107 Book Reviews/Comptes Rendus conclusion, imprégnée de réalisme, est à lire et à relire. Surtout, si on se souvient qu'elle est rédigée sur un campus qui fut plus que bien d'autres porté par le messianisme syndical. Pierre Savard Département d'histoire Université d'Ottawa Louis Cohen and Lawrence Manion. Research Methods in Education - Second Edition Croom Helm, London, 1985. It may be that educational research is an area of inquiry in trouble. Indeed, it may well be that it never did amount to much. Cohen and Manion seem to be of the view that the problems of educational research are of long standing, and that the progress of education itself has been "fitful and uneven". This unsatisfactory progress, in the main, is attributable to "too great a dependence on ... experience ... as a means of advancement and a corresponding reluctance to apply the principles of research..." (p. 6). This "rather gloomy ... assessment... has now, fortunately, to be tempered by the knowledge that in the past few years modest advances have been made as a result of the application of the methods of social sciences to the study of education ..." (p. 6). However, this sanguine assessment of what has been happening recently seems strangely out of place in the book. The new concern for the methods of social sciences, they say, " . . . has itself resulted in controversy and debate for, ... educational research has ... absorbed two competing views of the social sciences the established, traditional view and a more recently emerging radical view" (p. 6). It may well be that controversy and debate spur on the development of an area of study, but Cohen and Manion don't make a case for this having happened in research dealing with education. In fact, their introductory chapter leaves the reader with the impression that educational research is now self-consciously engaged in a conservative-radical contest at the expense of sustained attention to what is happening in and to the organizations, processes and experiences usually referred to as education. The first chapter is the longest chapter in the book. A reader gets the impression that it is the chapter on which the authors tried hardest and which they consider the most important part of their book. One even gets the impression that the authors wrote this chapter as a separate article or free-standing essay, and then decided that the strength of its arguments and the precision with which its theme is presented warranted turning it into a book (the authors acknowledge that parts of the chapter are taken from their earlier work, Perspectives on Classrooms and Schools, Holt-Saunders, Eastbourne, 1981). Unfortunately, the chapter is only a prettygood essay on the conservative and radical positions in social sciences research, 108 Book Reviews/Comptes Rendus and really is not a suitable foundation for a long book (383pp) on research methods. The authors are in trouble from p. 11 on. On that page they present one of their many boxes. These boxes appear frequently throughout the book, and seem to serve the purpose of a digest of important notions for readers in too much of a hurry to extract ideas paragraph by paragraph. This is probably a good pedagogical device, but only when the boxes correctly present the information carried by endless paragraphs of plain old prose (we hasten to add, though, that Cohen and Manion write easy-to-read paragraphs, and lucidly enough that the many visual-aids to understanding or reviewing their presentations are certainly not there to compensate for inadequacies in the prose). This first box purports to be a dichotomous presentation of the philosophical, methodological and technical differences between objectivists and subjectivists (who are, it seems, the traditionalists and radicals, respectively, on p. 6). The dichotomy is adopted from Thomas B. Greenfield's "Theory about Organizations: A New Perspective and its Implication for Schools" in Administering Education: Institutional Challenge (M.G. Hughes, ed., Athlans Press, London, 1975). Of more importance, it presents a picture of differences between traditionalists and radicals, and between objectivists and subjectivists, that doesn't exist and never has. The dichotomy is not even a good heuristic device, unless one wants to teach readers that objectivists and subjectivists are both quite silly in their approaches to research, and that objectivists are probably down-right stupid. This, in fact, is the weakness of the whole essay that is the pretty-good introductory chapter. It presents a picture of educational and social science researchers which suggests that they are really not very bright, but, on net, the subjectivists have more going for them than the objectivists. Nothing in the chapter prepares the reader for its concluding comment: The particular value of scientific research in education is that it will enable educators to develop the kind of sound knowledge base that characterizes other professions and disciplines; and one that will ensure education a maturity and sense of progression it at present lacks (p. 44). That sentence will remind every teacher of the groped-for termination that students so often use to end assigned essays when they don't want to write any more. It has little to do with all that went before, and, most regrettable in the case of an introduction to a book, it doesn't connect in interesting ways with what is to come. The content of this chapter justifies the "rather gloomy ... assessment..." with which it began more than the brave and hopeful assertion with which it ends. The fourteen chapters to follow range from pretty bad to pretty good, and none is as interesting as the introduction. Sadly, it must be said that there is virtually nothing to connect one of these chapters with any other, and the connections between any one of them and the introduction don't add much excitement to either the whole book or some part of it. Frankly, the fourteen chapters lack parallelism and congruity. Each of them is at least a little bit more defensible as a paper by itself than as a part of this book. 109 Book Reviews/Comptes Rendus Our candidate for