Abstract: Like many others, we believe the observational, correlational, and experimental literature on research on teaching has been very fruitful. From observational and correlational findings, for example, we have determined that the presence of opening homework reviews in mathematics classes are associated with higher achievement. This idea was studied further by creating levels of review and using these levels as variables in experiments. Opening homework reviews were again confirmed as important. Re-

Abstract: The contention of this paper is that the claim of compatibility and the call for cooperation between quantitative and qualitative inquiry cannot be sustained. Moreover, these claims have the unfortunate effect of closing down an important conversation. To elaborate these points, this paper briefly reviews the transition from conflict to cooperation between the two perspectives and then notes how compatibility is based on a confusion over two different definitions of method. Finally, the discussion focuses on why this conversation, because it invokes issues crucial to our understanding of who we are and what we do as inquirers, must be reinvigorated.

Abstract: This paper proposes an alternative to both meta-analytic and traditional reviews. The method, “best-evidence synthesis,” combines the quantification of effect sizes and systematic study selection procedures of quantitative syntheses with the attention to individual studies and methodological and substantive issues typical of the best narrative reviews. Best-evidence syntheses focus on the “best evidence” in a field, the studies highest in internal and external validity, using well-specified and defended a priori inclusion criteria, and use effect size data as an adjunct to a full discussion of the literature being reviewed.

Abstract: Research on the educational facets of television has been marred by a number of blunders that stand in danger of being repeated in the new wave of educational research on computers. Despite the clear differences in purpose and scope between television and computers, some useful lessons can be drawn from earlier assumptions about learning from television. In this paper, it is argued that "computer researchers" should do the following: (a) avoid asking whether computers teach better than some putatively comparable medium; (b) utilize holistic as well as standard experimental research paradigms, particularly during the early phases of research; (c) realize that learners bring many assumptions, proclivities, and active learning strategies to any encounter with a new medium or technology; and (d) expect a range of usages and experiences and a variety of outcomes from any encounter between an individual and a computer. It is particularly important to carry out background research during the period before comput...

Abstract: only of meaning-bearing connectives, but simple articles as well. Return to check answers. If incorrect, repeat starred exercises noted. Me Tarzan, you Jane. Finally, producers of educational software make considerable use of nonverbal symbols in their interactive programs. This often militates against development of verbal abilities in two ways. First, users draw information from the visual representation in ways that do not always promote (or even involve) verbal refinement. Second, the visual and aural components are often irrelevant gimmicks, such as fireworks when a correct answer is given or cuckoo sounds to accompany static verbal messages like 'Tha t ' s wrong, try again.\" I hope, along with Olson, that future research demonstrates the existence of cognitive spillover into verbal realms when students interact in different ways with computers. But the explicitness involved in computer programming

Abstract: This article provides support for the assertion that significant and enduring changes in federal educational policy have occurred during the Reagan administration. The administration’s procedural and substantive policy preferences in education are defined, and evidence is provided of progress in achieving those preferences. Finally, a set of contextual conditions are identified that suggest the likelihood of continued progress on the Reagan federal educational policy agenda during the remainder of the Reagan term and beyond.

