagne and Klopfer, 1977; National As sesament of Educational Progress, 1983a). Not only do schools emphasize the learning of facts, but the current methods of presenting science and mathematics appear to alienate female children and minority children of both sexes, who often decide in elementary school and again at the Junior high level that these subjects are not for them. Although research to date has not fully explained how young children develop scientific reasoning and concep- tual knowledge, powerful levers are available for the improvement of curriculum and instruction. One is the advance of cognitive and instructional sciences, areas that were ~ ust emerging 25 years ago . The knowledge base they have created permits the development of instruc- tional materials and systems that can facilitate the learning of concepts and skills now largely absent in mathematics and science education. Also, current techno- logical developments are stimulating application of new understandings about cognitive learning to mathematics and science: see, for example, the discussion in the committee's prev~iow report of the work by Larkin and Reif (1976~; Anderson (1981~; Riley et al. (1983~; Schoenfeld (1979~; Champagne et al. (1980~; Gleotent (1982~; and McCloskay (1983~. Such research has produced better understanding of the components that make good instruction effective, which include: (a) models of correct performance (e.g., of physics or mathematics problem-solving), (b) models of uninstructed performance (e. g., "preconceptional of scientific phenomena that interact with theories being taught), and (c) models of effective instruction (e.gO, principles of design for effective instruction). It appears that problem-solving, comprehension, and effective reasoning are based on subJect-specific knowl- edge. l~erefore, it seems best to teach reasoning skills i n the context of specific sub] acts that students are learning. This finding implies that research on reason- ing needs to involve experts in a particular discipline as well as cognitive scientists and experienced teach- ers. As an example, sus~cained collaborsti~re work involving physicists and cognitive scientists has pro- duced an effective means of teaching beginning students the difficult problem-solving techniques required for exploiting Newton's laws (Reif and Heller, 1982~.
16 In summary, to solve the problem of presenting ancient tific and o - kinematics subject matter in ways that enhance understanding and knowledge about natural phenomena and build competence in quantitative Winking requires the involvement of cognitive scientists, scholars in the scientific and mathematical disciplines, teachers, designers and developers, ant computer experts. The Context of Instruction Advanced scholarship in a subject is based on theories and concepts that maice a domain accessible to sub; ect°matter experts . However, a theory for the expert may not be good as a pedagogical theory for ache novice. Recent work in cognitive psychology has described how acquired knowledge is organized and represented and how cognitive models can facilitate reasoning and thinking as students use and test these models to solve problems and revise what they already Stow (Estes et al., 1982; Rumelhart and Norman, 19819 Resnick, 19-87), but such research has had little influence on the rigidly hierar chical conception of science and o~athematics that under- girds most classroom instruction. However,, effective teachers use their experience on how students learn to shape the subject matter they present (Brophy and Evertson, 1976; Collins and Stevens9 1982~. This craft knowledge provides an important source for developing instructional theory for teaching science and mathematics deco students at different levels of competence and education. "other source of knowledge is research on how young children learn (Ts~comina, 197S; Donaldson, 1978 ,, DeLoache arid Brown, 1979; Carey, 1985~. This research suggests that a fundamental way of changing how much time a child takes for a par~cicular task is to change the context of the task by a~ibedd-ing it in so~ larger activity in~rolv- ing familiar situations. Knowledge is lacking on how the insights gained from the work with young children can be applied at hither levels of the curriculum in ache areas of science 9 mathes~atics, and technology. However, there is evidence that science curricula combining activity- based instruction win appropriate text o~aterials are more effective than traditional curricula in teaching hi~er-order skills (Shymanaky et al., 1983; Holdzkom ant Lutz, 1984) 0
17 Because of the great importance of curricular orienta- tion and context to the learning of mathematics, science, and technology, ache committee urges special emphasis on this research area. Priorities include research on how important tasks can be embedded in contexts that reduce time needed for learning; under what circumstances and in what ways hands-on experience and laboratory experiments as well as models based on systematic laws can be used to enhance learning; and what makes theory~oriented instruc- ~cion work. The kinds of collaboration needed to carry out this sort of work, since it is more applied, are even broader than those required for developing a better under standing of reasoning processes and will require--in addition to subject m~t~cer experts, cognitive scien~cists, computer experts, and experienced teachers--the involve- ment of sociologists, psychologists, and anthropologists Integrating New Content Areas Educating for Scientific Literacy lathe goal of elementary science education should be to prepare children to understand science and mathematics, to teach them a few representative concepts, to acquaint ached with abstraction, and to introduce them to scien- tific reasoning and the nave of scientific evidence. The