Determining the effects of intraclass correlation on factorial experiments

This paper considers the effects of the presence of eqmi-correlation between observations in a k-way factorial experiment, A technique to provide unbiased F-tests is also constructed.     

Frequency of Citations as Criterion for the Ranking of Departments, Journals, and Individuals

Interest in the structure and dynamics of sociology as a discipline has led to recurrent attempts to measure the prestige, productivity, and quality of education of the various sociology departments. While several previous papers have suggested the potential value of the Science Citation Index (SCI) as a measure of scientific standing, few attempts have been made to utilize a citation-based measure of scientific prestige. This paper considers the meaning of the citation in science and in stratification theory. Procedures are described whereby citations to individual sociologists and journals are employed to yield rankings of departments of sociology and journals. The differences between these rankings and other rankings are discussed.     

Commitment in same-sex love relationships

This study sought to desribe and compare commitment structures among lesbian and gay male living-together relationships, and to develop an instrument which would measure different dimensions of commitment among gay dyads without some heterosexual biases inherent in traditional measures of interpersonal commitment. Questionnaire data collected from 32 lesbians and 50 gay males who had lived together for at least six months were factor analyzed, yielding three factors: (1) intradyadic commitment, (2) extradyadic commitment, and (3) commitment-as-trust. On all three of these indices the lesbians scored somewhat higher than did the gay males, but in fact there were surprisingly few significant differences between the gay men and the lesbians. Suggestions in the data that lesbians report somewhat less social supports and are more concerned with couple boundary maintenance, whereas gay males report somewhat more jealousy and dependency, are discussed in terms of their implications for such relationships.     

Characteristics of female perpetrators and victims of dating violence

Although many researchers have explored the topic of dating violence, limited attention has been paid to female perpetrators. Very little research has examined variables that facilitate aggression for females in dating relationships. In an effort to investigate distinct types of violent behavior, the present study separated females who experience dating violence into three categories (bi-directional aggression, perpetrator-only, and victim-only) and compared them with a control group not previously exposed to interpersonal violence. The purpose of this study was to examine variables that discriminate violent females from non-violent females. Variables that were hypothesized to be associated with aggressive behavior and investigated in the current study were interparental aggression, self-esteem, love attitudes, and alcohol use. Three hundred female college students responded to multiple self-report questionnaires examining psychological correlates of dating violence. Females in the bi-directional aggression group were more likely to have witnessed their father abuse their mother and scored significantly lower on a measure of self-esteem than non-violent controls. Females in the control group demonstrated higher scores on a measure of mature and selfless love style than did the victim or perpetrator-only participants. There were no significant group differences regarding general alcohol consumption. Implications for prevention and intervention are presented and discussed.     

Working with women in secure environments

1. Women's adverse experiences as patients in secure and other mental health services appear to reflect sexism in wider social and historical contexts. 2. A Women's Service Development Group in a medium secure unit in the United Kingdom has made improvements in services for female patients, including an independent advocacy and befriending service, a Well-Woman Clinic, and other gender-sensitive activities. 3. In general, services were evaluated positively by women. Some dissatisfaction primarily was related to activities, issues concerning seclusion, and the lack of a day area for women only. 4. Women had mixed views about their contact with male staff and patients. Because of this, policies and services in this and related areas should account for individual needs.     

Quantile plots for mean effects in the presence of variance effects for 2k-p designs

In robust parameter design, variance effects and mean effects in a factorial experiment are modelled simultaneously. If variance effects are present in a model, correlations are induced among the naive estimators of the mean effects. A simple normal quantile plot of the mean effects may be misleading because the mean effects are no longer iid under the null hypothesis that they are zero. Adjusted quantiles are computed for the case when one variance effect is significant and examples of 8-run and 16-run fractional factorial designs are examined in detail. We find that the usual normal quantiles are similar to adjusted quantiles for all but the largest and smallest ordered effects for which they are conservative. Graphically, the qualitative difference between the two sets of quantiles is negligible (even in the presence of large variance effects) and we conclude that normal probability plots are robust in the presence of variance effects.