Constrained canonical correlation

This paper explores some of the problems associated with traditional canonical correlation. A response surface methodology is developed to examine the stability of the derived linear functions, where one wishes to investigate how much the coefficients can change and still be in an -neighborhood of the globally optimum canonical correlation value. In addition, a discrete (or constrained) canonical correlation method is formulated where the derived coefficients of these linear functions are constrained to be in some small set, e.g., {1, 0, –1}, to aid in the interpretation of the results. An example concerning the psychographic responses of Wharton MBA students of the University of Pennsylvania regarding driving preferences and life-style considerations is provided.Wayne S. DeSarbo, Robert Jausman, Shen Lin, and Wesley Thompson are all Members of Technical Staff at Bell Laboratories. We wish to express our gratitude to the editor and reviewers of this paper for their insightful remarks.     

Since the eruption of Mt. St. Helens,has anyone beaten you up? Improving the accuracy of retrospective reports with landmarkevents

When people are questioned about past experiences, it is common for eventsto be reported as happening more recently than they actually did. One technique for reducing the instances of this “forward telescoping” is tested in the current research: It involves the provision of landmark events that can clearly mark the beginning of the reference period. Such atechnique was shown to reduce the telescoping problem in five experiments, involving 1, 694 subjects. The landmark event used in two of the studies was the first major eruption of Mt. St. Helens, and use of this event was shown to significantly reduce the incidence of forward telescoping of crime victimizations. Asking subjects to provide their own personal landmarks had similar beneficial effects. Finally, a more usual public landmark event, New Year’s day, substantially reduced forward telescoping. At least part, but not all, of the benefit of landmark events appears to be due to their being dated rather precisely.     

Genetic mediation of sex differences in horizontality: A rejoinder

Neither M. G. McGee (Developmental Review, 1981, 1, 289–295) nor M. J. Allen, M. A. Wittig, and K. Butler (Developmental Review, 1981, 1, 284–288) suggest any alternative explanation for our finding that water-level performance appears to have an X-linked genetic basis. The power calculations of Allen et al. are found to be faulty, and McGee confuses the hypothesis we tested with a weaker hypothesis. Although the X-linked genetic model is not an adequate model of water-level performance the water-level data fit the X-linked model far better than color blindness and HCN data McGee presents as exemplars of X-linked characteristics.     

A Longitudinal Assessment of Teacher Perceptions of Parent Involvement in Children's Education and School Performance

This study examines the ways in which parental involvement in children's education changes over time and how it relates to children's social and academic functioning in school. Teachers provided information on parent involvement and school performance for 1,205 urban, kindergarten through third-grade children for 3 consecutive years. They rated the following four dimensions of parent involvement: frequency of parent-teacher contact, quality of the parent-teacher interactions, participation in educational activities at home, and participation in school activites. As predicted, the frequency of parent-teacher contacts, quality of parent-teacher interactions, and parent participation at school declined from Years 1 to 3. Every parent involvement variable correlated moderately with school performance and parent involvement in Years 1 and 2, and accounted for a small, but significant amount of variance in Year 3 performance after controlling for initial performance level. Participation in educational activities at home predicted the widest range of performance variables. Results suggest that enhancing parental involvement in children's schooling relates to improvements in school functioning.     

Boys will be boys (and girls will be girls): The attribution of gender role stereotypes in a gaming situation

This study tested one hypothesis concerning the attribution of gender role stereotypes about competitive behavior and three hypotheses concerning differences in attribution of sex between male and female subjects. The study used a Prisoner's Dilemma Game setting to expose subjects to one of three conditions (competitive, cooperative, or tit-for-tat) to measure attribution of sex to an unknown confederate. A chi-square analysis revealed significant differences in the attribution of sex to the anonymous confederate between the competitive and the combined cooperative and tit-for-tat groups. In the competitive condition, subjects were more prone to think that the anonymous confederate was male than were subjects in the cooperative and tit-for-tat condition. This finding is consistent with the gender role stereotype that generally associates competitive behavior with masculinity and not with femininity. Post hoc chi-squares also revealed no difference between male and female subjects in the attribution of sex in any of the three conditions. Implications of these findings are discussed.     

