Generalized approaches to the maxbet problem and the maxdiff problem,with applications to canonical correlations

Van de Geer has reviewed various criteria for transforming two or more matrices to maximal agreement, subject to orthogonality constraints. The criteria have applications in the context of matching factor or configuration matrices and in the context of canonical correlation analysis for two or more matrices. The present paper summarizes and gives a unified treatment of fully general computational solutions for two of these criteria, Maxbet and Maxdiff. These solutions will be shown to encompass various well-known methods as special cases. It will be argued that the Maxdiff solution should be preferred to the Maxbet solution whenever the two criteria coincide. Horst's Maxcor method will be shown to lack the property of monotone convergence. Finally, simultaneous and successive versions of the Maxbet and Maxdiff solutions will be treated as special cases of a fully flexible approach where the columns of the rotation matrices are obtained in successive blocks.The author is obliged to Henk Kiers for computational assistance and helpful comments.     

Evaluation of a mutual help network for the elderly residents of planned housing

We set up a mutual help network among the elderly residents of a government-subsidized apartment building and evaluated its impact on their social ties, social support satisfaction, and depression. A quasi-experimental design was used, with the residents of a similar building located in the same neighborhood serving as the control group. A total of 230 individual services were exchanged, and 28 group activities were organized during the study period. There was almost no difference between experimental and control groups with respect to the change over the study period in the number of social ties with other residents of the building. Support satisfaction decreased in both groups, but the decrease was larger in the experimental group. The control group had a slight decrease in the frequency of their depressive symptoms, whereas the experimental group had a moderate increase. More research is needed on the ability of social support interventions to produce beneficial effects on mental health among individuals who are not experiencing a particular life stress.     

The association between teacher attitudes, behavioral intentions, and smoking and the prevalence of smoking among seventh-grade students.

Seventh-grade students and teachers from twenty-three middle and junior high schools were surveyed to determine the association between teacher attitudes, behavioral intentions, and smoking behavior and the prevalence of student smoking. Teacher attitudes toward smoking policies were found to be strongly related to the current smoking behavior of the teacher but not consistently related to student smoking. In addition, teacher likelihood of intervening showed a modest association with teacher smoking status, with current smokers generally indicating being the least likely to intervene for student possession or use of cigarettes. Teacher intentions to intervene were strongly associated with the prevalence of smoking among boys but not girls.     

Covariation detection in treated and untreated spider phobics.

Twenty treated and 18 untreated spider phobics were exposed to a series of 72 slides. Three different categories of slides were used: phobia-relevant slides (spiders), alternative fear-relevant slides (weapons), and neutral slides (flowers). Slides were randomly paired with either a shock, a tone, or nothing at all. Despite the absence of a systematic correlation between slides and outcomes, untreated phobics strongly overestimated the covariation between spider slides and shock. Treated phobics did not show a covariation bias, suggesting that such bias can be modulated by behavioral treatment. In addition, untreated subjects were more confident about their contingency estimates than were treated subjects. The present results fit with earlier studies.     

Integration of the big five and circumplex approaches to trait structure.

To integrate the 5-dimensional simple-structure and circumplex models of personality, the Abridged Big Five Dimensional Circumplex (AB5C) taxonomy of personality traits was developed, consisting of the 10 circumplexes that can be formed by pitting each of the Big Five factors against one another. The model maps facets of the Big Five dimensions as blends of 2 factors. An application to data consisting of 636 self-ratings and peer ratings on 540 personality trait adjectives yielded 34 well-defined facets out of a possible 45. The AB5C solution is compared with simple-structure and lower dimensional circumplex solutions, and its integrative and corrective potential are discussed, as well as its limitations.     

Abstract versus modality-specific memory representations in processing auditory and visual speech

Serial recall of lip-read, auditory, and audiovisual memory lists with and without a verbal suffix was examined. Recency effects were the same in the three presentation modalities. The disrupting effect of a suffix was largest when it was presented in the same modality as the list items. The results suggest that abstract linguistic as well as modality-specific codes play a role in memory for auditory and visual speech.     

Effects of protein malnutrition and the administration of cortisol on the reflexologic development in rats.

The use of neurological tests in various states of undernutrition has confirmed the presence of delays in reflex ontogeny. Glucocorticoid administration in conditions of normal nutrition produces diminished growth and alterations in brain development similar to those observed in malnutrition, but there is not enough evidence about its effect on reflex development. In addition, some facts lead us to think that the use of permissive doses of these hormones during gestation could alleviate some of the effects of protein malnutrition on the development of pups. In the present study the ages at which certain reflexes appeared in the offspring of Wistar rats (Rattus norvegicus) fed a 20% (N) or 10% (M) protein diet from a week before conception to the end of lactation were observed. Each dietary group was subdivided into two: mothers receiving a daily injection during gestation either of a 0.9% NaCl (S) solution or of a 0.5 mg/100 g weight hydrocortisone acetate (C) solution. Results showed that the NS group of pups generally presented earlier ages of appearance of the variables under study than the remaining groups. Pups of the NC group presented delays in the age of appearance of reflexes. Finally, protein malnutrition delayed the appearance of reflexes and this effect seemed to be attenuated by treatment with hydrocortisone. Thus, hydrocortisone administration in protein malnutrition conditions seems to be beneficial although the results are harmful when administered under normal nutritional conditions.     

The effect of a non-informative cueing signal in a three-choice reaction-time task

Summary Many studies have shown that in a simple reaction-time (RT) task a non-informative leading cue decreases RT to a subsequent target presented in the same region as the cue. This facilitation lasts about 150 ms; it is then replaced by a sustained inhibition. Experiment 1 tests the hypothesis that the initial facilitation is due to motor, rather than perceptual, processes by using choice RT. Facilitation is greater and lasts longer in choice than in simple tasks, thus supporting the hypothesis. Experiment 2 shows that facilitation occurs even when the subject actively orients his or her attention away from the cue.     

Alternating least squares algorithms for simultaneous components analysis with equal component weight matrices in two or more populations

Millsap and Meredith (1988) have developed a generalization of principal components analysis for the simultaneous analysis of a number of variables observed in several populations or on several occasions. The algorithm they provide has some disadvantages. The present paper offers two alternating least squares algorithms for their method, suitable for small and large data sets, respectively. Lower and upper bounds are given for the loss function to be minimized in the Millsap and Meredith method. These can serve to indicate whether or not a global optimum for the simultaneous components analysis problem has been attained.Financial support by the Netherlands organization for scientific research (NWO) is gratefully acknowledged.     

Least-squares approximation of an improper correlation matrix by a proper one

An algorithm is presented for the best least-squares fitting correlation matrix approximating a given missing value or improper correlation matrix. The proposed algorithm is based upon a solution for Mosier's oblique Procrustes rotation problem offered by ten Berge and Nevels. A necessary and sufficient condition is given for a solution to yield the unique global minimum of the least-squares function. Empirical verification of the condition indicates that the occurrence of non-optimal solutions with the proposed algorithm is very unlikely. A possible drawback of the optimal solution is that it is a singular matrix of necessity. In cases where singularity is undesirable, one may impose the additional nonsingularity constraint that the smallest eigenvalue of the solution be , where is an arbitrary small positive constant. Finally, it may be desirable to weight the squared errors of estimation differentially. A generalized solution is derived which satisfies the additional nonsingularity constraint and also allows for weighting. The generalized solution can readily be obtained from the standard unweighted singular solution by transforming the observed improper correlation matrix in a suitable way.