temptation to include tithe most important basic concepts" in each scientific discipline probably needs to be resisted, became the list will soon be very long O In the committee ' ~ view, it is more important for students to learn a few concepts well, with sufficient underatand- ing to teach the concept to another person, than to have superficial exposure to many concepts. True understand- ing will make it much easier for students to master addi tional concepts later and will make science enjoyable rather than intimidating. The secondary school curriculum needs to address the goal of improving scientific literacy for all students National Science Foundation, 1983~. School leavers, no matter when they stop their education, need to find employment in a labor market increasingly driven by tech- nological innovation. For everyone, an understanding of the scientific method, of the ways in which inferences are drawn from evidence, of-the nature of controlled experiments and scientific proof will be essential for making individual and public choices in coming decades. - -
18 A general conceptual ~terstasading will be more useful than a small amount of specific, technical knowledge that soon will be out of date. People will also need a con- tlnulng capability for learning and comprehension of science and mathematics throughout life. Consequently, curriculum and instructional alternatives of be con- cerned with methods of learning chic prepare people deco take advantage of the lifelong availability of info~a- tion technology. Curricula depend upon and are built around ed.uca- tional materials. Texthoo" are a central factor in de~cermining what is learned on what the schedule (Stake arid Easley, 1978~. The content of textbook is influ- enced by their authors' sense of appropriate learning goals, the publishers' perception of ache demands of the education market, and the priorities Id procedures for textbook approval arid selection of states and local school districts. ~. Neither scientists from the dis- ciplines taught in school nor cognitive acien~cists have been much involved in the process in Scenic years. In contrast. the attempts to reform mathematics and science curricula in the late 1950s and 1960s (~e Chapter 3) did Isolde scientists and astheo~aticians, and such attention was paid deco the balance between emphasis on facts and emphasis OR concepts and learning how to learn. Shout tiae then new curricula often were difficult for studen~cs asked misma~cched to teachers' coaDetencies. ~ar~cicularly in _, · . the elementary grates, they led to so~ improvement in the learning of inferential and critical thinking skills le%~-tr~ -- -1 ~ qua\ `~, A_ An., ~___,. However, those curricula have been replaced by more traditional materials, and the effects have not beast maintair~ed. The other component of teaching science, hantls°on experience in ele~cary school and laboratory experi- ments in the upper grades9 is in danger of disappearing from many schools. At the same time, the potential of computers to strengthen science and mathematics education resins largely untapped. The educational computer soft ware that is commercially available shows little of the creativity of pilot programs developed by scientists working with cognitive researchers. ~ concerted interdisciplinary research effort is needed to improve the science and mathematics curricula and their use. First, scientists and learning special- ists mat agree on the appropriate content for de~relop- ment of higher~order skills . Second, new sub] ect-ma~cter content within various fields ant at various grade levels
19 must be explored as potentially more appropriate than traditional topics. As a specific instance, systematic attention must be given to changes in the knowledge struc- tures and the processes of the sciences and mathematics as a result of readily available computation and the implications of these changes for the school curriculum. Just as the advent of calculators made traditional drill in using logarithm tables superfluous, the agent of more powerful computers has implications for all the science and mathematics courses taught in school. Third, the substantive content in current textbooks and software must be reformulated to support the learning of reason- ing, thinking, and problem-sol~ring skills as well as lower-order recall and memorization tasks. And fourth, curricula design out take into acco~c the role of hand~-on experience and the laboratory in the learning of science. Each of these efforts involves the cooperation of several disciplines and specialties. Technology Education In addition to these research issues on curricula content for science and mathematics, there are specific questions surrounding two aspects of technology educa- tion: the subject matter appropriate to understanding technological system and the teaching of computer science. Unlike the more traditional donating of science and mathematics, technology and computer science do not have well-established curricular With respect deco under standing technological systems (as contrasted to the basic science that gives rise to them), the traditional school curricula is essentially a blank. With respect to computer science, many schools are now introducing "computer literacy. courses. Some districts are even requiring such a course for high school graduation. Often, such courses focus on teaching programing in a particular computer language. In other instances, computer literacy courses deal with the capa- bilities and functioning of computers, either with or without hands-on experience, and only include topics on the effects of computers on the workplace and society. Knowledge is lacking about the age and grade levels at which computers and programing should be introduced and about the effects of alternative curricula in computer literacy.