PROBLEMS IN ESTABLISHING NORMS FOR CROSS-CULTURAL COMPARISONS

Problèmes de l'établissement des normes dans les comparisons interculturelles. — En examinant plusieurs des méthodes employées pour comparer les attitudes et les pratiques de parents allemands et américains relativement à l'éducation des enfants, nous avons abouti aux conclusions suivantes, susceptibles de s'appliquer à toutes les études comparatives interculturelles. 1. Les observations rollectées dans une culture et réduites à des moyennes ne peuvent servir à en évaluer une autre. — 2. Il faudrait étudier des problèmes spécifiques plutôt que des problèmes généraux ou globaux; on peut s'attendre à trouver des configurations complexes dans les différences observées. — 3. Les variables susceptibles ou non de modification selon l'âge de l'enfant, augmentent la complexité des comparaisons entre cultures. — 4. Une norme basée sur une culture n'est probablement pas valable pour une autre. — 5. Les méthodes basées sur l'utilisation d'observateurs bi-culturels semblent réduire au minimum la possibilité d'erreurs systéamatiques et la dépendance vis-à-vis de normes non-équivalentes. — 6. Les questions et les résponses de type opératoire et comportemental augmenteront, mais ne garantiront pas, la précision et la validité de l'instrument employé. — 7. Les questions ouvertes ne peuvent permettre de différencier les éléments non importants, très évidents ou uniformes des comportements ou des attitudes. — 8. Actuellement, on ne peut que rarement détecter les causes du comportement. Il n'est peut-être pas possible de les mettre en évidence, en dehors des cas où plusieurs méthodes différentes et différentes sortes de questions donnent des résultats cohérents. — 9. Il est nécessaire d'isoler un très grand nombre de variables dans chaque culture avant de pouvoir faire des comparaisons interculturelles satisfaisantes. l'utilisation de moyens électroniques pourrait faciliter le travail.     

Abacs for determination of a correlation coefficient corrected for restriction of range

Abacs approximating the product-moment correlation for both explicit and implicit selection are presented. These abacs give accuracy to within .01 of the corresponding analytic estimate.     

Murphy on Postmodernity, Science, and Religion

Nancey Murphy claims that a shift in "thinking strategy" from modern to postmodern modes of thought makes it easier to exhibit the intellectual respectability of theology vis-à-vis the sciences. Her case for this proposition depends on modernist interests, most notably in systematizing the sciences for reasons that have their origin in Plato's divided line.     

Narrative aspects of a doctor-patient encounter

In this essay I want to examine the ways that doctors communicate with their patients, in particular the stories that doctors tell their patients. I contend that the kinds of stories that physicians tell patients have a direct bearing on the nature of their relationship as a whole. Specifically, I shall argue that physicians who tell their patients the most complete stories possible about their illnesses, acknowledging uncertainties wherever they exist (to the extent that this is possible), show the greatest amount of respect for their patients as persons. This quest for completeness must be supplemented by an attempt on the part of the physician to explore the patient's understanding and beliefs so as to formulate not only a more complete narrative, but also to formulate one that better serves the needs of the patient.     

IF OUR GENES ARE FOR US,WHO CAN BE AGAINST US? THOUGHTS OF A PRAGMATIST ON SCIENCE AND MORALITY

Abstract. The philosopher Michael Ruse accounts for the difference between hypothetical and categorical imperatives, and thus the origin of distinctively moral obligations like that of altruism, in genetic terms. This is part of an attempt to develop a philosophy that takes Darwin seriously by substituting respectable scientific entities, specifically those of evolutionary biology, for suspect theological or philosophical ones, like God or the transcendental ego, as a basis for addressing philosophical questions. Pragmatists take Darwin seriously, but in a very different way from that proposed by Ruse. Darwin introduced a “logic” into the study of living things—including human beings, the human mind, and culture— that leads philosophers to ask new and different questions about morality rather than trying to supply new answers to the same old questions. This essay contrasts these two different ways of taking Darwin seriously for purposes of philosophy and claims certain advantages for the pragmatist way over Ruse's.