Gender Roles,Power Strategies,and Precautionary Sexual Self-Efficacy: Implications for Black and Latina Women's HIV/AIDS Protective Behaviors

This cross-sectional study tested a conceptual model of women's HIV/AIDS protective behaviors using gender roles, relationship power strategies, and precautionary sexual self-efficacy as predictors in a predominantly Black and Latina community sample of heterosexual women (N = 125). Results revealed no support for the full model, but partial confirmation for several components of the model. Education significantly predicted gender roles, and gender roles and use of direct power strategies were significant predictors of sexual self-efficacy. Most of the participants were married or partnered and were unconcerned about contracting HIV, suggesting that if women perceive that they are at low or no risk, their gender roles, power strategies, and precautionary sexual self-efficacy will be inconsequential to their HIV/AIDS risk reduction practices.     

An Alternate Method for Content-Oriented Test Construction: An Empirical Evaluation

This study empirically evaluates the relationships between rationally and empirically estimated item indices used in the development of a test of land navigation knowledge for the US Marine Corps. Three land navigation instructors provided ratings on three item indices (item content validity ratios [CVRs], difficulty, and discrimination). These ratings were correlated with empirically derived item indices (difficulty, discrimination, and item-criterion correlations) obtained from the scores of 359 Marines on measures of land navigation knowledge, skill, and performance. Contrary to previous research, item CVRs were significantly correlated with all three empirically determined item indices. The increased effectiveness of rational item indices obtained in the present study over those from previous research appears to be related to the following areas: qualifications of expert raters, match of raters to the rating task, and the nature of the rating task. The implications for content-oriented test design and for judgmental test validation strategies are discussed.     

Predicting memory completeness and accuracy: emotion and exposure in repeated autobiographical recall

Using a repeated testing procedure modified for an autobiographical event, individuals provided narrative accounts of their memory for the announcement of the verdict in a widely publicized murder trial. The goal of the study was to determine whether different factors would predict initial versus final recall performance. Ratings of negative emotional reaction during the event predicted initial amount recalled, and frequency of exposure to the event predicted final amount recalled. Judged completeness of recall and phenomenological ratings of the vividness of the memory were also related to final amount recalled. The measured variables were unrelated to recall accuracy. Examining memory over repeated recall attempts offers a fruitful way to examine processes in autobiographical memory. Copyright © 2001 John Wiley & Sons, Ltd.