~ ~ d~ 20 Interdisciplinary research is ne:eded~co provide characterizations of the cognitive skills and knowledge needed for understanding of and successful performance in technological systems and to develop usable school curricula in this area and in computer literacy. Vocational Education A special case in the school eurricul~ that involves both scicnti£ic literacy and technology education is ache rela~cionshlp of vocational instruction to the com~en- tional oasthe~aatics and science experience in elementary and secondary schools. As work in industry entailing computers and robots becomes extensl~re, the boundaries between vocational and academic learning will become ever more blurred. Employers are increasingly concerned with instruction in technical utters that involves some com- bination of apprenticeship, vocational training, and precollege academic learning. I of the unresolved questions is how asked where these learning activities should be provided°-in schools, in industry, or some Combination of industry and schools. For examples is a 1984 survey the Industrial Research Institute identified some 161 cooperative ventures between its member induce trial research laboratories arid high schools and colleges chat emphasize mathema~cics, science, and technology train- ing in practical set~cinge. As of 1983, nearly 2 out of 3 of the nations mayor companies were involved in coopers Live programs or provided internships for students from local schools (Lund and McGuire9 1984~. In~cerdisci- plinary research projects could utilize the existing collaborative education and training ventures to assess their potential for instruction in science and mathe- matics and in Marion vocations. Curricula and Testing Most present classroom methods of deducing what stu- dents know emphasize ache recall of facts--as does most leaching O If tests are not to trivialize instruction even further, new approaches to assessing student achieve- ment oust be developed thee aim at conceptual understand- ing, the ability to reason and than with scientific or mathematical subject otter, and competence in ache key processes that characterize science or mathematics (Frederiksen, 1984; Romberg, 1986~.
21 Work on two different kinds of Scents is needed. Assessments of student achievement that reflect the expectations of the life- long social and economic roles of an individual need to be developed. And more limited tests that bear on the curriculum of a particular grade or course need improvement. Scares or school districts should be assisted in developing tests that measure student reasoning and problem-sol~ring skills. If more adequate tests were available ~ school district evalus- ~cions and certifications could be tied to overall test scores as well as to patterns of variance in achievement across social class, race, language, gender, and economic characteristics of pupils. Thus, test outcomes would be directly related to what a school is expected to achieve, and test results would have meaningful impact on the schools . There is a new and promising line of research that links traditional psychometrics to the growing under- standing of reasoning skills (Hunt et al ., 1973 ; Glaser, 1981; Sternberg, 1977, 1984) . Cheap, powerful computers provide possibilities for more effective interactive testing--making tests more accurate and less time- cons~ning for students as well as less labor-ir~tensive for those administering them (Weiss, 1983--but further research is required to explore those possibilities. Tests play a role in the lea' sing process itself. They tell students what in the curriculum is important and shape the -teaching and learning process (Frederiksen, 1984; Romberg, 1986) . If, for example, testing is con- fined to recalling facts, students will concentrate on memorizing those facts, ignoring the more sophisticated levels of understanding and reasoning to which teachers and text materials may be rendering lip service. Teach- ers and school administrators also use tests as a guide to curriculum emphasis, especially when student perfor- mance on given tests is wed as ~ measure of teacher and school performance. Research needs in testing include the development of practical tests that reliably assess reasoning ability, possibly using interactive teasing made possible by microcomputers; improving the testing of mathematics and science achievement to reflect important instructional goals and objectives; and techniques for educating teachers to become better writers of test questions, particularly of questions that test for higher-order intellectual skills and levels of learning. Cognitive scientists, psychologists, mathematicians, physicists,
