Robust estimation of ability in the Rasch model

Estimating ability parameters in latent trait models in general, and in the Rasch model in particular is almost always hampered by noise in the data. This noise can be caused by guessing, inattention to easy questions, and other factors which are unrelated to ability. In this study several alternative formulations which attempt to deal with these problems without a reparameterization are tested through a Monte Carlo simulation. It was found that although no one of the tested schemes is uniformly superior to all others, a modified jackknife stood out as the best one in general, it was also super efficient (more efficient than the asymptotically optimal estimator) for tests with forty or fewer items. It is proposed that this sort of jackknifing scheme for estimating ability be considered for practical work.This research was funded through a grant from the Law Enforcement Assistance Administration (78-NI-AX-0047) to the Bureau of Social Science Research, Howard Wainer, Principal Investigator. We would like to thank Ronald Mead, Anne Morgan and James Ramsay for kind, generous, and invaluable help at various stages of the project.     

The Effects of Measurement Strategy on Attributions for Marital Problems and Behaviors1

The present study examined the relation of attributions for spouse behaviors, attributions for global conjugal conflict, and marital adjustment. The sample consisted of 74 French-Canadian couples who completed the Marital Attribution Style Questionnaire, the Conflict Rating Scale, and the Dyadic Adjustment Scale. Results showed that the relation between attribution and marital satisfaction is stable across culture. In addition, specific and general attribution measures were low to moderately correlated. However, multiple regression analyses demonstrated that attributions for global conjugal conflict entered the regression equations more often than attributions for hypothetical spouse behaviors. These findings suggest that the comparability of marital attribution measures should not be taken for granted. The need to develop standardized measures is underlined.     

Growth coping,work satisfaction and turnover: A longitudinal study

A longitudinal organizational field study examined work satisfaction and employee growth coping as joint predictors of turnover. Three employee categories were examined: 1) No Turnover, 2) turnover within the same occupational field (Intraoccupational Turnover), and 3) turnover to a new occupational field (Interoccupational Turnover). Work satisfaction was moderate in the first two groups and low in the third group. As predicted, a specific strategy of employee coping, growth, was lowest in the first group, moderate in the second group, and highest in the third group. Through the use of linear discriminant analysis, 66.4% of the employees were correctly classified into No Turnover, Intraoccupational Turnover, and Interoccupational Turnover groups based only on information regarding work satisfaction and growth coping. The first squared canonical correlation was found to equal .30, far surpassing the modest work satisfaction/turnover correlations found in all previous studies.The authors wish to thank the two anonymous reviewers for their thoughtful comments.     

Acceptance patterns of retaliatory aggression among adjudicated youth by sex and race

This study investigates sex and race differences in normative beliefs about the acceptability of aggression across dimensions of severity of provocation (weak versus strong) and sex of provoker (male versus female). Students in the sample of 311 included those required to attend three Department of Youth Services schools after placement by the juvenile court system from across the state of Alabama (N=392). Results show that males were significantly more likely than females to approve of retaliation to weak provocation and against females. There were no significant differences between males and females for retaliation with strong provocation and against males. Results also show the Black group was significantly more likely than the White group to approve of retaliation to weak provocation, strong provocation, and against females.     

The number line effect reflects top-down control

Recent evidence indicates that central directional stimuli, such as eyes and arrows, trigger rapid, reflexive shifts of spatial attention. A study by Fischer, Castel, Dodd, and Pratt (2003) suggested that a similar effect might also apply to central numbers, as if a digit's meaning causes attention to be oriented to its relative position on a left-to-right mental number line. However, unlike central eyes and arrows, the orienting effect for central digits emerges slowly, suggesting that top-down endogenous processes may be mediating this effect. Here, we report a series of three experiments that strongly support this hypothesis. Experiment 1 replicated Fischer et al.'s left-to-right number line effect. Experiment 2 showed that this effect could be completely reversed by merely asking participants to imagine a number line running from right to left. Experiment 3 showed that a left-to-right number line effect could be abolished by presenting targets above and below central fixation, as well as to the left and right of center. Experiment 3 also showed that other mental sets, such as imagining a clock, result in attention's being oriented in accordance with where the central digits are represented on a clock face. Together, these data indicate that the spatial representations and attentional orienting related to the perception of digits are both fragile and flexible and depend critically on the top-down spatial mental sets adopted by individuals.     

Will both spouses have careers?: Predictors of preferred traditional or egalitarian marriages among university students

This study examined the significance of work preference variables, family preference variables, socialization influences, and needs for achievement and affiliation in predicting preferred traditional versus egalitarian marriages among university students. Results showed that, overall, male university students preferred traditional marriages while female university students preferred egalitarian marriages. Sex also interacted significantly with the following variables: desired family task involvement, desired job task involvement, and mother's family involvement. The data suggest that a perspective encompassing a wide variety of variables (including work, family, and socialization considerations) may be useful for conceptualizing individuals' preferences for traditional or egalitarian marriages. Implications of the results for management and career counseling are discussed.     

Online counselling: learning from writing therapy

Online counselling is new and still contentious, with little UK-based research available to practitioners, providers or potential clients. This article aims to extend an earlier review of some of the research into writing therapy \[Wright &; Chung (2001) Mastery or mystery? Therapeutic writing: a review of the literature, British Journal of Guidance and Counselling, 29(3), 277-291] and to indicate how it could be applied to online counselling. It also refers to some of the literature on online counselling, which, together with the writing therapy research, informed the decision to offer an online service to staff in a university setting.     

Inhibiting children's memory of an interactive event: the effectiveness of a cover‐up

Children are generally more susceptible than adults to suggestive interview techniques. Children's memories of an event can be altered and added to by presenting post‐event information (PEI). What is not known is whether embedding silence about a particular scene within the PEI makes that scene less likely to be reported. Children aged 5–6 years made cakes with an agent ‘Mrs Flour’. The following day they received PEI in which a target scene from the original event was omitted, resulting in children reporting the target scene significantly less often than did controls (control= 57% and omit= 23% correct responses). There was direct evidence from the children's language that the omission led to a detriment in memory for the original scene itself. Allowing children to draw during the interview did not reduce the effect. Implications are discussed in terms of child victims and witnesses particularly regarding child sexual abuse. Copyright © 2002 John Wiley & Sons, Ltd.