Some properties of two measures of multivariate association

Two kinds of measures of multivariate association, based on Wilks' and the Bartlett-Nanda-Pillai trace criterionV, respectively, are compared in terms of properties of the univariateR ² which they generalize. A unified set of derivations of the properties is provided which are self-contained and not restricted to decompositions in canonical variates. One conclusion is that asymmetric index based on allows generalization of the multiplicative decomposition ofR ² in terms of squared partial correlations, but not the additive decomposition in terms of squared semipartial correlations, while the reverse is true for anasymmetric index based onV.We are indebted to Jos M. F. ten Berge for some fruitful discussions.     

Reliability issues in the penile assessment of incarcerants

Although there has been increased reliance upon the direct physiological measurement of penile tumescence in the assessment of sexual deviations, the reliability of these measures has never been empirically validated. This investigation evaluates the reliability of the rape index as a technique in differentiating rapists from nonrapists. Eighteen individuals incarcerated on non-sexual-related offenses were presented visual and audio stimuli depicting mutually consenting intercourse, rape, and aggression. These stimuli were presented under two instructional sets: (1) enchance arousal and (2) suppress arousal. Reliability indices for the rape index were computed among various conditions of modality presentation, instructional set, and test-retest conditions. The overall reliability of the rape index was found to be low. The implications of these results upon the assessment and classification of offenders is discussed.     

Addressing Weight Suppression to Improve Treatment Outcome for Bulimia Nervosa

Cognitive behavioral therapy (CBT) is regarded as the gold-standard treatment for bulimia nervosa (BN), yet despite impressive empirical support for its effectiveness, over 50% of patients fail to achieve abstinence from binge eating and purging by the end of treatment. One factor that may contribute to reduced efficacy rates in CBT is weight suppression (WS; the difference between a person's highest weight ever at their adult height and current weight). A growing body of research indicates that WS in patients with BN may have a clinically significant effect on symptomatology and prognosis. However, the current cognitive behavioral framework for BN does not explicitly acknowledge the role of WS in the onset or maintenance of BN symptoms and does not provide guidance for clinicians on how to address WS during treatment. The relationship between WS, biological pressure to regain lost weight, and the maintenance of BN symptoms suggest that current cognitive behavioral models of BN may be improved by considering the role of WS and exploring needed treatment modifications. Indeed, a reconceptualization of existing models may offer insight into potential strategies that can be used to reduce the susceptibility to treatment dropout, nonresponse, and relapse. It is therefore necessary to consider whether, and how, clinicians' consideration of WS during case conceptualization and treatment planning could serve to improve CBT outcomes. The current review explores ways in which high WS could contribute to poor CBT outcomes, provides preliminary clinical recommendations for incorporating WS into existing cognitive behavioral treatments based on extant data and clinical wisdom, and proposes suggestions for future research needed in this domain.     

Cultural representations of thinness in women, redux: Playboy magazine's depiction of beauty from 1979 to 1999

Playboy's portrayal of the male ideal of feminine beauty, in terms of overall body size, percent normative weight, and waist-to-hip ratios (WHRs), was analyzed for the years 1979–1999. Trends were examined through body measurements obtained from Playboy centerfolds. Results reveal a continuation of the low body mass index (BMI) found in the Playmates by earlier studies; however, for the 21 years examined, the trend towards increasing thinness seems to have stabilized and may have begun to actually reverse. There was also an increase in the centerfolds’ WHRs over the 21-year time period. Overall, the results support the continued valuing in American society of a thin ideal for women; while the images of beauty have become somewhat heavier over the 21-year period reviewed, the Playmates remain markedly below weights normative for their age group.     

The swimsuit issue: Correlates of body image in a sample of 52,677 heterosexual adults

Past research on adults’ body image has typically used small convenience samples, limiting the ability to examine associations of personal characteristics to body satisfaction. This study of 52,677 heterosexual adults ages 18–65 examined associations of body satisfaction to age, height, gender, and body mass index (BMI). Age and height were mostly unrelated to body satisfaction. Consistent with an Objectification Theory perspective, fewer men than women reported being too heavy (41% versus 61%), rated their body as unattractive (11% versus 21%), or avoided wearing a swimsuit in public (16% versus 31%). Men felt better about their bodies than women across most of the weight span, although among underweight individuals, women felt better than men. Slender women (BMIs 14.5–22.49) were more satisfied than most other women (BMIs 22.5–40.5). Among men, underweight and obese men were least satisfied. These findings highlight gender differences in the association of weight to body satisfaction.     

