Robustness of normal theory methods in the analysis of linear latent variate models

The structure of the covariance matrix of sample covariances under the class of linear latent variate models is derived using properties of cumulants. This is employed to provide a general framework for robustness of statistical inference in the analysis of covariance structures arising from linear latent variate models. Conditions for normal theory estimators and test statistics to retain each of their usual asymptotic properties under non-normality of latent variates are given. Factor analysis, LISREL and other models are discussed as examples.     

Abstraction of contingency in concept learning

It is generally agreed that concept learning involves the abstraction of some general representation or schema. Just what is abstracted, however, and how it is used in the classification of sets of stimuli in the natural world or in the laboratory, remain outstanding questions. In this paper a hypothesis involving contingency abstraction is described as a possible solution to these questions. An experiment which manipulated measured contingency in a concept-learning task, and which offered empirical support for the hypothesis, is reported. The advantages of a contingency-abstraction theory of concept learning are briefly discussed.     

Quality of life is a process not an outcome

Quality improvement mechanisms increasingly use outcome measures to evaluate health care providers. This move toward outcome measures is a radical departure from the traditional focus on process measures. More radical still is the proposal to shift from relatively simple and proximal measures of outcome, such as mortality, to complex outcomes, such as quality of life. While the practical, scientific, and ethical issues associated with the use of outcomes such as mortality and morbidity to compare health care providers have been well rehearsed, the specific concerns associated with the use of quality of life measures in quality of care research have received little attention. As with much research on quality of life there is a tendency to assume that the disadvantages are outweighed by the general virtue of ??listening?? to patients. In this paper we disagree with this assumption and argue that quality of life is a process, not an outcome.     

Low Prospects and High Risk: Structural Determinants of Health Associated with Sexual Risk Among Young African American Women Residing in Resource-Poor Communities in the South

African American women at increased risk of HIV/sexually transmitted infection (STI) may engage in risky sex as a coping mechanism for depressed economic conditions. This study examines the association between high-risk sexual behavior and structural determinants of sexual health among a sample of young African American women. 237 young African American women (16–19 years old) from economically disadvantaged neighborhoods in North Carolina were enrolled into a randomized trial testing the efficacy of an adapted HIV/STI prevention intervention. Logistic regression analyses predicted the likelihood that young women reporting lack of food at home, homelessness and low future prospects would also report sexual risk behaviors. Young women reporting a lack of food at home (22 %), homelessness (27 %), and low perceived education/employment prospects (19 %) had between 2.2 and 4.7 times the odds as those not reporting these risk factors of reporting multiple sex partners, risky sex partners including older men and partners involved in gangs, substance use prior to sex, and exchange sex. Self-reported structural determinants of sexual health were associated with myriad sexual risk behaviors. Diminished economic conditions among these young women may lead to sexual risk due to hopelessness, the need for survival or other factors.     

Disentangling Child-Specific and Family-Wide Processes Underlying Negative Mother-Child Transactions

The current study examines the transactional processes between maternal negativity and child disruptive behavior during early childhood, using a sibling comparison design. This design allows for a more sophisticated test of the behavioral training hypothesis than is possible when only one child per family is included in the analysis as it excludes two confounders (passive gene-environment correlation and family-wide environmental influence). Three hundred and ninety-seven families were visited on three occasions when the target child was 1.5, 3 and 4.5 years old (920 children, 51.5% female). The target child and up to three older siblings per family were included in data collection and analysis. Mothers and fathers reported on children’s disruptive behavior and mothers reported on their negativity to children. Within-family cross-lag pathways, as well as a significant indirect effect through which children increased their own disruptive behavior via maternal negativity confirmed a behavioral training effect. Family level maternal negativity and sibling disruptive behavior showed high levels of stability over three to four years and no context effect for maternal negativity or sibling disruptive behavior was seen. The theoretical and clinical implications of these findings are discussed.     

