The accurate measurement of examinee test performance is critical to educational decision-making, and inaccurate measurement can lead to negative consequences for examinees. Person-fit statistics are important in a psychometric analysis for detecting examinees with aberrant response patterns that lead to inaccurate measurement. Unfortunately, although a large number of person-fit statistics is available, there is little consensus as to which ones are most useful. The purpose of this study was to compare 36 person-fit indices, under different testing conditions, to obtain a better consensus as to their relative merits. The results of these comparisons, and their implications, are discussed.Sound decisions in educational settings hinge largely on accurate measurement of student characteristics. Such measurements can help identify those individuals who are qualified enough to enter a particular school, or receive a particular educational degree. Also, these measurements can be used to monitor students' learning progress. This may, for example, enable educators to productively tailor their curriculum, or help policy makers decide on important educational issues.In contrast, the inaccurate measurement of test performance can lead to negative consequences. On the one hand, spuriously high test scores can lead to unqualified individuals being enrolled into an educational program (e.g., undergraduate, graduate, or professional), or being awarded an educational degree. On the other hand, qualified individuals with spuriously low test scores may be unfairly excluded from academic programs, or unfairly denied a degree. Furthermore, the inaccurate measurement of test performance undermines the assessment of students' learning progress, and curriculum planning efforts.Requests for reprints should be sent to George Karabatsos, College

Based on scientific literature and interviews with clinicians and patients, we developed a quality of life instrument for use with people with MS called the Functional Assessment of Multiple Sclerosis (FAMS). The initial item pool consisted of 88 questions: 28 from the general version of the Functional Assessment of Cancer Therapy quality of life instrument, plus 60 generated by patients, providers, and literature review. The validation samples comprised a mail survey cohort (N = 377) and a clinical cohort (N = 56). Both cohorts provides evidence for internal consistency of the derived subscales, test-retest reliability, content validity, concurrent validity, and construct validity. Principal components and Rasch measurement model analyses were applied sequentially to survey sample data, reducing test length to 44 questions, divided into six subscales: mobility, symptoms, emotional well-being (depression), general contentment, thinking/fatigue, and family/social well-being. Fifteen initially rejected questions were added back as miscellaneous (unscored) questions for their potential clinical and empirical value. The mobility subscale was strongly predictive of the Kurtzke Extended Disability Status Scale and the Scripps Neurologic Rating Scales. The other five subscales were not, indicating they measure aspects of patient quality of life not captured by the neurologic exam. The final 59-item English language instrument (FAMS version 2) is available for inclusion in clinical trials and clinical practice.

Alcohol is associated with risk of sexual assault among women and with increased risk of experiencing completed rape once attacked. In particular, alcohol use prior to sexual assault by both offenders and victims may affect the severity of sexual victimization experienced by women. Little research has explored the mechanisms (e.g., social context, behavior) through which alcohol may affect outcomes of sexual attacks using multivariate analysis. This study analyzed the role of alcohol in sexual assaults experienced by a national sample of female college students. A hierarchical multivariate regression showed that victim alcohol abuse propensity and both victim and offender alcohol use prior to attack were directly associated with more severe sexual victimization to women as measured by the Sexual Experiences Survey. This study suggests that alcohol use plays both direct and indirect roles in the outcomes of sexual assaults. Rape and alcohol abuse prevention efforts can benefit from incorporating information about alcohol's role in different assault contexts.

Alcohol use prior to sexual assault by both offenders and their victims may be associated with the severity of sexual aggression men commit against women. Little research has explored the pathways (e.g., social context, behavior) through which alcohol may affect outcomes of sexual attacks. The present study analyzed the role of alcohol in sexual assaults ( N = 694) committed by men identified from a national sample ( N = 2,972) of male college students completing a survey. Interactions of alcohol use with assault variables did not suggest any synergistic role of alcohol use in predicting sexual aggression severity. Path analysis showed, however, that offender propensity to abuse alcohol and victim preassault alcohol use were each both directly and indirectly related to sexual aggression severity, whereas offender preassault alcohol use was not directly related to sexual aggression severity. This study suggests that alcohol use plays both direct and indirect roles in the outcomes of sexual assaults. Rape and alcohol abuse prevention efforts can benefit from incorporating information about alcohol's role in different assault contexts.

consensus theory, Bayesian inference, Markov Chain Monte Carlo, posterior predictive model evaluation, Bayesian model selection,

No abstract

No abstract

This study introduces an order-constrained Bayes inference framework useful for analyzing data containing dichotomous-scored item responses, under the assumptions of either the monotone homogeneity model or the double monotonicity model of nonparametric item response theory (NIRT). The framework involves the implementation of Gibbs sampling to estimate order-constrained parameters, followed by inference with the posterior-predictive distribution to test the monotonicity, invariant item ordering, and local independence assumptions of NIRT. The Bayes framework is demonstrated through the analysis of real test data, and possible extensions of it are discussed.