22 chemists,, and biologists onset all be involved in such efforts . lisle KNOWIEDGE AND SKITt5 OF TEACHERS Teachers must understand the material they are pre- ses~ting in order to teach it effectivelyO the natural curiosity of children leads them to ask basic questions. This tendency will be encouraged by teachers who under- stand basic concepts theo'selves arid who can we this understanding to lead students to a comprehension of fundamental ideas of physics, chemistry, biology, and mathematics. Subject-matter understanding is essential because it allows a teacher to use whatever examples are at hand to show students how evidence is used to build knowledge. Furthermore, a teacher Aunt understand the nature of scientific reasoning and know the difference between evidence and inference. Teachers must also under- stand how students acquire knowledge and understanding in science and mathematics. New curricula and teaching materials will be useless without teacher understanding. The committee has identified three areas that would benefit greatly by interdisciplinary research on science and mathematics leaching I (1) the content of teacher education; (~) elementary teachers' knowledge and under- standing; and (3) assessment of initiatives related to the teaching profession. Content of Teacher Education SubJect-matter courses taken by prospective teachers visually the standard courses offered by science depart meets--often cover too much content at too rapid a pace ant seldom pay explicit heed to developing reasoning capacities (Axons, 19831. Hence, prospective and practic- ing teachers often lack a genuine understanding of con- cepts and lines of reasoning that characterize the sub- ]@Ct(~) they are teaching and9 having missed effective training themselves' are unable to cultivate and enhance the basic reasoning capacity of their students. Thus, additional teacher understanding of science cannot be achieved simply by sending teachers back to the uniters sities to take a few more science courses. Moreover, methods courses in education are not very helpful in elucidating how students learn specific subject matter,
23 because these courses do not often include recent ad. vances in the cognitive sciences. Thus, courses in the universities, whether located in schools of education or in the disciplinary departments, are inadequate to train teachers in science and mathematics education. Teacher competence involves adequate cognitive mastery of the subject matter to be taught and, in the case of science, proficiency in handling experimental materials that can lead students to form new concepts from observation and evidence. Even the best curricula will be ineffective unless teachers are trained to deal with various modes of abstract reasoning, for example: the logic of arithmetic involved in ratios and division, the logic of control of variables, dealing with proposi- tional statements, recognizing gaps in available informa- tion, making inferences and predictions from mental models, doing hypothetico-deductive reasoning, and the like (Arons, 1981~. In fact, the processes and problems involved in educating teachers to acquire these capaci- ties are not very different from those i m olved for younger learners. New undergraduate and in-service courses that empha- size conceptual learning and comprehension in science and mathematics need to be designed. What is needed is not only new course development, but also new educational research oriented toward specific subject matter. Cogni- ti~re scientists and sub] ect-~tter scientists oust col- laborate in studying how scientific knowledge is acquired by older students, i.e., prospec~ci~re or practicing leachers O Communication between these groups for the purpose of teacher education has been minimal up until now9 but it must be increased. And both sides will have deco master knowledge of the ocher discipline if they are to collaborate fruitfully in research on scientific learning and in developing improved courses useful for teacher education. Elementary Teachers' Knowledge and Understanding Conventional views of public education presume that Ache teachers of elementary school science and mathematics understand the subjects sufficiently well to transfer comprehension to the great diversity of students they face in the classroom. Evidence from teacher examina- tions (Vobe~da, 1985), and research on teacher perfor- mance and student achievement ~ Schalock, 1979; Fisher et
24 al., 1980; Lain hardt and Greeno, 1984; Limpest, 1986) does not bear out this presumption. Assessments of teacher performance should take into account teachers' perceptions of how and what students are learning and teachers' ability to recognize the failure to learn. Thuds teacher education would concentrate on the relax tively new area of teachers' interpretation of students' thought processes and ability to use new skills rather than traditional estimates of factual knowledge. To develop such education, further investigations by subJect°matter experts in mathematics and science and behavioral and cognitive scientists are needed con whether and how teachers effect student learning. One approach to i-mpro~ring mathematics and science learning in elementary school in the use of specialist teachers. Specialized teaching, starting in grades 4 or S. characterizes the education system of many other industrialized nations, particularly those in which students outperform U.S. students. A few such specialist teachers work in selected school districts in several states, but at present the practice is uncommon in U.S. elementary schools. If the use of specialist teachers were to spread in grades 4-6, what training would be appropriate for them? How could school systems w e specialist teachers without the investment of additional E6SOUrC68 that Day It be available? What should be the role of the specialist leacher O student contact? works ing with regular grade teachers? both? Cooperative efforts involving several specialties from the natural and social sciences as well as experienced teachers and administrators are needed to develop better approaches to the education of those who teach mathematics and science in the elementary grades, whether they are specialist or regular teachers. Initiatives to Improve the Teaching Profession Many of the current initiatives for improving educa- tion are aimed at making teaching a more rigorous profess sign. Thus, 32 states have changed teacher certification requirements, 28 states have changed teacher education curricula, and 20 states have raised entrance require- ments for teacher education programs (Goertz et al., 1984~. Other policies that have been proposed are to require liberal arts mayors for all teachers and a 5-year rather than 4-year degree program (Scanner and Guenther,
