David E. Wiley

Studies of a class of covariance structure models

Abstract This article demonstrates the usefulness of analysis of covariance structure procedures for estimating variance components under more general assumptions than are provided in the typical mixed-model analysis of variance. A sequence of eight models is presented based on varying degrees of restrictive assumptions. Maximum likelihood procedures are employed in the estimation of the parameters of the models. The procedures are applied to simulated data and to an empirical example which shows the necessity of the more general assumptions.

Teacher Reports on Opportunity to Learn: Analyses of the 1993 California Learning Assessment System (CLAS)

The California Learning Assessment System 1993 OTL Study allowed examination of the extent to which California public school mathematics teachers had been exposed to the practices consistent with those advocated in the California Curriculum Frameworks; the extent to which those teachers implemented these practices in their classrooms; and their impact on student performance in mathematics. The findings support the conclusion that curriculum and instruction that require higher level skills benefit students’ achievements. The data are the best available measures of California students’ learning of higher level skills, and provide consistent information for discussion and debate among educators about reform strategies.

Who takes what math and in which track? Using TIMSS to characterize U.S. students' eighth-grade mathematics learning opportunities

This article examines the range of eighth-grade mathematics learning opportunities in the United States, drawing on data gathered for the Third International Mathematics and Science Study (TIMSS). Sources of variation in the provision of learning opportunities are identified, and patterns in eighth-grade mathematics course offerings are compared across schools. Comparison of students’ learning opportunities includes consideration of the specific course in which they were enrolled, the type of textbook employed for the course, and the proportion of time teachers devoted to teaching specific topics. Analyses revealed a mismatch between the mathematics course title and the textbook employed in the course for nearly 30% of U.S. eighth-grade students. Course-textbook combinations demonstrated significant relationships with the time teachers devoted to specific topics and the international topic difficulty score. Some differences in mathematics learning opportunities were found on the basis of a school’s location (urban, rural, suburban), size, and percentage of minority enrollment. We contend that the significant relationships found were not consistent enough to provide a satisfactory explanation for the observed variation and that individual student differences, which often provide an implicit rationale for tracking, also represent an inadequate explanation for the observed diversity in curricular opportunities. We conclude that recognizing the multiple definitions for students’ classroom mathematics learning experiences is an important step in reform and policy discussions.

Latent partition analysis

Latent partition analysis has been formulated to study the relationships between two or more partitions of the same set of items. The major structural hypothesis is that a latent partition underlies the manifest partitions; that is, it is assumed that each item belongs to a latent category and that the manifest categories are derived by dividing and combining the latent categories. We have found that by examining manifest categories it is possible to reconstruct information about the latent partition and about its relation to the manifest partitions.

Conceptual Issues in Models of School Learning

†This paper is a translation of ‘Kernkonzepte des Schullernens’, Zeitschrift fur Entwicklungspsychologie und Padagogische Psychologie, Vol. 9, No. 3 (July, 1977). It is published here with the permission of Verlag fur Psychologie, Dr C. J. Hogrefe. The authors assume joint and equal responsibility for this work. David E. Wiley is also affiliated with the University of Chicago.

ESTIMATING CONTENT-ACQUIESCENCE CORRELATION BY COVARIANCE STRUCTURE ANALYSIS

A topic of continuing interest in the measurement area is response acquiescence. A recent study has demonstrated the feasibiliy of studying acquiescence or, more importantly, content/acquiescence correlation in the MMPI. Utilizing the components of variance approach, this study found that the variance due to acquiescence in scores on the Pt and Hg scales was small relative to content variance, but that the correlation between acquiescence and content may be substantial for the Pt scale. The present paper describes a general statistical procedure for investigating content variance, variance due to non-content characteristics of items, and the covariances of content and various item characteristics. The data from a previous paper are reanalyzed, using alternative covariance structure models. Maximum likelihood procedures which allow for a statistical test for parameters of interest are used. The results point to the significance of the content- acquiescence correlation in the Pt scale, but not in the Hy scale. The previous findings are verified statistically, and procedures which hold promise for other investigation into the properties of behavioral tests are described.

