An empirical comparison between some models for multidimensional scaling

Abstract.— The results of three methods for multidimensional scaling—Torgerson's metric analysis, a non-metric method (TORSCA), and a method dealing with individual differences in multidimensional scaling (INDSCAL)—were compared in experiments on rhythm experience and on perception of sound quality. The INDSCAL analysis seemed to be the most adequate method for treating the data in these experiments.     

An empirical comparison of several interval estimation procedures for coefficient alpha

Following a general introduction to Tukey's Jackknife technique and the construction of approximate confidence intervals, ten variants of the procedure are defined for giving interval estimates for coefficient alpha. A data base of known population characteristics was generated and used to compare the robustness of the ten Jackknife alternatives with Feldt's well-known sampling theory based upon the assumptions of normality. The empirical results indicate that out of the ten variants, defined by five transformations and two methods of data subdivision, only two are justifiable competitors to Feldt's approach.     

An empirical comparison of methods of communality estimation

This paper shows the formulation of nine methods of estimating the unknown communalities. Each of these methods has been used on experimental data and the results tabulated for comparison. The results show that the most accurate approximations are obtained from the Centroid No. 1 and theGraphical methods.The author wishes to express his appreciation to Professor L. L. Thurstone for his advice and for providing the facilities of the Psychometric Laboratory for this investigation.     

An empirical comparison of Maslow's and Murray's needs systems

Summary: The Work Motivation Inventory (WMI), a measure of Maslow's hierarchy of needs, and the Edwards Personal Preference Schedule (EPPS), a measure of Murray's manifest needs, were administered to 372 undergraduates. The two instruments were compared using canonical analysis. The analysis revealed three significant relationships between components of the two instruments. The first relationship supported Maslow's need hierarchy in general and its measurement by the WMI. The second suggested a fluctuating relationship between giving and receiving help and the levels of Maslow's hierarchy. The third relationship sumested that need for Achievement is associated with the intermediate levels of Maslow's hierarchy.     

An empirical comparison of rhythm in language and music

Musicologists and linguists have often suggested that the prosody of a culture's spoken language can influence the structure of its instrumental music. However, empirical data supporting this idea have been lacking. This has been partly due to the difficulty of developing and applying comparable quantitative measures to melody and rhythm in speech and music. This study uses a recently-developed measure for the study of speech rhythm to compare rhythmic patterns in English and French language and classical music. We find that English and French musical themes are significantly different in this measure of rhythm, which also differentiates the rhythm of spoken English and French. Thus, there is an empirical basis for the claim that spoken prosody leaves an imprint on the music of a culture.     

An empirical comparison of restricted and general latent distance analysis

Latent distance analysis provides a probability model for the non-perfect Guttman scale; the restricted latent distance structure is simpler to compute than the general structure. Since no sampling theory for latent structure analysis is available, the advantages of the general structure cannot be expressed formally. The two structures are compared in terms of their fit to fifteen sets of empirical data. The computation schemes used are summarized.     

An empirical comparison of natural-father and stepfather family systems.

This study examined the family system differences between 40 volunteer natural-father and stepfather families. Family triads consisting of the husband, the wife, and a child whose age ranged from 12 to 15 years were studied. Four instruments were used: (a) the Family Concept Q-Sort; (b) a Semantic Differential; (c) a demographic questionnaire; and (d) an interaction-reaction questionnaire. Analyses of variance on the data obtained from the Q-sorts and the Semantic Differentials indicated that stepfather family systems are different from natural-father family systems along several salient dimensions including psychological adjustment, satisfaction with family, reciprocal understanding, and perceived goodness and potency. It was concluded that the differences between the family systems in terms of their interpersonal relations and perceptions affect the entire stepparent family system and its ability to function adequately.     

The structure of emotion: An empirical comparison of six models

There are a number of conflicting approaches to the problem of the relationship between different emotions. One category of models focuses on the valence of emotional experience and typically reports evidence for dimensional approaches to emotion. A second category of models argues for the possibility of discrete basic emotions, but typically focuses on evidence from emotion systems other than conscious experience. In the present study, a list of emotion terms was drawn up that were derived conceptually from a set of basic emotions. A group of participants filled out a questionnaire that asked how much in general they experienced each of the emotions. A structural equation modelling approach was then used with the resultant dataset, an approach that permitted the comparison of six different models that ranged from the positive-negative affect models to models of discrete basic emotions. The analyses gave support for a set of five basic emotions but only when these were allowed to correlate with each other. Implications for theories of emotion are considered.     

Predicting group-level outcome variables: An empirical comparison of analysis strategies

This study provides a review of two methods for analyzing multilevel data with group-level outcome variables and compares them in a simulation study. The analytical methods included an unadjusted ordinary least squares (OLS) analysis of group means and a two-step adjustment of the group means suggested by Croon and van Veldhoven (2007). The Type I error control, power, bias, standard errors, and RMSE in parameter estimates were compared across design conditions that included manipulations of number of predictor variables, level of correlation between predictors, level of intraclass correlation, predictor reliability, effect size, and sample size. The results suggested that an OLS analysis of the group means, with White’s heteroscedasticity adjustment, provided more power for tests of group-level predictors, but less power for tests of individual-level predictors. Furthermore, this simple analysis avoided the extreme bias in parameter estimates and inadmissible solutions that were encountered with other strategies. These results were interpreted in terms of recommended analytical methods for applied researchers.     

