Sue Pinzur Rasher

Preenrollment variables and academic performance as predictors of freshman year persistence, early withdrawal, and stopout behavior in an urban, nonresidential university

Multiple group discriminant analysis was employed to determine the utility of preenrollment traits and academic performance in identifying freshman students who persisted, stopped out, or withdrew early from an urban, nonresidential university. An equation based on nine preenrollment variables significantly discriminated among the three groups and correctly identified 48.1% of an independent validation sample (p<.001 for the hypothesis that overall correct classification was a significant improvement on chance). The clearest separation based on preenrollment traits was between stopouts on the one hand and both persisters and withdrawals on the other. It was only after first-quarter academic performance was added to preenrollment traits that a sharp discrimination was found between persisters and early voluntary withdrawals.

Time, Teacher Comments, and Achievement in Urban High Schools

AbstractObservations of time usage and teachers statements were made in 175 classrooms in twenty-six high schools in Chicago. The measures of time were combined into one variable that was then tested for its association to the achievement scores of the eleventh grade class. The correlation showed that schools with higher reading achievement had a lower amount of lost time. Regression models showed that better use of time and more use of positive comments were each associated with increases in achievement. The low use of reinforcing comments by teachers and the high proportions of student time lost through absences, interruptions, and noninvolvement were cited as areas that could be affected by school policies.

Predictive Validity of Student Perception: A Cross-Cultural Replication

To test the cross cultural generalizability of measures of student perception of the social environment of learning, the mean end-of-course achievement scores of 166 groups of studious and non-studious members of 83 general science classes and 134 similar groups in 67 social studies classes randomly sampled in the State of Rajasthan, India were correlated with IQ and the 15 scales of the Learning Environment Inventory, (LEI, translated into Hindi). With IQ controlled, the median partial correlations of perception and achievement are .43 for general science and .58 for social studies. Multiple regression analysis shows that the LEI scales account for substantial achievement variance beyond that accounted for by IQ.

Childhood Traits and Environmental Conditions of Highly Eminent Adults.

respected biographies show their distinctive intellectual competence and motivation, social and communication skills, general psychological wholesomeness, and both versatility as well as concentrated perseverance during childhood. Most were stimulated by the availability of cultural stimuli and materials related to their field of eminence and by teachers, parents, and other adults. Although most had clear parental expectations for their conduct, they also had the opportunity for exploration on their own. Our research combines psychology and &dquo;Cliometrics,&dquo; the quantitative study of history, after Clio, the muse of history. The criterion of eminence employed (the number of words written about each man in biographical dictionaries and encyclopedias) proved technically reliable; but the ratings of traits and conditions are no doubt distorted to

Validity and reliability issues in measurement instrumentation

Abstract This article provides a brief overview of validity and reliability as concepts related to the overall quality of test instruments. We describe the nature and interpretation of content, face, criterion, and construct validity and identify several approaches for measurement and improvement of reliability. We urge that test developers, users, and interpreters recognize the characteristics of good instruments as well as the limitations of measurement instrumentation.

IQ Correlates With High Eminence

Indicators of eminence derived from word and citation counts in primary biographical articles in encyclopedias published at the turn of the century, in 1935, and 1974 correlate positively. 33 overall with IQ estimates made from biographical sources on a select sample of 282 philosophers, scientists, non-fiction and fiction writers, musicians, artists, religious leaders, statesmen, revolutionaries, and soldiers. These results are striking since the sample is restricted to the higher end of the eminence distribution; the mean estimated IQ for the total group is 158.9. Indicators of eminence for some fields-philosophers, musicians, and artists--vary from one period to the next. Individuals also shift in estimated eminence during the three time periods examined.

Eminence and Citations in Educational Psychology

pact of their writing on the field; and citation counts provide an admittedly imperfect index but one that has shown validity in predicting Nobel prizes and other recognitions of eminence (Wade, 1976). The primary question addressed here is whether the relation between eminence and citation holds in educational psychology. Since 1969, the Institute for Scientific Information (1975) has annually listed all writing that has been cited in a large sample of educational, psychological, and social science journals published during the year. The list includes the name of the cited author and reference in-

English Acquisition as a Diminishing Function of Experience Rather than Age.

Self-ratings of English fluency of 352 Japanese students who lived from 0 to 12 years in Japan and 0-12 years in the United States and ratings of English competence made by American teachers of an over-lapping sample of 360 students were regressed on linear and logarithmic forms of numbers of months lived in the U.S. and in Japan, and their product. A simple regression of the ratings on the log of months in the U.S. yields Rs of from .52 to .62 which are generally not increased significantly by additional terms. The hypothesis of special competence of young children for language acquisition is unsupported. The rate of acquisition, however, diminishes with time in this country in children of all ages in the sample.

Improving Regression Models.

This paper illustrates cut-and-try techniques that point to appropriate transformations of variables and to the selection of sets of variables for an equation that may improve understanding of a social process. The substance of the research reported — the relation of mental test results to state population, cultural, and school resource indexes (Walberg and Rasher, 1974) — illustrates typical problems of behavioral data: multi-colinearity, outliers, abnormal distributions, and the lack of a consensually-validated, explicit theoretical model. Despite these problems, data originally collected for purposes other than the investigator’s may yield tentative confirmations or cautions about prior findings and provisional indications for theory or policy; such inferences may be at least partially checked by cross-validation on independent or semi-independent sets of data. After discussing the sequence of analyses and the results, we conclude by mentioning a number of uncertainties and reservations about drawing substantive or policy implications.

An operational test of a three-factor theory of classroom social perception

To test the cross-cultural generalizability of measures of student perception of the social enviornment of learning and an operationalization of the social environment of learning and an operationalization of a three-factor theory of the social-psychological properties of small groups (Moos-Bales), the mean end-of-course achievement scores of 166 groups of studious and nonstudies members of 83 general science classes randomly sampled in the State of Rajasthan, India, were regressed on group perception of affect, task orientation, and competitveness (derived from scales of the Learning Enviornment Inventory, LEI, translated into Hindi) and control variables–school and teacher characteristics, and student IQ sex, and socioeconomic status. Nine variables selected for unique contribution to accountable variance in achievement included the four perceptual factors, produced a multiple correlation R of .86, and a cross-validated R of .83 in an independent sample of 67 social studies casses.