Determining the power of multiple regression analyses both with and without repeated measures

Power analysis guides researchers in planning how much data to collect. This article describes BWPower, a computer program for the Windows 95 environment that performs power analyses for research designs that may or may not include both between- and within-subjects factors. We discuss how BWPower easily accommodates both between- and within-subjects factors and provide examples of BWPower’s use in performing power analyses on designs with only between-subjects factors, designs with only repeated measures, and with mixed between- and within-subjects designs. We highlight the major features of BWPower’s user interface, such as the ability to iteratively increment or decrement the number of subjects and the automatic recalculation of power when the number of subjects or effect sizes is changed.     

Placement Stability Following Short-Term Residential Care

Youth discharged from a short-term residential facility and reunified with their parents were contacted at least one year after discharge to determine whether or not they had a disruption in their placement since reunification. Follow-up intervals of 149 youth contacted (82% of the eligible sample) ranged from a minimum of 12 months to a maximum of 30 months post-discharge. Differences in the time to placement disruption were analyzed using survival analysis techniques. After controlling for adjustment differences during the treatment program, youth behavior problems, and family problems, a significant interaction of youth age and treatment intensity was found: Compared to the others, younger youth who received additional treatment components had placement disruptions later and less frequently.     

Gender-Related Knowledge Variations Within Geography

In a survey of students from junior high through the undergraduate college level, males consistently out-performed females on a test covering map skills and physical, human, and regional aspects of geography. The gender gap was consistent across all four subfields. It widened with increased education and was significant even after accounting for numerous personal information items that might have explained the difference. This consistent performance difference involving the broad spectrum of geographic knowledge argues against explaining the gender gap as symptomatic of differences in a single underlying factor such as map skills or spatial abilities. Knowledge-based and nonknowledge-based explanations for the gender difference are discussed.     

The Future of Psychology in Pain Management

Pain treatment facilities have proliferated in recent years and psychology has enjoyed considerable success in the clinical and research arenas of pain management. However, changes in the health care environment present significant challenges to the future of psychology in pain management. This article discusses concerns in three important areas of pain treatment that psychologists must address if they are to maintain a strong presence in this field: (1) evaluating and treating patients in the pain treatment setting, (2) evaluating and enhancing the efficacy of pain treatment, and (3) applying pain treatment services to other patient populations. With appropriate attention to these issues, psychology can continue to thrive in pain treatment settings.     

Descartes and the tree of knowledge

Descartes' image of the tree of knowledge from the preface to the French edition of the Principles of Philosophy is usually taken to represent Descartes' break with the past and with the fragmentation of knowledge of the schools. But if Descartes' tree of knowledge is analyzed in its proper context, another interpretation emerges. A series of contrasts with other classifications of knowledge from the seventeenth and eighteenth centuries raises some puzzles: claims of originality and radical break from the past do not seem warranted. Further contrasts with Descartes' unpublished writings and with school doctrines lead to the ironic conclusion that, in the famous passage, Descartes is attempting to appeal to conventional wisdom and trying to avoid sounding novel.     

On the misuse of manifest variables in the detection of measurement bias

Measurement invariance (lack of bias) of a manifest variableY with respect to a latent variableW is defined as invariance of the conditional distribution ofY givenW over selected subpopulations. Invariance is commonly assessed by studying subpopulation differences in the conditional distribution ofY given a manifest variableZ, chosen to substitute forW. A unified treatment of conditions that may allow the detection of measurement bias using statistical procedures involving only observed or manifest variables is presented. Theorems are provided that give conditions for measurement invariance, and for invariance of the conditional distribution ofY givenZ. Additional theorems and examples explore the Bayes sufficiency ofZ, stochastic ordering inW, local independence ofY andZ, exponential families, and the reliability ofZ. It is shown that when Bayes sufficiency ofZ fails, the two forms of invariance will often not be equivalent in practice. Bayes sufficiency holds under Rasch model assumptions, and in long tests under certain conditions. It is concluded that bias detection procedures that rely strictly on observed variables are not in general diagnostic of measurement bias, or the lack of bias.Preparation of this article was supported in part by PSC-CUNY grant #661282 to Roger E. Millsap.     

Attribution in conversational context: Effect of mutual knowledge on explanation-giving

Attribution theorists typically have conceived the attribution process in terms of universal laws of cognitive functioning, independent of social interaction. In this paper we argue for the notion, grounded in recent ordinary language philosophy, that any consideration of the form of everyday explanation must take into account its function as an answer to a ‘why’ question within a conversational framework. Experiment 1 provides support for the idea that speakers should identify as causally relevant that necessary condition for the occurrence of an event about which the enquirer is ignorant. Experiment 2 replicates this basic finding and further demonstrates that speakers will change their explanations to enquirers believed to be sharing different knowledge about the same target event. Experiment 2 also assessed the role of individual differences in conversational rule-following, and found in apparent contrast some previous predictions that high self-monitoring individuals were no more likely than lows to tailor their explanations to suit the enquirer's knowledge state. If anything, the reverse occurred. Taken together, these experiments support the central contention of the abnormal conditions focus model (Hilton and Slugoski, 1986), that the common sense criterion of causality is that of an ‘abnormal condition’ rather than constant conjunction as instantiated in the ANOVA model of causal attribution (Kelley, 1967, 1973).     

Body dissatisfaction,eating disorders,and depression: A developmental perspective

McCarthy (1990) contends that a cultural ideal of thinness (the thin ideal) causes depression and eating disorders to occur more frequently in women than men. She believes eating disorders are a way of coping with depression and hypothesizes that the thin ideal has its greatest impact during puberty when sexual attractiveness becomes important and changes in physical appearance increase the discrepancy between the adolescent girl's actual and ideal body image. The present study used a cross-sectional design of 5th, 8th, and 12th grade males and females (N = 599) to test several predictions from McCarthy's model. As expected, 8th and 12th grade females preferred a thinner than average body shape. They were more dissatisfied with their bodies than males and scored higher on measures of disordered eating. In contrast, before puberty (5th grade) boys were more depressed than their female peers but this pattern was reversed in 8th grade boys and girls. Paralleling changes in depression, disordered eating scores were highest in 8th and 12th grade girls. Both of these findings underscore the role of maturational factors in the onset of eating disorders. Although the sample was ethnically diverse and mainly from middle and low socioeconomic groups, there was a surprisingly high incidence of body dissatisfaction and symptoms of disordered eating. This illustrates the pervasiveness of thinness as a standard of feminine beauty in our society and the damaging effects it can have.     

EFFECTS OF SELF-MANAGEMENT TRAINING AND REINFORCEMENT ON THE TRANSFER OF IMPROVED CONDUCT IN THE ABSENCE OF SUPERVISION

The instruction, maintenance, and transfer of training of social skills of 3 seriously emotionally disturbed adolescents were accomplished by a self-management training and reinforcement package. During baseline sessions these students, who were covertly filmed in their classroom, averaged over 90% off-task or socially inappropriate behavior while their teacher was out of the room. They showed similar behaviors when walking between classes, unattended by their teacher. Treatment was introduced in the classroom and consisted of social skills and self-management training and reinforcement. Treatment procedures included instruction, modeling, and role playing of social skills, as well as self-assessment, self-recording, and self-reinforcement for correct approximations of these skills. After 5 weeks of training, all subjects demonstrated substantial improvements in the classroom during the teacher's absence and when distracted by other students; however, transfer of social skills did not occur to the between-class setting until students were given explicit instruction to initiate self-managing procedures in this setting.