Diagnosing Autism using ICD-10 criteria: A comparison of neural networks and standard multivariate procedures

In a sample of 976 consecutive cases derived from the recent world-wide Field Trial of Autism and other Pervasive Developmental Disorders, we tested the accuracy of the 15 ICD-10 criteria for the diagnosis of Autism, by comparing neural network models (NN) to more conventional multivariate competitors, namely, linear and quadratic discriminant function analyses and logistic regression. NNs were less accurate than competitors, both in terms of cross-validation results as well as in levels of shrinkage from training to test conditions. The clinical research implications of these results are discussed.     

A comparison of several procedures for eliminating behavior

The pecking response of pigeons was reinforced when a minimum period of time had elapsed since the last response (DRL schedule of food-reinforcement). Punishment, satiation, extinction, and stimulus change were employed separately to reduce responding. When the effects of the four procedures were compared, punishment was found capable of producing a more immediate, complete and long lasting response reduction than the others. Punishment had its maximum effect on the responses that were least relevant to reinforcement. The punishment reduced the frequency of the short inter-response times to a greater extent than did either extinction or satiation. In this way, punishment actually increased the efficiency of the DRL responding.     

Stimulus-driven attentional capture: An empirical comparison of display-size and distance methods

Four experiments examined attentional capture by colour as assessed by two different investigative methods. Subjects performed a visual search task for a vertical-target line embedded among tilted-distractor lines, presented inside 4, 8, or 12 coloured discs. Interestingly, when the colour singleton was task irrelevant, and data were analysed by means of the display-size method combined with the zero-slope criterion, no evidence for attentional capture by colour was found. However, when data were analysed by means of the distance method, which consists of monitoring the spatial relationship between the target and the singleton, results showed that the target was found faster and/or more accurately when it was inside the singleton than when it was in a nonsingleton location. This provided evidence for a stimulus-driven attentional capture. In addition, the application of signal detection methodology showed that attentional capture, as revealed by the distance method, resulted from a perceptual modulation at the singleton location, rather than from a criterion shift. We conclude that, at least with the kind of stimuli used here, the display-size method combined with the zero-slope criterion is less than ideal for investigating how static discontinuities can affect the automatic deployment of visual attention.     

Overt integrity tests versus personality-based measures of delinquency: An empirical comparison

This study examined the ability of both an overt integrity test and a personality-based measure of delinquency to predict on-the-job theft and counterproductivity. A heterogeneous sample of 105 employees anonymously completed both tests and a criterion checklist that included a 9-item theft subscale and a 12-item counterproductivity subscale. Results indicated that the overt integrity test was a superior predictor of both on-the-job theft and drug abuse compared to the personality-based measure. The overt integrity test was also a slightly better predictor of general counterproductivity. These findings and their implications are discussed.     

A model for ordered metric scaling by comparison of intervals

This paper presents a model of individual choice behavior for application to experimental situations in which a subject is required to compare utility intervals (differences in subjective value). This model is contrasted with a weaker model, which is also derived. Both models generate ordered metric scales, but differ in predictive power. An experiment on the utility of grades, which provides a test and comparison of the models, is presented.This research was supported in part by Group Psychology Branch, Office of Naval Research, under Contract Nonr 225(17), with Stanford University, and in part by U. S. Public Health Service grant M-2046.The author wishes to express his indebtedness to E. W. Adams, University of California, whose constructive criticism did much to improve the quality of this work.     

An evaluation and comparison of time-out procedures with and without release contingencies

A common recommendation for implementing time‐out procedures is to include a release contingency such that the individual is not allowed to leave time‐out until no problem behavior has occurred for a specific amount of time (e.g, 30 s). We compared a fixed‐duration time‐out procedure to a release contingency time‐out procedure with 4 young children (3‐ and 4‐year‐olds) using a reversal and multielement design. Results demonstrated that both time‐out procedures were effective at reducing problem behavior outside time‐out, problem behavior occurred in time‐out during both procedures, and problem behavior in time‐out was not predictive of problem behavior outside time‐out.     

An individual differences model for multidimensional scaling

A quantitative system is presented to permit the determination of separate multidimensional perceptual spaces for individuals having different viewpoints about stimulus interrelationships. The structure of individual differences in the perception of stimulus relationships is also determined to provide a framework for ascertaining the varieties of consistent individual viewpoints and their relationships with other variables.This research was supported in part by the National Institute of Mental Health, United States Public Health Service, under Research Grants M-2878 and M-4186 to Educational Testing Service, in part by Educational Testing Service, and in part by the Office of Naval Research under Contract Nonr-1834(39) and the University of Illinois. The authors wish to thank Drs. Harold Gulliksen and Douglas N. Jackson for their helpful comments and Miss Henrietta Gallagher for supervising the computations. Portions of this paper were presented at the American Psychological Association meetings in Chicago, September 1960.This paper was written while Dr. Messick was a Fellow at the Center for Advanced Study in the Behavioral Sciences.