25 1981; Boyer, 1983; National Commission for Excellence in Teacher Education, 1984; Holmes Group, 1986; Carnegie Fond on Education and the Economy, 1986), and to ensure teacher competence through a nationally recognized licensing examination (Shanker, 198S). Still other proposals are intended to make precollege education a more attractive profession in order to encourage entry into teaching of individuals of high ability and keep competent teacher and administrators from leaving the schools. The se proposals include salary increases and structural changes in compensation for teachers and the creation of more conducive working conditions in schools. Few of the policy changes that here been implemented are based on research evidence about the decisions of individuals deco enter teaching (Mu~-..ane, 198S ~ or the acquisition of knowledge and skills deemed necessary for science and mathematics teaching. Some of the new poli- cies may procure effective; others may have some us~desir- able consequences. For examples Seers and Wolfe (1977) found a statistically significant negative correlation between teachers' scores on the National Teachers Exam and their students' test score gains. Possibly more critical than initial certifications - the ob] eat of many of the policy changes--is recertification, since specific college learning, even if pertinent to teaching, fades unless used and renewed. A 3-year cycle of evaluation for recertification might be considered, during which teachers would be asked to produce their own evidence on their competence. The current experimentation with incentives, teacher education programs, and cretentialing sharpens the need to understand better (a) who gets access to various teacher preparation program, (b) the content of these programs, (c) the regulation of access to teaching posi- tions, and (d) the maintenance of teacher competence. These factors are poorly understood even for the current pool of teachers. Gaining a better understanding will require cooperative work by economists, sociologists, sub] act-matter specialists, teacher educators, and educational administrators. SETTINGS FOR LEARNING Uha~c factors influence students ' willingness or interest in learning science and Attics? To what extent do experiences in the elementary grades influence
26 choice of courses at the secondary level and later careers? Research suggests that the school emirorment, family, and co~uni~cy all influence students' a~cti~des and confidence in their own abilities to comprehend science and mathematics. Three area in this domain appear particularly in need of interdisciplinary research: (1) assessing Ache effects of school and classroom settings; (2) understand- ir~g parental effects on st:udent learning; and (3) examine ing educational models that cross institutional bounda- ries O School and Clasarooo' Settings Schools should be responsible for n~o~civating stu- dents. The success of some schools in improving the perfo~ne. or comment of students should be studied to learn how schools can persuade students that scien tific li~ceracy, like learning to read, is expected of allO Research points to a number of ques~cions that need deco be addressed (~e reviews by Chipman ant Thomas, 1984; Chipman et al.9 19859 Eccles°Parsons et all 1983; Ameri- can Association for the Advancement of Science, 1984)0 How does Ache school anviror~nt influence the perception that science and mathematics are difficult? Are teacher expectations and at~citudes or i~titu~cional expectations influential? Do the different reaction of different cultural and gender groups to science and Mathematics result at 1~t is part from the attitudes of coaches and school administrators toward them? How should schools and classroom be organized for effective learning of o~a~ches~a~cics and Science? Schools organize elassroo~ in various ways (Bidwell and Kasardla, 1975~. Group placement based on ability is common to reduce heterogeneity and match ins~cruction to the students' skills. Such placement has been shown to be Vocable over ti~--once in the low group, it is hard to get out--and to hare differential impact on high- and low-group students. Studies conalatently and robustly document that ability grouping has detrimental effects within classrooms on average and low-ability groups (Persell, 1977; Good and Marshall, 1984~. Inappropria~ce grouping may amplify relatively minor differences at the beginning of let grade into orator differences in later grades; it certainly affects students' perceptions about their ability to do mathematics and science (Hallinan and Sorensen, 1984) .