Data, model, conclusion, doing it again

This paper explores the robustness of conclusions from a statistical model against variations in model choice (rather than variations in random sampling and random assignment to treatments, which are the usual variations covered by inferential statistics). After the problem formulation in section 1, section 2 presents an example from Box and Tiao in which variation in parent distribution is modeled for a one sample location problem. An adaptive Bayesian procedure permits to use a weighted mixture of parent distributions rather than choosing just one, such as a normal or a uniform distribution.In section 3 similar considerations are applied to an event history model for the influence of education and gender on age at first marriage, but here the conclusions are hardly influenced by the choice of the duration distribution. In section 4 a brief discussion of model choice in factor analysis and structural equation models is followed by a more elaborate treatment of the choice of integer valued slopes for item response functions in the OPLM model which is an extension of the Rasch model. A modest simulation study suggests that Adaptive Bayesian Modeling with a mixture of sets of slopes works better than fixing one set of postulated slopes.The paper concludes with some remarks on the roles and sources of prior distributions followed by a short epilogue which argues that simultaneous consideration of a class of models for the same data is sometimes superior to exclusively analyzing the data under one specific model chosen from such a class.This article is based on the Presidential Address Ivo W. Molenaar gave on June 20, 1998 at the 1998 Annual Meeting of the Psychometric Society held at the University of Illinois in Champaign, Illinois. Thanks is given to John Wiley & Sons and George C. Tiao for granting permission to reprint three figures from the book George C. Tiao wrote with George E. P. Box titled Bayesian Inference in Statistical Analysis.—EditorThanks are due to Anne Boomsma, Jeffrey Hoogland, Mark Huisman, Tom Snijders, Marijtje Van Duijn, and Norman Verhelst for suggesting improvements and/or assisting with data analyses.     

THEORETICAL,PRACTICAL, AND SOCIAL ISSUES IN BEHAVIORAL TREATMENTS OF OBESITY

Although showing superior maintenance, behavioral treatments of obesity typically produce small weight losses at a decelerating rate. Rather than reflecting poor compliance with treatment, these findings are consistent with known compensatory metabolic changes that operate to slow weight loss and promote regain. Other problems associated with dieting include failure of caloric regulation, hyper-responsivity to food palatability, and hunger, which is greatest under conditions of moderate restriction and unpredictability of access to food. The inevitability of treatment failure in many instances must be faced and efforts made to prevent further worsening of the obese patient's self-esteem. Prognosis and treatment planning may be aided by consideration of the historical difficulties of weight loss, the degree of hunger experienced on diets, which may reflect important physiological differences among individuals, and the use of food to optimize arousal level. Full involvement of the patient in setting goals and planning treatment is recommended.     

COMPARISON OF ESTABLISHED AND INNOVATIVE WEIGHT-REDUCTION TREATMENT PROCEDURES1

A central component of a commonly used weight-loss procedure developed by Stuart (1967), which emphasizes situational engineering (SE), was compared to eating behavior control (EBC), a treatment employing reinforcement of changes in eating habits. A third treatment encouraged development of individuals' own applications of a model of self-control and attempted to provide effective feedback regarding progress toward weight-loss goals. This self-initiated treatment (SI) was designed to be less directive than many behavioral treatment packages and to be more realistic about the expectations of subjects and the actual eating habits of overweight and normalweight people. Following baseline, treatment lasted six weeks and followup lasted 16 weeks. While all treatments produced weight losses, EBC surpassed SE during treatment. SI losses during treatment did not differ from SE or EBC, but SI was the only group that lost weight throughout followup, indicating the possible utility of more subtle, less directive interventions than those that have often been associated with behavioral approaches. The results of SE question the applicability to clinical procedures of the external cue hypersensitivity theory of obesity (Schachter and Rodin, 1974), to which SE is closely related.     

Culture and body image: Body perception and weight concern in young Asian and Caucasian British women

The comparatively lower prevalence of eating disorders among Asian and Afro-Caribbean than Caucasian women in the UK has often been attributed to cultural differences in pressures for slimness. However, there have been no attempts to evaluate cultural differences in ideals for female physique directly among women in the at risk age groups. In the present study, cultural influences on body image were evaluated by comparing the body size ideals, body image and dieting concerns in a sample of 274 young white and Asian British women. The results indicate that Asian women are less likely to describe themselves as too fat, were less dissatisfied with their body size, less likely to want to lose weight and less restrained. However, Asian women were slimmer than white women and, after controlling for the difference in body size, the effects were reduced. The hypothesis that any body satisfaction differences could be explained by differences in ideal body size between the two groups were not supported; Asian women favoured even slimmer ideal body sizes than white women. Alternative explanations in terms of cultural differences in evaluation of fatness are discussed.     

School performance and IQ-test scores at age 13 as related to birth weight and gestational age

The cohort in the present longitudinal research program consisted of 873 children in an entire school grade, in a Swedish community. The present results showed a main effect of birth weight; low birth weight (LBW) children had lower school performance and intelligence-test (IQ) scores at age 13 than did normal birth weight (NBW) children irrespective of parental SES. Second, there was no significant main effect of gestational age (GA) on scholastic performance and IQ-test scores. Third, there was a significant main effect of the combination of birth weight and GA on scholastic performance and IQ-test scores. The LBW children born at term (38-40 pregnancy weeks; pw) had significantly lower scores and school grades as compared to the control group while the LBW children born with short gestational age (34-37 pw) and with very short gestational age (less than 34 pw) had significantly lower scores and marks in fewer areas of academic attainment.