Abstract: is also notable for its discussion of problems in the retrieval of data from studies and the difficulties posed by different choices of unit of analysis. The material dealing with data analysis and interpretation begins with an argument for the importance of quantitative methods in analyzing the results of a large number of studies. A particularly important addition is the discussion of situa­ tions in which quantitative methods should not be used. There are readable introductions to many of the standard quanti­ tative techniques of research synthesis: vote counting, omni­ bus combined significance tests, RosenthaΓs fail-safe number, and the possibility of combining raw data. The standardized mean difference and correlation coefficient as indices of ef­ fect magnitude for combining study results are described, as well as methodology for combining effect size estimates and for testing variation among effect sizes. Each of the methods is clearly illustrated with an explicit example. The chapter is also notable for its criticism of the use of conven­ tional statistical methods (such as analysis of variance and regression analysis) with effect sizes or correlations. Public presentation of research reviews is an important part of meta-analysis, and advice and examples of how best to sum­ marize the evidence in research reviews are provided. This material takes on special relevance because conventions in reporting evidence in research reviews are not well estab­ lished. Cooper uses the basic division of primary research reports into four sections (introduction, methods, results, and discussion) as the basic structure for reporting research reviews. Several examples of tabular summaries are provided to illustrate the choices faced by the reviewer. In general, The Integrative Research Review provides an excellent introduction to methods for conducting rigorous re­ search reviews. It is generally clear and well-reasoned and cites references that can be the basis for further study. The principal limitations of the book stem from its brevity. Some topics must be treated in less than ideal detail in a book of 143 pages. The treatment of coding sheets for research reviews is well-reasoned but brief. Only one example of a coding sheet is given. Coding sheets are the instrument for collection of data on between-study variations. Adequate cod­ ing of study characteristics is therefore essential in detect­ ing patterns of variation in study outcomes. For example, some treatments such as open education, modern mathema­ tics instruction, or psychotherapy are by nature diffuse. The book contains neither a discussion of how to capture dimen­ sions of variation in diffuse treatments nor a discussion of the pervasive problem of incomplete reporting of details of study design and treatment implementation. In many re­ views, such incomplete reporting leads to a trade-off between what it would be desirable to code for each study and what it is possible to code. Some guidance on this matter would have been desirable. Finally, although the chapter on analysis and interpreta­ tion is up to date, it does not address many of the more sophis­ ticated and useful analyses for effect sizes, such as analogues to the analysis of variance and multiple regression. These methods are often needed in meta-analysis, and some cover­ age of them would have been desirable. This omission was undoubtedly planned, however, because the companion vol­ ume by Rosenthal goes into greater depth in its presenta­ tion of statistical methods. Under the circumstances, the deci­ sion to limit coverage of statistical methods is understandable. In spite of a few limitations, The Integrative Research Re­ view is an important contribution. Experienced researchers and reviewers will learn something by reading it. Its clear presentation of issues makes it particularly useful as a text for advanced undergraduates or beginning graduate students who are psychologically oriented. Perhaps most important is that it helps to demonstrate that thinking about rigorous procedures is as essential in every phase of research review­ ing as it is in every stage of original research.

Abstract: In recent years there has been a great deal of methodological debate among educational researchers, theoreticians, and practitioners concerning issues such as relativism raised by the so-called “new,” “Kuhnian” or “postpositivistic” philosophy of science. The intensity of this debate notwithstanding, the fundamental principles and their relations that comprise the postpositivistic view have not always been carefully spelled out. Some of the principles discussed will include (a) the problem of confirmation, (b) the underdetermination of theory by logic, (c) the underdetermination of theory by experience, (d) the Quine-Duhem thesis, (e) the theory-ladenness of experience, and (f) the incommensurability of theories. No attempt will be made to evaluate these principles. However, those who are prepared to accept all of these will be hard pressed to avoid the dangers of relativism. I will argue that these dangers, if they exist, may be lessened if not eliminated by practicing the pragmatic virtues of epistemolo...

Abstract: Commissioned reports calling for the reform of U.S. secondary schools have been followed by reports and recommendations for the reform of teacher education. Within the broader framework of the calls for school reform and the relatively brief history of institutionalized teacher education, this article critically examines the nature and implications of these calls. The article focuses on meanings given to reform, including the possibility of reform as ritual, and the technical rationality that seems to underlie the reform proposals. Consideration is then given to how teacher education reform efforts might address critical questions–particularly questions of purpose, substance, and value–that are largely neglected by the recent calls. (Note: Unless otherwise indicated, teacher education refers to the initial or preservice preparation of teachers.)