~1 Successful methods of classroom' instruction fre- quently involve breaking classes into smaller activity groups that combine theoretical understanding with hands-on familiarity. Unfortunately, these conditions have only been maintained in specialized program (Goodlad, 1984; Hawkins and Sheingold, 1983; Mall and Diaz, 1982; Peterson en al., 1984~. Even programs with demonstrated success, like the acti~rity-based elementary school science curricula of the 1960s, currently languish and are used only when individual teachers make heroic efforts or an especially well~educated parent group creates a demand. therefore, research is needed on how to make student activity groups successful, especially in multi ~ ethnic classrooms, over a range of mathematics and science tasks. In particular, the combined effects of curriculum, school organization, teacher training, and small group dyn~ice--especially the question of group heterogeneity--need to be better unde£stoodO That kind of research requires not only sub ~ eat -matter experts and cognitive scientists, but also sociologists, anthro- pologists, and social psychologists who have studied the environments of classrooms and whole schools. Another important area of research concerns the relation between those used for instruction and student learning. A recent study by Stevenson et al. (1986) showed that the mathematics achievement of American children compared with the achievement of Japanese and Chinese children consistently declined from Ist through Sth grades, and that differences in the amount of instruc- tional time and direct instruction by the teacher appear to be important factors. In the 5th grade, American children spend 64.5 percent of their classroom time involved in academic activities; Chinese children spend 91.5 percent and Japanese children spend 87.4 percent. American teachers spent proportionally less time during the school day on ache subject matter (21 percent) than did the Chinese teachers (58 percent) or the Japanese teachers (33 percent). The Chinese and Japanese children also spent a considerably larger portion of the school day on mathematics than did the American children, who averaged less than 20 percent of the school day on mathematics . School organizational factors that affect teachers' use of time, such as scoff support and professional autonomy, and factors that increase or decrease teacher incentives and motivation to improve their teaching must be better understood before interventions to increase teacher effort can be designed.
2-8 Parental Effects on Student Learning lathe values' attitudes, and beliefs of African parents are important elements in how much and how well American children learn. Parental attitudes appear to make a difference for both minority and other students and for males and females (Gemill en al. 9 1982) 0 A review of researcia ore gender and mathematics (Fox' 1977) shows that support and encouragement from parents is crucial deco participation in mathematics, but that parents give less encouragement to their daughters thence their Sony A survey on science dirts ire conjunction with the National Assessment of Educational Progress (BLEEP) (~uef~cle et al., 1983) found that Americar~ female students were 1~s likely to participance in science- related activities at home char were Ales. Females watches fewer science program on television, read fewer booles on science, and were less likely to work on science projects or hobbies Several analyses suggest that the presence of educational resources in the home facilitates learning (e.g.,, Wslberg en al.9 1981; Ralcow, 1984), but resources matter only if they are used. Research on student learning in different countries provides insight into the importance of parental influ- encesO American elemes~cary students spend far less time of homework than Japanese and Chinese students (Stevenson 6t alO 9 1986; Fetters et al. ~ 1983; Walberg et al., n. doJo For example, in the Sth grade, A`aerican mothers estimated that their children spent 46 Minutes a clay on homework; for the Chinese9 the estimate was 114 minutes and for the Japanese, 57 minutes. At the same the, howler 69 percent of African mothers believed the small ueount of homework of their children was use right, and the Chinese and Japar`~ac Lao the rs were not dissatisfied with larger amounts of homework assigned to their children (Stevenson en al., 1986~. Troost (1985) also found that Japanese parents have a high participa- tion rate in schools and that parents ant schools are consistent in placing 0th demar~ds on studen~ca. African students receive less assistance fro. parasite than s~cu- dents in china and Japan; a lower percentage have deal" in the homes and fewer receive mathematics or science workbooks from parents (Stevenson et al., 1986~. Mothers in different countries have different perceptions about the factors important in student achievement: Japanese mothers assigned the highest ranking to the child's effort; American mothers gave the
