John L. Wasik

Sociometric Characteristics of Kindergarten Children at Risk for School Failure.

Abstract The purpose of this study was to determine if children identified by teachers at different levels of risk for school failure were distinguished by their peers in terms of social status and behavioral descriptors. Two sociometric measures, nominations and ratings, were used to assess social status. Childrens perceptions of their peers were also obtained for a set of behavioral descriptors. A total of 348 kindergarten children in one school for three successive years were included in the data analysis. Children were classified by teachers as at no risk, moderate risk, or high risk for retention. The results from multivariate analyses of variance showed a significant effect for risk groups for both sociometric measures and descriptors. Univariate test results showed the no-risk group was viewed more positively than the other two groups on the sociometric measures. The high-risk group was judged to be more disruptive and aggressive and to prefer to be alone more. Gender was a significant factor for the peer nomination data and behavioral characteristics but not for the class rating. The results are discussed in relation to early intervention with at-risk children.

Basic Concepts and Conservation Skill Training in Kindergarten Children

This study investigated the effects of basic concepts training on conservation acquisition in kindergarten children. A total of 41 children, 17 white boys, 15 white girls, 6 black girls, and 5 black boys, were randomly assigned within level of race and sex to four groups: Group 1, Basic Skills Training and Conservation Training; Group 2, Basic Skills Training; Group 3, Conservation Training; and Group 4, Contact Control. The experimental design called for Groups 1 and 2 to receive training on basic concepts, specifically “more than,” “less than,” “same,” and “different.” Following a second test session, Groups 1 and 3 then received training on conservation while Group 4 served as a contact control group for both training sessions. Only the conservation training program resulted in significant effects and that was for the white students alone. A retention testing administered several months after the conservation training showed a significant drop for those students who had received the conservation training and a significant increase for the contact control group.

Gentest: a Computer Program to Generate Individualized Objective Test Forms

A computer program to generate individualized objective test forms for use in a Student Paced Statistics (SPS) course is described. The program features disproportionate sampling from different item domains and enhanced character generation facility for test printing purposes.

Teacher Perceptions of Behaviors Associated With Creative Problem Solving Performance

This paper presents the results of a study designed to determine the perceptual cues used by teachers to identify creative problem solving ability. A battery of eight Structure of Intellect (SI) ability tests were administered to 162 10th grade students of one school in a large suburban school district on the Mid-Atlantic seaboard. Ratings of student creative problem solving ability obtained from four teacher teams were related to the SI measures by the use of the Christals Judgement Analysis Procedure. It was concluded that the four teacher teams were using the same cues (SI abilities) in rating students in terms of creative problem ability and that these cues were related to types of abilities measured by scholastic aptitude tests.

SEGMENTED POLYNOMIAL MODELS IN QUASI- EXPERIMENTAL RESEARCH

An introduction to the use of segmented polynomial models in behavioral science research is presented. Examples of suitably defined design matrices of dummy variables are given for the least squares analyses of time series and discontinuity quasi-experimental research designs. Linear combinations of dummy variable vectors are shown to provide tests of effects in the two quasi-experimental designs. Interpretation of results from the segmented polynomial analyses is also demonstrated.

Book Reviews : Fienberg, Stephen E. The Analysis of Cross-Classified Categorical Data. Cambridge, Massachusetts: the MIT Press, 1977. × + 151 pp,

ments might come about through insufficient revision. One example is the advice to use &dquo;disc recorders&dquo; to gain verbatim protocols when, to my knowledge, cassette tape recorders have gained total supremecy in the clinical market. Another element is an all too brief section on &dquo;gaining rapport.&dquo; Aside from my dislike of the word (it suggests manipulating the subject to gain data from him), there is the implication of a very narrow relationship. Currently, in most settings, the psychologist is more than a tester. The patient or client is interviewed and

10.95:

This text was originally published in 1942 as one of the first statistics texts written specifically for a behavioral science audience. The fact that the text is still in print suggests it continues to meet the need of potential and practicing behavioral scientists. The last review of this text to appear in Educational and Psychological Measurement was by James Walsh in 1966 on the fourth edition. Since that review, the text has undergone two more revisions and acquired a co-author, Professor Benjamin Fruchter of the University of Texas at Austin. This reviewer first came in contact with the text through the third edition while taking a graduate course in educational statistics. Thus, reading the sixth edition was like renewing an acquaintance with an old friend, a feeling heightened by the fact that a majority of the textual material has not been changed over the years. The emphasis of this review, therefore, will be on reporting changes in content that have occurred since the fourth edition, the last edition to have been reviewed in Educational and Psychological Measurement. References will also be made to changes observed in the fifth edition since this was the time Professor Fruchter’s first became associated with the writing of the text.

Book Reviews : J. P. Guilford and Benjamin Fruchter. Fundamental Statistics in Psychology and Education, 6th edition, New York: McGraw-Hill Book Co., 1977, Pp. xix + 555.

meaning of the actions of others and disturbances of ability to perceive oneself normally. Self-concept and body-image are said to be related. This volume ranges from over-simplification to expectations for ascertaining highly complex behavior disorders on the basis of relatively routine teacher observations. Psychologists might not gain much from it but if interpreted cautiously, it should be helpful to some educators, especially perhaps to teachers with limited classroom experience.

13.95

descriptions of the studies, there is no way to know if these 33% are themselves methodologically flawed. In the new epilogue Rosenthal also deals with some of the criticisms of Pygmalion in the Classroom. He briefly presents the criticisms that have been levied by Jensen, Thorndike, Elashoff, and Snow, and responds to them. Neither the criticisms nor the responses are as extensively presented as the research warrants, especially in light of the considerable publicity the work has received, and in light of it being the research category of greatest increase since 1969.

Book Reviews : Evaluation Studies Review Annual, Volume I. Edited by Gene V. Glass, Beverly Hills, California: Sage Publications, 1976.

names and characteristics of the children studied, their basic ratings, lists of correlations, lists of child-rearing emphases rate by Topeka mothers, and a long inventory of coping variables. The book concludes with series of notes relevant to earlier chapters. This book is a valuable contribution of our knowledge of child behaviors. It is a delight to read. It suggests numerous hypotheses and problems worthy of further longitudinal research.