Traditional and Rank-Based Tests for Ordered Alternatives in a Cluster Correlated Model

Psychometrika - Methods for the analysis of one-factor randomized groups designs with ordered treatments are well established, but they do not apply in the case of more complex experiments. This...     

BEWILDERMENT AND THEREAFTER: Some Reflections in Response to Lee Yearley

The following reflections were originally an oral response to issues raised in Lee Yearley's presentation in May 2009 at Harvard Divinity School. As written here, they follow upon his oral and now written comments, highlighting key issues and points for development, drawing on this respondent's expertise in comparative and Hindu studies.     

Measurement Equivalence of 360°-Assessment Data: Are different raters rating the same constructs?

This study uses data collected using a 360°-assessment instrument to investigate the structural equivalence of 360°-assessment ratings, according to rater type, controlling for organizational level. Data from 533 managers and their raters were employed in the study, which used multi-group structural equations modeling. Issues central to the implementation and use of 360°-assessment data are also considered within the context of current research and practice.     

Adult-child interaction during invasive medical procedures.

Adult-child interactions during stressful medical procedures were investigated in 43 pediatric patients videotaped during a venipuncture procedure in the course of cancer treatment. Relations among six adult behavior categories (explain, distract, command to engage in coping behavior, give control to the child, praise, and criticize/threat/bargain) and three child behavior categories (momentary distress, cry/scream, and cope) were examined using correlational and sequential analysis. Results indicated that adult distraction resulted in increased child coping and reduced momentary distress and crying. Adult explanations, although a likely response to child distress and crying, did not result in a reduction of these behaviors. Attempts to give the child control reduced child crying. Implications for clinical interventions during painful medical procedures are discussed.     

Sex-role change, anomie and female suicide: a test of alternative Durkheimian explanations.

Durkheim's influential book, Suicide, provides general (economic) anomie, conjugal anomie, and sex-role convergence explanations of changes in suicide rates under conditions of social change. We used trend analyses of American suicide rates and female/male suicide ratios from 1950 to 1984 and the regression of the ratios on female educational attainment, white female labor force participation, white fertility rates, and divorce rates to examine these explanations. The general anomie explanation of female suicide trends is supported for the middle-aged (30 to 54 years) but not for the young (10 to 30 years) or the elderly (55 to 74 years). The conjugal anomie proposition is at best supported for age groups between 15 and 44 when general anomie is not pronounced. The role convergence explanation is rejected for all 13 5-year-age-groups. General anomie may not be a viable explanation of suicide trends for groups actively supporting relevant social changes or not yet tradition-bound or for groups whose retirement status reduces the importance of some social changes.     

Response error and processing biases in confidence judgment

Previous studies showed that random error can explain overconfidence effects typically observed in the literature. One of these studies concluded that, after accounting for random error effects in the data, there is little support for cognitive‐processing biases in confidence elicitation. In this paper, we investigate more closely the random error explanation for overconfidence. We generated data from four models of confidence and then estimated the magnitude of random error in the data. Our results show that, in addition to the true magnitude of random error specified in the simulations, the error estimates are influenced by important cognitive‐processing biases in the confidence elicitation process. We found that random error in the response process can account for the degree of overconfidence found in calibration studies, even when that overconfidence is actually caused by other factors. Thus, the error models say little about whether cognitive biases are present in the confidence elicitation process. Copyright © 2008 John Wiley & Sons, Ltd.