29 highest ranking to ability. The critical role of effort in achieving success, as perceived in Japan, no doubt contributes to the development of after- school schools (juku) and greater time spent by Japanese parents discussing school work with their children (Fetters et al., 1983; Stevenson et al., 1986~. Research on the home in relationship to mathematics, science, and technology education has provided some impor- tant ins ights, but it is limited in several respects . Variables selected for analysis and the measures of these variables vary considerably among research studies ~ and many fail to look simultaneously at both home and non- home influences. Also, most of ache studies relating parental and home influence to achievement use tests of basic skills; possibly, the result would be different if tests of reasoning were used instead. Moreover, the studies seldom examine how effects of home and parents differ for various groups of students. Cooperative efforts will be needed among researchers who have been working from the perspectives of their own specialities. A theoretical framework must be constructed that relates critical variables pertaining to parental and home influ- ence to different types of learning outcomes specifically ire science and mathematics; effects for different seg- ments of the student population need to be disaggregated by age, ability, ethnic group, and type of school dis- tr~ct; and studies should distinguish factors associated with the home from those in the wider comity (e.g., influences of peers, neighborhoods, mass media) but examine their interactions and Joint effects on the learning of mathematics and science. Educational Models That Cross Institutional Bour~dlaries Research reviewed in the paper on contextual factors in education prepared for the committee (Gore and Griffin, 1987 ~ makes it clear that coordinated attention should be given to educational activities that cross ache boundaries between school and out-of-school learning. For example, after-school learning activities could effectively increase active learning time for mathematics and science, using such settings as community centers, churches, libraries, and school facili~cies themselves. A variety of prototypes that suggest the range of possible activities and institutional arrangements already exists (Moll and Diaz, 1982; Woodson, 1982~. What does not
30 exist and needs interdisciplinary research to develop is an overall understanding of the potential and limitations of different configurations and sponsorships for such activities. Joint school-comaunity program and school-muse~ program. (Fantini and Sinclair, 1985) also hold promise for o~then.^tice and -science education. The American Association for the Advancement of Science (1984) summarized a large s~uoibar of exemplar educational program searing women and olinori~ciesO to report inclu-s much informal knowledge based ore practi- cai eXpeEionca with educa~cional program that are success ful with these populations O Many of the program exhibit a key structural property°°they create a ~ys~cem of educa- tios~ that is integrated bow vertically (from early edu- eation through later years) and horizontally (they coor- dinate and draw support frog' a range of depar~cments, inSti~tio~9 and bureaucracies) For exile, the Com~ity Educational Resource asked Research Center of the University of California San Diego, brings adults, col- l@ge students, and high -school students into a single activity setting after school, creating ver~cical integra- tion. Horizontal integration is achieved by involving multiple parties: university, school ~y~tem9 comity. The problem with such be dary°crossing systems, even where they are demonstrated suceesses9 is that they are difficult to fit into existing bureaucratic arrange- men~cs. As the AAAS report notes9 demonstrations of success based upon short-term funding of experience does not ensure continuation by the sponsoring institutions. T~ova~cive educational successes have had long~term failure built into Ocher ( - urinary Associa~cion for the Advancemen~c of Science, 1984, Stage et ale ~ 1985) o This history suggests a need for analysis ova how to create mixed institutional systems for mathematics and science education that can be sustained in the face of existing bureaucra~cic and social structures. CHANGE IN SCHOOLS Although structural and professional aspects of schools appear to work against significant educational change, some schools achieve sizable and lasting improve- ments in science and mathematics teaching. What are the structural features of those schools that encourage educational reform7 How did ache desires improvements take place?
