A cautionary note on using internal cross validation to select the number of clusters

A highly popular method for examining the stability of a data clustering is to split the data into two parts, cluster the observations in Part A, assign the objects in Part B to their nearest centroid in Part A, and then independently cluster the Part B objects. One then examines how close the two partitions are (say, by the Rand measure). Another proposal is to split the data into k parts, and see how their centroids cluster. By means of synthetic data analyses, we demonstrate that these approaches fail to identify the appropriate number of clusters, particularly as sample size becomes large and the variables exhibit higher correlations.The authors express their thanks to the Sol C. Snider Entrepreneurial Center, Wharton School, for support of this project.     

A time sharing cross-sectional study of monolinguals and bilinguals at different levels of second language acquisition

Several methodological factors associated with the concurrent activities (finger-tapping) paradigm were considered in a cross-sectional study investigating cerebral patterns of asymmetry in three groups of English-speaking non-Hispanic dextral males at three levels of second language (Spanish) acquisition and one control group of monolinguals. Results revealed the fluent bilinguals to be bilateral and significantly different from other groups for native language tasks in English. Moreover, a priori contrasts indicate that greater right- than left-hand disruption in concurrent tapping may be typical of monolinguals, but can be influenced by other factors. Monolingual reliability test-retest correlations were .77 and .47.     

Response to Michael Sells

In an era when lies and misrepresentations about historical events easily become firmly rooted, Michael Sells's discussion illustrates the importance of careful historical research as a moral enterprise. In addition to the skills of the historian, however, there is also room in this enterprise for those of the ethicist. In particular, I warn against confusing the truth or falsity of claims about one narrow historical period with larger questions about the moral meaning and significance of those claims. Illustrating this, I argue one cannot assess the legitimacy of competing nationhood claims solely on the basis of the deeds of specific actors. Nor should the actions of a single individual like the Grand Mufti of Jerusalem be converted into a totalizing claim about the rights of the Palestinian people.     

THE PRESIDENT'S COUNCIL ON BIOETHICS—REQUIESCAT IN PACE

In mid‐June 2009, the Obama administration dissolved the President's Council on Bioethics (PCBE), a group established by President George W. Bush in August 2001 and whose nearly eight‐year life was marked from beginning to end by controversy. While some will regret the PCBE's passing, others will regard the Council as a failed experiment in doing public bioethics.     

Body weight and body image among college women: perception, cognition, and affect

This study examined the relations between body weight and multiple parameters of body image in a nonclinical sample. Thirty-six female undergraduate students with stable body weights served as subjects; 12 were underweight, 12 were normal weight, and 12 were overweight, as determined using recent normative tables. Each subject responded to a general weight information questionnaire and to the Body-Self Relations Questionnaire, a standardized instrument assessing affective, cognitive, and behavioral components of appearance-related body image. A newly developed apparatus and procedure for estimating body size, the Body Image Detection Device, was utilized for perceptual and cognitive measures of body image. Each subject estimated and also subjectively appraised the width of her own body at five points and that of a realistic female mannequin. A caliper was used to record actual widths, and a standard laboratory scale and rule measured weight and height. The general finding of the study was that the perceptual, affective, and cognitive components of body image differed as a function of body weight, but the nature of the differences varied with the measure employed. The pattern of results for the perceptual measure suggest a cautious view of its reliability and validity. The multidimensional approach of our study, in marked contrast to earlier fractionated studies, offers a more integrated perspective on body image and provides new directions for future research.     

The nonequivalence of behavioral and mathematical equivalence.

Sidman and his colleagues derived behavioral tests for stimulus equivalence from the axiom in logic and mathematics that defines a relation of equivalence. The analogy has generated abundant research in which match-to-sample methods have been used almost exclusively to study interesting and complex stimulus control phenomena. It has also stimulated considerable discussion regarding interpretation of the analogy and speculation as to its validity and generality. This article reexamines the Sidman stimulus equivalence analogy in the context of a broader consideration of the mathematical axiom than was included in the original presentation of the analogy and some of the data that have accumulated in the interim. We propose that (a) mathematical and behavioral examples of equivalence relations differ substantially, (b) terminology is being used in ways that can lead to erroneous conclusions about the nature of the stimulus control that develops in stimulus equivalence experiments, and (c) complete analyses of equivalence and other types of stimulus-stimulus relations require more than a simple invocation of the analogy. Implications of our analysis for resolving current issues and prompting new research are discussed.     

Predictors of adolescent female decision making regarding contraceptive usage.

The relationship of cognitive capacity, cognitive egocentrism, and experience factors to decision making in a contraceptive usage problem was examined. Fifty sexually active, unmarried females, ages 14-19, served as subjects. Using correlational, regression, and canonical correlational analyses, cognitive capacity and cognitive egocentrism variables, not experience with contraceptives, were found to be significantly related to, and predictive of, five of seven decision-making variables. Forty-one percent of the variance was accounted for in predicting the canonical decision-making variable. The implications of these results for future research are discussed.     

Clarification of the dual task dilemma: Lateralized effects for perfunctory and purposeful tasks in left- and right-handed males

The hypothesis that behavioral asymmetries with the dual task paradigm represent manual dominance was investigated with right- and left-handed males performing verbal and spatial tasks ordered by complexity. Lateralization was assessed for nonideational (perfunctory) and ideational (purposeful) components of tasks with multivariate and ANCOVA procedures. The outcomes of prerequisite tests showed the assumptions for conducting ANCOVA procedures were not satisfied with different handedness groups in the same design. However, results of the multivariate analyses suggest lateralized effects are more likely to represent the cognitive task when interference is high and may represent manual dominance when interference is low.     

DEFINING,VALIDATING, AND INCREASING INDICES OF HAPPINESS AMONG PEOPLE WITH PROFOUND MULTIPLE DISABILITIES

In this study we attempted to operationalize, measure, and increase happiness among people with profound disabilities. Happiness indices were defined and observed among 5 individuals. Validation measures indicated that (a) increases in happiness indices were observed when individuals were presented with most preferred stimuli relative to least preferred stimuli, (b) increases in unhappiness indices were observed when they were presented with least preferred relative to most preferred stimuli, and (c) practitioner ratings of participant happiness coincided with observed indices. Subsequently, classroom staff increased happiness indices through presentation and contingent withdrawal of activities. Results suggested that a behavioral approach can enhance happiness as one aspect of quality of life among people with profound disabilities. Research directions are offered that focus on using a behavioral approach to investigate other private events that are important among people with disabilities.