the worst chapter is Chapter 2, Historical Research. It says nothing about the classic art of writing history, nothing about the evolution of the partially-scientific profession of history in the last century, and nothing about the place of educational histories in the larger art and science of history. The most egregious fault with this chapter is some of the erroneous assertions it contains, some, indeed, that the authors went to the trouble of quoting from other writers. For example, it is wrong to say that history "cannot depend ... on direct observation or experimentation..." (p. 47), it definitely can and sometimes does. Some of the assertions in this chapter simply reflect a lack of careful editing by the authors themselves. To say that "historical research does have some features in common with both normative and interpretive approaches to research..." (p. 47) really makes no sense. History is hardly an approach to research in the same sense as are "normative and interpretive approaches". And it is not at all clear what it would mean for the art and science of history to have something in common with these "approaches". Chapter 3, Developmental Research, is concerned with cross-sectional, longitudinal and trend studies of students and populations. "Collectively they are termed developmental research because they ... describe what the present relationships are ... and ... account for changes occurring in those relationships as a function of time" (p. 68). There are many other things developmental research might mean, especially coming after a chapter on historical research. Unfortunately, the reader never learns why developmental research is limited to so narrow a meaning in this chapter. Chapter 4, Surveys, does offer some clear advice on the planning and doing of surveys. It touches upon some of the problems and techniques of this popular tool of social science research. But the reader is left to wonder why surveys are treated as conceptually equal to historical research and developmental research. The same goes for Chapter 5, Case Studies, and Chapter 6, Correlation Research. Whatever the purpose of Chapter 6, it is not to teach statistics. Chapter 7, Ex Post Facto Research, one would expect to be related to Historical Research and the chapters between 1 and 7. Alas, it is not. Chapter 8, Experiments, Quasi-Experiments and Single-Case Research, is twenty pages long. It deals quite succinctly with much of what might be most interesting to traditionalists or objectivists, if any continue with this book beyond the first chapter. For readers who have been captivated by the authors' approaches to research, 20 pages is probably quite enough for what is covered here, and what once would have been the heart of a book on social science research. In Chapter 9, Action Research, the sort of enthusiasm for their work that characterized the authors in their first chapter reappears. " . . . (action research) is essentially an on-the-spot procedure designed to deal with a concrete problem ... the step-by-step process is constantly monitored over varying periods of time and by a variety of mechanisms so that the ensuing feedback may be translated into modifications, adjustments, directional changes, re-definitions, as necessary, so as to bring about lasting benefit... " ( p . 213). The authors do, though, acknowledge 110 Book Reviews/Comptes Rendus s o m e of the criticisms of action r e s e a r c h , e v e n q u o t i n g R . M . W . T r a v e r s : The writer's evaluation of the last fifty studies which have been undertaken which compare the outcome of one teaching methodology with another is that they have contributed almost nothing to our knowledge of the factors that influence the learning process in the classroom, (p. 215). The authors are not put off by such an assessment. They conclude the chapter with this comment, "... in the United States, it was found that the teachers taking part (in a sample of action research studies) were generally enthusiastic. They seemed to feel that the staff worked more as a unit than before the research, that staff members were drawn closer together with the knowledge that they shared problems and goals, and that respect for subordinates, both teachers and pupils, had increased." (p. 227) In Chapter 10, Accounts; Chapter 11, Triangulation ("the use of two or more methods of data collection in the study of some aspects of human behaviour", p. 254); Chapter 12, Role Playing; Chapter 13, The Interview; and Chapter 14, Personal Constructs, the authors continue to introduce their readers to the exciting aspects of educational research that are complementary to action research, that are likely to appeal to subjectivists, and are in accord with the "emerging radical view" they mentioned back in Chapter 1. There is no "concluding" chapter. Indeed, there are no concluding remarks. The final chapter, 15, is called Multidimensional Measurement. In fewer pages than are needed to deal effectively with the matter, they offer some mildly interesting comments on such subordinate topics as elementary linkage analysis, cluster analysis and factor analysis. It would have seemed appropriate to put Chapter 15 after Chapter 6, Correlational Research. As with Chapter 6, it is not quite clear what Chapter 15 is intended to do. It clearly is not intended to teach statistics. This is the second edition of Research Methods in Education, the first came out in 1980, and was reprinted in 1981, 1982 and 1984. The book is a success. We are, nonetheless, hard put to say why a student or a researcher would read it when the time spent could be put to studying methodology more methodically. The best use for the book demonstrated so far is as a source of facile descriptions of the methods of subjectivist approaches to research, and of disarming defenses of those approaches against criticisms such as Travers'. And what are such facile descriptions and disarming defenses good for? Well, they come in very handy in oral defenses of theses of the subjectivist genre now in vogue. John W. Holland and Saeed Quazi Higher Education Group The Ontario Institute for Studies in Education