31 General research on change in organizations suggests that both the mix of organizations and the character of individual organizations change over time, but that those changes are not ordinarily attributable to the fore- sithted-intention of organizational leaders (March and Simon, 1958; Cyert and March, 1963; Cohen and March, 1974~. Rather, they reflect birth and growth of dif- ferent organizational forms (Greiner, 1972; Kimberly and Miles, 19803, incremental trial-by- trial learning from experience (Herriott et al., 198S), and diffusion of ideas (Rogers, 1962~. Education organizations, for example, respond to the pressures exerted by interest groups and emerging societal issues rather than in accord with plans and initiatives of governing boards or admin- istrators (Dreeben, 1976; March, 1981; Cusick, 1983 ~ . Research is needed to pinpoint those aspects of schools or school systems that work for or against educational reform. As noted, few elementary schools retained the hands-on science experiences recommended in the curriculum reforms of the 1960s. Was this important and valuable facet of science teaching dropped because of funding problems, space limitations, class size, teacher attitudes, or administrative attitudes? Mat is happen- ing to laboratory instruction in high school science courses? Why? Why are computers being so little used in effective and imaginative ways for science and matiae- matics teaching? Models for Change in Schools To foster development of capabilities for change, NSE should consider supporting the design and pilot io~plemen- tation of modified kinds of schooling to serve as inter- pretive models for effective education. These models should be designed and implemented in existing school districts in order to demonstrate what kind of capability ts needed and how it can be created. In addition, basic research is needed to identify mayor organizational and social factors affecting change within educational insti- tutions. A research program should examine factors affecting the generation of new alternatives for o~athe- matics, science, and technology instruction and learning; the dissemination of information about such alternatives and experience with Echo; incentives within educational sys~ces~s for considering and adopting new alternatives; and capabilities for implementing effective new
~ - ~. 32 alternatives. loo avoid the predation that chang,. is always good, a research program should also consider both the conditions under which educational institutions fail to adopt effective stew programs or methods and the conditions under which they adopt programs or methods that should not have been adopted The problem of affective we of microcomputers in schools is a variation an the problem of fostering change ire organizations. Computers represent a means for prow sensing scierlce in an attractive more with broad appeal that can emphasize abstractions9 scientific reasoning, and experimentation. Computers can also be used as ~is~alligent tutored to make instruction more effective. Excessive practical use of computers would, however 9 require I different classroom organizations new teach- ing materials, easily available and usable software, and different teacher skills in managing differena~c social interaction. It is not enough to focus Just on training teachers,, or on developing software, or on changing learn- ing groups. All these factors ~t be related. Tnterdis- ciplinary research is needed to exploit findings on how schools can produce better learning through educational system that mith~c be quite different-card more effective --than familiar ones. Information and Evalua~cion of Alternatives Efforts to develop new curriculum material or instruc- tios~al systems are by themselves nose sufficient to have powerful posi~ci~re effects on the extent and quality of mathe~ties and sciences instruction. Another necessary but flOt sufficient condition for improvemen~c is that schools be informed about alternatives. Evaluation and dissemination of educational materials and practices should be supported in the areas of mathematics, science, and technology. Interdisciplinary collaboration drawing upon the knowledge and expertise of mathematicians, physicists, chemists' biologists, asked social scientists as well as mathematics and science educators will be needed to evalua~ce existing educational materials in science and mathematics from both a scientific and pedagogical perspec~cive. The reviews and evaluations should include curriculum materials and instructional al~eQrnatives developed with the support of NSF and other public and private funding agencies, textbooks and ancillary o~a~cerials prepared by publishing houses, and
33 materials produced by state departments of education or local public and priorate schools. Tests of student achievement should also be examined, particularly those used to make Judgments about statewide or nationwide performance in mathematics and science. In addition to evaluation of materials, exemplary and innovative educational practices should be evaluated, including new ways of using computers, use of specialist and resource teachers, and preser~rice and insenrice education. Evaluation result- should be given careful scrutiny by scientific and pedagogic experts and then widely disseminated to schools districts, teachers ~ prin- cipals9 and other concerned with educational improvement.
