Factor analysis by minimizing residuals (minres)

This paper is addressed to the classical problem of estimating factor loadings under the condition that the sum of squares of off-diagonal residuals be minimized. Communalities consistent with this criterion are produced as a by-product. The experimental work included several alternative algorithms before a highly efficient method was developed. The final procedure is illustrated with a numerical example. Some relationships of minres to principal-factor analysis and maximum-likelihood factor estimates are discussed, and several unresolved problems are pointed out.The authors wish to thank the Factor Analysis Work Group (supported, in part, by ONR) for valuable criticisms and suggestions made in the course of a discussion of the present work in April, 1965.     

The Development and Evaluation of a Measure of Counselor Effectiveness

The purpose of this research was to construct and evaluate a measure for assessing counselor-trainee effectiveness. An eight-category rating instrument was constructed and two standardized interview situations emphasizing problems commonly encountered by secondary school counselors were developed. The standardized interview situations provided by the use of coached counselees permitted a panel of six trained judges to evaluate via closed-circuit television 30 counselor-trainees under conditions designed to insure comparability and, hence, fair appraisal. Using a definition of reliability as the degree of consistency between judges, the average correlation between ratings given by pairs of independent raters was .50. It was concluded that the rating instrument developed and the techniques employed for its use constituted a usable, realistic measure for assessing the counseling proficiency of counselor-trainees.     

Children's judgments of body size and distortion

Body size estimates of 69 children aged 5–13 years were obtained using a TV video methodology. Tasks included an adjustment procedure where children manipulated the width of their body image on a TV, and a discrete task, where they judged the accuracy of their TV image which was presented as either normal or distorted, that is, too wide or too thin. On the continuous task, judgments were more accurate on ascending trials where children had to increase the width of the image. On the discrete task, a signal detection analysis revealed older children were better in detecting size distortion in their images and that the differences were due to differences in the biological sensory system and not due to a bias to report that they were too fat or too thin. However, there were response bias differences between genders. Females were more likely to report size distortion with increasing age while males were less likely.     

Preemployment screening procedures: How human resource managers perceive them

A survey of Human Resource practitioners' perceptions of pre-employment screening procedures was conducted. The results indicated differences between tests in terms of perceived accuracy and offensiveness, as well as other attributes. Regression analyses suggested that beliefs about validity, the number of other companies using these procedures, perceived offensiveness of the test, and ease in passing the test were important factors in the willingness to use a particular test. Conversely, perceptions regarding the discriminatory impact of the test and the cost of using the test generally were not related to willingness to use the test. Limitations of this study, which was primarily exploratory, are noted.     

VSearch: Macintosh software for experiments in visual search

In this paper, we describe software that turns a Macintosh computer into an off-the-shelf tool for experiments on visual search. Our design goals included portability (between members of the Macintosh computer family and between various research settings), user-friendliness (equivalent to Macintosh programming standards), flexibility (to allow replication and extension of important experiments on visual search), and adaptability (very short design-to-data and data-to-analysis turnaround times). We describe how the software meets these goals in three major phases of an experiment: stimulus construction, experimental control, and statistical analysis. We then list several landmark studies of visual search that can be easily designed and extended with the software. Finally, we outline plans for expanding the experimental variations that will be supported in future versions of the software.