Peter A. Taylor

The Use of Factor Models in Curriculum Evaluation: A Mathematical Model Relating two Factor Structures

AN earlier paper presenting a model of curriculum evaluation, written with Maguire (1966), proposed the sectioning of the evaluation process into a number of steps. The steps involved the identification of broad, societal-institutional objectives; the rephrasing of these latter into operational terms more suitable for measurement operations; the translation of operationally-stated objectives into teaching practices; and the student outcomes after being subjected to these practices. In the Taylor-Maguire model, it was stressed that different (though not necessarily disjoint) sets of individuals were involved at each step, and in passing judgment upon the values involved in the transition from step to step. It is quite obvious to all evaluators that these different sets of

Book Reviews : Testing Problems in Perspective: Twenty-fifth Anniversary Volume of Topical Readings from the Invitational Conference on Testing Problems by Anne Anastasi, Editor. Washington, D. C.: American Council on Education, 1966. Pp. xiii + 671.

tion in the name of science of empirically-derived scales measuring largely only two dimensions has involved much effort which might better now be re-directed. While the number of such scales has probably not yet reached three thousand, the stable (which now includes Block’s modest contribution) is becoming crowded. It is time to consider the task of cleaning. Or, because no Hercules has appeared in our midst, we had better seek fresher pastures and more modern quarters for a somewhat more carefully selected herd of oxen. Had Block laid the foundation for such a search with a fresh approach to personality assessment, I could have in honesty hailed it as attacking an important problem. Instead, we find only a further routine application of empirical item selection to produce yet another set of &dquo;scales&dquo; almost totally lacking in discriminant validity. In the future, rather than further overpopulate a barn already acknowledged by all to be inadequate, let us aspire to blaze

10.00

the Educational Process” by Duncan N. Hansen, VIII. “Multivariate Analysis” by Elliot hf. Cramer and R. Darrell Bock. The quantitatively oriented reader mill find much to offer in the chapters on design methodology, Bayesian statistics, nonparametTic statistics, factor analytic methodology, and multivariate analysis. The evaluative nature of the factor analytic chapter makes it particularly valuable for a student in the concerned field. The dedication of a chapter to the burgeoning field of multivariate analysis represents a long-needed recognition of the importance of this area. The chapter on computer-assisted instruction is highly informative and provides reference to many important documents of which the typical reader may not be aware. The authors of the chapter on the observation and recording of behavior are to commended, not only for excellent description of the pertinent literature but also for developing a classificatory system to describe the category systems which have been developed to record and describe behavior. The chapter on sample survey designs is essentially a “how to do it” guide for school administrators. Although the chapter achieves its objectives well, and although there was probably need for such a guide, there is a question of the appropriateness of including a chapter such as this in a review of educational research methodology. I n view of the large number of published and unpublished books and articles cited, the review can be valuable as a source of reference material alone. Despite minor shortcomings, Methodology of Educational Re search, should be a welcome addition to the library of anyone seriously interested in educational research and related topics.

Book Review: Assessing Behavior: Readings in Educational and Psychological MeasurementFlynnJohn T. and GarberHerbert. Reading, Mass.: Addison-Wesley Publishing Co., 1967. Pp. vi + 377 (paper),

many psychologists have tried to test these so-called postulates by various kinds of experiments (Bush, 1960, p. 127).&dquo; As we take leave of Psychomathonoma, the many splendored homilies disappear into the subsets in the basis of Mt. Olemmas. While up on the saddle points, reflecting barriers reverberate with the sound of the psychomathic credo, &dquo;Models are more often destroyed by better models than by experiments (p. 467).&dquo; But not all of the noise from Olemmas is signal; some of it is noise. For something is brewing up there besides the nectar of Dionysus; the lesser psychonomes are rumbling. Long disenchanted with their demi-status, they are demanding the immediate repeal of the Cauchy-Schwarz and Tchebycheff inequalities. The day will soon arrive when every psychonome on Olemmas will have the titular right to replace the sound of Jones, Smith, and Hall with Melodious monikers like Safarevic, Krasnoselskii, Nemyckii, Cebotarev, Ljusternik, or Hooke.