Abstract: This article presents a rationale for changing the current emphasis of precollege computer courses. It suggests that as an interim approach, to be used before computers are integrated effectively across the curriculum, computer courses deemphasize BASIC programming skills and instead focus on teaching applications software skills. The research evidence regarding the quality of computer literacy courses is reviewed, and the parallel cognitive and affective consequences of programming and applications software are discussed. The primary source of evidence regarding student interest in specific undergraduate majors comes from reports on college-bound seniors prepared for the College Board (College Board, 1983b, 1984, 1985). More than 800,000 high school seniors annually complete information regarding their intended college majors, choosing from a list of over 115 fields or specializations. Computer science was first listed as a possible major choice in 1974, with subcategories of systems analysis and data processing added in 1975. Between 1973 and 1983, interest in computer science rose dramatically. Less than 1% of high school seniors chose computer science as their intended major in 1974 (College Board, 1974), w ereas in 1983, more than 10% of high school seniors chose it (College Board, 1983b). In the next two years, however, interest in computer science took a sudden downward turn, with the level of interest in 1985 approximating that of 1981 (see Figure 1). This trend represents a decline of 27%, with women's interest decreasing 35% and men's interest 22%. No other intended major suffered such dramatic shifts during the same period of time. Although this rise and fall may simply reflect the fortunes of the computer industry in general, another explanation for this decline in interest also may be tenable. Students in the 1970s enrolled in untried and novel computer curricula. By 1985, due to their increased exposure to computers and computer literacy courses at elementary and secondary school levels, students had more information with which to evaluate computer science (Becker, 1983a, 1983c; Lockheed, 1985a). The increase in students taking precollege computer courses began around 1979. In 1980, 15% of elementary and 50% of secondary schools provided students access to microcomputers in their instructional programs (Goor, 1982); by 1985, 82% of elementary and 93% of secondary schools provided students access to computers (Becker, 1985). The most frequent use of microcomputers was for national, state, and locally mandated computer literacy courses (Boyer, 1983; College Board, 1983a; National Science Board Commission on Precollege Education in Mathematics, Science and Technology, 1983). In general, the mandates provide for required, not elective, computer course participation. It is the contention of the authors that the generally poor quality and restricted curriculum of these courses, which in 1985 were required for many high school seniors, accounts at least in part for the declining interest in computer science at the college level. Inadequate Computer Literacy Courses Early computer literacy courses, which stressed learning to program in BASIC, were generally of poor quality in terms of teacher training, course curSupport for the preparation of this paper was provided by Educational Testing Service. Marlaine E. Lockheed is a Sociologist in the Education and Training Department, The World Bank, 1818 H Street, NW, Washington, DC 20433. Her specializations are the sociology of education, education technology, and gender in education. Ellen B. Mandinach is Associate Research Scientist, Educational Testing Service, Princeton, NJ 08541. Her special zations are individual differences and educational uses of microcomputers.

Abstract: As economies of scale increase, school districts, by means of policy strategy, trade class size against salary incentive for teachers. This idea is developed and related to district-level achievement through an explanatory, structural equation model. The model is tested on data on student achievement in 102 Colorado school districts. The results show that the model is moderately successful for explaining reading achievement in grades 5, 6, 11, and 12. However, factors not yet included in the model are believed to have strong effects.

Abstract: This article presents a rationale for changing the current emphasis of precollege computer courses. It suggests that as an interim approach, to be used before computers are integrated effectively across the curriculum, computer courses deemphasize BASIC programming skills and instead focus on teaching applications software skills. The research evidence regarding the quality of computer literacy courses is reviewed, and the parallel cognitive and affective consequences of programming and applications software are discussed.

Abstract: 1. Introduction 2. Organizing a Reviewing Strategy 3. Quantitative Procedures 4. Numbers and Narrative: The Division of Labor 5. What We Have Learned 6. A Checklist for Evaluating Reviews Reference Index

Abstract: at the 1986 AERA annual meeting. Those who heard Futrell speak, and those who will now read her words, will recognize an important call for collaboration and mutual support between teachers and educational researchers. This call comes at a time when teachers, through their professional organizations, are striving for recognition qua professionals. As students of the professions have noted (see especially Lortie's 1977 landmark study of the status and functioning of teachers, School Teacher: A Sociological Study, published by the University of Chicago Press), professions depend critically on technical knowledge. Members of professions have and use, and are recognized by society as having and using, knowledge and technique special to their field. This knowledge sets members of the profession apart from occasional dabblers or amateurs at their art; it also distinguishes expert from novice members of the profession since knowledge takes time and practice to acquire. Futrell points out that as teachers seek professional recognition for their work, they are looking to research on education to provide the base for their technical knowledge. Research-on learning and instruction, on the social organization and functioning of schools, and on instruments for monitoring and managing the process of education-is what teachers need for building the tools of their technical expertise. Teachers and researchers need each other. Without users, most research would be empty of import; without increasing knowledge on which to build their expertise, teachers will find their claims to professional status unheeded. That is why this appeal for an alliance between teachers and researchers comes at this time, and why it is so important to heed the appeal. LAUREN B. RESNICK, President, AERA