Fitting direct covariance structures by the MSTRUCT modeling language of the CALIS procedure

This paper demonstrates the usefulness and flexibility of the general structural equation modelling (SEM) approach to fitting direct covariance patterns or structures (as opposed to fitting implied covariance structures from functional relationships among variables). In particular, the MSTRUCT modelling language (or syntax) of the CALIS procedure (SAS/STAT version 9.22 or later: SAS Institute, 2010) is used to illustrate the SEM approach. The MSTRUCT modelling language supports a direct covariance pattern specification of each covariance element. It also supports the input of additional independent and dependent parameters. Model tests, fit statistics, estimates, and their standard errors are then produced under the general SEM framework. By using numerical and computational examples, the following tests of basic covariance patterns are illustrated: sphericity, compound symmetry, and multiple‐group covariance patterns. Specification and testing of two complex correlation structures, the circumplex pattern and the composite direct product models with or without composite errors and scales, are also illustrated by the MSTRUCT syntax. It is concluded that the SEM approach offers a general and flexible modelling of direct covariance and correlation patterns. In conjunction with the use of SAS macros, the MSTRUCT syntax provides an easy‐to‐use interface for specifying and fitting complex covariance and correlation structures, even when the number of variables or parameters becomes large.     

Testing differences between nested covariance structure models: Power analysis and null hypotheses

For comparing nested covariance structure models, the standard procedure is the likelihood ratio test of the difference in fit, where the null hypothesis is that the models fit identically in the population. A procedure for determining statistical power of this test is presented where effect size is based on a specified difference in overall fit of the models. A modification of the standard null hypothesis of zero difference in fit is proposed allowing for testing an interval hypothesis that the difference in fit between models is small, rather than zero. These developments are combined yielding a procedure for estimating power of a test of a null hypothesis of small difference in fit versus an alternative hypothesis of larger difference.     

Gender and victimization by intimates

Recent data demonstrate that, although gender has an impact upon the experience of being a victim of an intimate's violence, there is no particular personality pattern that leads one to become a victim. Rather, women--who are socialized to adapt and submit, and who are likely to become victims of men's sexual violence or physical abuse--may not develop adequate self-protection skills as children, especially if they come from childhood homes in which females are victimized, leading to a later vulnerability to physical and sexual abuse. Men, however, socialized to express anger and aggression in an outward manner, learn to model the abuse witnessed or experienced in childhood and often learn that women are the "appropriate" recipients of this violence. Social learning theories of modeling and aggression are used to explain how such personality patterns develop, and the theory of learned helplessness is used to explain battered women's coping responses to their partners' abusive behavior. The extreme situation, in which a battered woman kills her partner in self-defense, is analyzed in order to understand women victims' sense of desperation and entrapment in severely abusive relationships and the extent to which their behaviors are in reaction to the abuse perpetrated by the mate.     

Offender and Offence Characteristics of School Shooting Incidents

School shootings are a concern due to their impact in the local community. This paper aimed to (a) establish frequent characteristics of the offender and offence, (b) explore the differences between offenders who are over the age of 18 years and those who are younger, and (c) consider the underlying themes of the offence characteristics. Data were collected on 28 cases through accessing resources such as West Law and case studies. The majority of the offenders were Caucasian and US citizens and suffered from depression. Their offences were primarily well planned, involved more than three deaths, and resulted in the offender committing suicide. Pearson's chi‐square test and Fisher's exact test identified significant differences between the two age groups. Offenders who were 18 years of age or under were more likely to experience depression, be US citizens and be linked to the school. Additionally, offenders who were 18 years of age or under were more likely to have stolen their weapons and made threats prior to the incident. Smallest space analysis revealed four thematic regions in relation to the offence characteristics: making an impact, delivering a message, doing unrestrained activity, and targeting specific individuals. These findings have implications for risk assessment and furthering understanding. Copyright © 2015 John Wiley & Sons, Ltd.