3.95.

cial science in general. The author writes well and includes numerous illustrations and examples, both within the chapters themselves as well as in the several questions and problems following each chapter. The topics surveyed are standard for elementary economics statistics, viz. frequency distributions, central tendency, dispersion, sampling, point estimation, confidence intervals, stochastic functions, significance tests, linear and nonlinear regression, time series, and index numbers. The majority of these topics are, of course, relevant to many of the other social sciences. The introductory material presented in the first four chapters of the book leads into the central topic of stochastic functions in Chapter 5, and the last five chapters deal with special applications. In summary, the book is well-written, representing more of a verbal than a mathematical exposition. The focus is on clear explanation and a practical approach through numerous examples. Consequently, it should prove to be more popular with students than most introductory statistics books.

Book Reviews : Basic Statistics: A Primer for the Biomedical Sciences by Olive Jean Dunn. New York: John Wiley and Sons, 1964 (second printing, 1966). Pp. xiii + 184:

achievement tests, standardized tests, and procedures for marking and reporting. Thus, the content presents nothing unusual or unique from the subject matter standpoint. Two condensed appendixes accompany the six condensed chapters. One appendix includes tables, and the other a well presented set of computational procedures for the statistical concepts presented in the main body of material. In summary, for those who like brief but authoritative attempts to treat a broad range of subject matter, Lindeman’s effort will be quite well received. Instructionally, the book might be considered for classroom use by the instructor who views a text as purely supplementary to his teaching. However, the instructor who treats a text as the basic anchor for instruction might very well go down with the ship using this book. Although only alluded to in the preface, its greatest utility might be as a brief and authoritative review for people who have received instruction in measurement and evaluation at an earlier period in their professional training. JAMES C. MOORE The University of New Mexico

Book Reviews : A Programmed Introduction To Statistics by Freeman F. Elzey. Belmont, California: Wadsworth Publishing Co., Inc., 1966. (Paper) Pp. viii + 376

a Commission on Plans and Objectives for Higher Education whose function was to be the continuing study of long-range problems facing colleges and universities, and to make policy recommendations where such were possible and appropriate. One of the early questions to which the Commission addressed itself was: &dquo;What are the present strengths and weaknesses of our graduate schools in providing well-trained scholars for both teaching and research?&dquo; Cartter’s monograph is a report on the attempts to answer this question.

Book Reviews : Introduction to Statistical Analysis and Inference for Psychology and Education by Sidney J. Armore. New York: John Wiley and Sons. 1966. Pp. xx + 545.

The KELP appears to have real potential as an aid to the kindergarten teacher in making progress reports and conducting teacher conferences. Since it requires keeping records and also draws the teacher’s attention to a variety of pupil behaviors, it should result in better focused conferences and reports. The norm tables for the KELP are not provided in the book, but may be obtained from the publishers. Little information is given regarding these norms. For example, the authors indicate that the 1965 norms are based on about 2000 cases. Virtually no information

8.95

H e calls for greater use of the institutional records already being kept to reveal the effects of changes, warns about various potential sources of invalidity (cf. Campbell and Stanley, 1963), and discusses several specific ways in which his suggestions can be carried out. I n summary, this book contains a wealth of information, and even more speculative discussion, which should stimulate any educational researcher. Bayesian analysis, research training, correlational analysis, incomplete experimental designs, programnintic research, and administrative experimentation are all discussed and many methodological and philosophical ideas are presented. With an exhaustive list of references and an adequate index, Improving Experimental Design and Statistical Analysis is one of the very best of the published Phi Delta Kappa Symposia, and a valuable addition to the libraries not only of educational researchers but also for all who are concerned with the methods and philosophies of the behavioral sciences.

Book Reviews : An Assessment of Quality in Graduate Education by Allan M. Cartter. Washington, D. C.: American Council on Education, 1966. Pp. xvi + 132.

on cognitive styles and Chapter 12 on race differences. The area of cognitive styles is very much to the fore at the present moment. This chapter gives a much-needed summary of the cognitive research to this point. The author sees the research of Witkin, Holtzman, Klein and many others as occupying a &dquo;strategic position in differential psychology.&dquo; The chapter on race differences is especially good. The author neither wholeheartedly accepts the notion that one race is inferior to another, nor does she deny the empirical existence of such differences. She simply calls for research concerning the possible