Book Reviews : Exploratory Data Analysis John W. Tukey Reading, Mass: Addison-Wesley, 1977, Pps. xvi +688.

the confidence with which it can be asserted that they will reappear in further samples from the same population. Since the emphasis and the language of contemporary statistical analysis is oriented almost exclusively toward confirmation, Exploratory Data Analysis (EDA) will strike many readers as being radically new; the style and format of the book will help to confirm the impression. Ironically, EDA returns to some old themes which have been much neglected recently. The principal theme is flexibility of technique. Good exploration requires many and varied analyses of the same data, and a certain amount of trial and error is expected. Creativity in approach may, therefore, be crucial. Precision and efficiency in any particular analysis are not nearly as critical as robustness and convenience. Consequently, all the techniques in this book can be performed manually and require minimal computation. Since the median is both robust and convenient, it is almost invariably employed as a location statistic; and the interquartile range is the favorite measure of spread. A secondary theme is that structures and patterns in data fall into two broad classes: the obvious and the surprising. The techniques discussed can be correspondingly divided into those designed to display clearly and simply the first level of structure and those intended to look beyond those relations to uncover the unexpected features underlying what confirmatory analysis might dismiss as residuals

17.95

ALTHOUGH the assessment of the meaning of concepts and attitudes has often been studied (e.g., Merton, 1957; Nemcomb, 1958), i t is still difficult to quantify word meaning. The Semantic Differential is a simple and systematic procedure for quantifying the connotative meaning of concepts (Osgood, Suci, and Tannenbaum, 1957). It offers an original approach to the old but important problem of measuring the meaning of words. I n the present study one facet of this important problem of measuring the meaning of words mas investigated. The connotative meaning among college students of the term “Public School Teachers” mas studied by the Semantic Differential. I n two other studies (Huselc and Wittrock, 1962; Wittrock, 1964), the Semantic Differential technique was used to determine the factors of connotative meaning for the concepts “School Teachers” and “Public School Teachers.” I n the latter study, six factors were found with a heterogeneous population. The factors Fere: Expressiveness, Tenacity, Stability, Potency, Predictability, and Evaluation. The conclusion of these studies mas that the factor structure depends upon the heterogeneity of the population of subjects as well as upon the concepts they are rating and the scales they use in the rating.

The Connotative Meaning of the Concept “Public School Teachers”: An Image Analysis of Semantic Differential Data

disservice to the students who are most likely to find the book useful. The bald statement that the concept of memory is less abstract than the concept of intelligence raises questions concerning the authors’ understanding of each of these ideas. The statements, &dquo;Because standardized intelligence tests exist, instances of varying levels of intelligence would not be difficult to obtain,&dquo; followed by: &dquo;On the other hand, the objective verification of instances of varying degrees of anxiety would present a problem,&dquo; lead to speculation about polygraphs which indicate anxiety and

Book Reviews : William E. Mendenhall. Introduction to Probability and Statistics. (2nd Ed.) Belmont: Wadsworth, 1967, Pp. xiii + 393.

a good textbook for an introductory course on experimental design for students in the behavioral sciences. The approach is practical and intended for students with a wide range of mathematical ability but presents a relatively rigorous and logical development of the theory. The comment relevant to the applicability of this book to an introductory course in design needs further clarification. The traditional approach to design in the behavioral sciences has been to introduce the subject in the context of the analysis of variance. The usual practice is to state the randomization procedure and the resolution of the sums of squares appropriate to the design. An ANOVA table is then presented which summarizes this resolution and indicates the methods of testing important hypotheses. One often finds the student memorizing formulas without understanding the nature of the theory associated with the underlying model. Mendenhall’s notions of model building consist of: specification of the model and design, estimation of the parameters of the model, parameter hypothesis testing and the determination of the lack of it. These are presented in a logical fashion leading the student through some of the most important notions of applied statistics. Fixed models which are linear in the parameters and the corresponding least squares theory receive most attention in the book. This approach to design and analysis gives the student a framework