Abstract: "mainstream." Proposal readers were not frequently seized by the originality of a research strategy or a fresh conceptualization of an old problem. There were some of these, to be sure, but not a lot. Possibly there were more in the early stages of the competition. Perhaps the peer review process itself, and the government rating systems behind it, tended to screen out less conventional approaches. But even if not exhilarating, the outcome was satisfactory. The centers we ended up with are all promising; each has a serious mission, able individuals, and a worthy plan, some worthier than others at this early stage. I feel good about the process, hopeful about the winners, and determined to do all we can to help them succeed-a judging process, I might add, in which we will use various peer review mechanisms. The situation with the regional labs is a bit more problematic. The new contracts describe an ambitious agenda of work that, if satisfactorily carried out, will surely benefit American education. So far, so good. But virtually all the "new" labs were also "old" labs or close facsimiles thereof. And of the old labs, while it is both conventional and true to remark that their work was "uneven," I am not aware of any evidence that American education was palpably improved in consequence of their activities as a group over the years. I say this with sadness, not satisfaction, and in the fervent hope that the future will yield a different record. We on the OERI side of this contractual relationship will do our level best to bring this about. But the record remains to be established. Yet the principles of peer review remain sound, and we shall apply them throughout OERI, even to the point of requiring outside participation in the review of proposals (such as contracts let by our Center for Statistics) that heretofore were appraised only by Department staff members. But we apply them in the full awareness that peer review is a help for decisionmakers, not a substitute for them, and that the ways a peer review process is constituted and managed matter greatly, especially in this field of education where a linear extension of past thinking and familiar practices into the future is simply not good enough.

Abstract: J ohn Leverett, who was nominated to be president of Harvard College in 1707, was the grandson of Governor John Leverett of Massachusetts. Leverett \"inherited wealth and readily took a place among the prominent men of affairs in Massachusetts\" (Hofstadter, 1961, pp. 106-107). Before his nomination at age 45, he had already served as a representative in the General Court, a justice of the peace, a Supreme Court judge, Speaker of the House, and a member of the provincial Council. His connections with his old friend Governor Dudley served him well, since he had been sent on \"important missions\" for the Province. What Hofstadter assumed, and what his readers assume, is that wealth and influence may be passed from one generation to another. However, I challenge the generality of that assumption. Leverett's sister, if he had one, could also have claimed Governor John Leverett as her grandfather, but she could not have assumed that his influence and wealth would lever her into prominence as a judge of the Supreme Court of Massachusetts or as a member of the provincial Council. There is no way that they could have promoted her presidency of Harvard

Abstract: That body of writing cannot be ignored, even if courses introduce readings that counterbalance its views by making women the Subject and men the Other. Hence, we have to undertake a continuing process of examination, of discussion about the paradigm that has traditionally been accepted. We have to reconceptualize the history of education in order to make both men and women, together, the Subject. Our task is to write new histories using that as our paradigm.

Abstract: The new scholarship on women and education will change the ways in which we think about educational research. The authors of the articles in this special issue of the Educational Researcher consider both conceptual and methodological issues. Jane Roland Martin argues that an education of the mind is an inadequate education if it ignores generative love. While her paper is anchored in the philosophy of education, she challenges the dualism of vocational versus liberal educa­ tion. Joan Burstyn discusses the history of education, but focuses particularly on the ways in which our current conception of this field is organized around male experience. Consideration of women's lives as \"subject\" rather than as \"other,\" she says, will change how we interpret educational events. In her report of efforts to create a history of women in educa­ tional research, Charol Shakeshaft poses methodological questions raised by the new scholarship on women. She suggests that traditional sources, questions, and approaches may be inadequate for characterizing women's contributions. Jacquelynne Eccles approaches the issue of gender and achievement from a psychologically framed perspective as she investigates the relationship between occupational preferences and gender-role stereotyping. Susan Klein and Karen Bogart recom­ mend sex-equity strategies for all levels of the educational system. Janice Scheuneman examines the quantitative areas of statistics and research methodology to probe the ways in which the new scholarship on women challenges certain methodological tenets. We have chosen articles that reflect a variety of perspectives, but the authors share an understand­ ing that the structure of educational disciplines as we know them, and the nature of methodology as we have come to understand it, will change as generalizations about human behavior are generated from women's as well as men's exper­ iences. These articles will hopefully stimulate dialogue around these important issues.

Abstract: only of meaning-bearing connectives, but simple articles as well. Return to check answers. If incorrect, repeat starred exercises noted. Me Tarzan, you Jane. Finally, producers of educational software make considerable use of nonverbal symbols in their interactive programs. This often militates against development of verbal abilities in two ways. First, users draw information from the visual representation in ways that do not always promote (or even involve) verbal refinement. Second, the visual and aural components are often irrelevant gimmicks, such as fireworks when a correct answer is given or cuckoo sounds to accompany static verbal messages like 'Tha t ' s wrong, try again.\" I hope, along with Olson, that future research demonstrates the existence of cognitive spillover into verbal realms when students interact in different ways with computers. But the explicitness involved in computer programming