结构方程模型检验：拟合指数与卡方准则

讨论了Hu和Bentler（1998,1999）推荐的检验结构方程模型的7个拟合指数准则,对这7个指数的历史、特点和表现做了比较详细的述评。指出了他们基于这7个指数的单指数准则和2-指数准则的不足之处。提出了超低显著性水平下的卡方准则,并部分重复他们的模拟例子,将卡方准则与这7个指数准则比较,结果说明新的卡方准则优于其中的6个,与另一个相当。最后简要说明了应当如何检视拟合指数进行模型检验和模型比较。     

Religious Fundamentalism and Gender Differences

The role of religious fundamentalism and its relationship to shame and guilt was evaluated in 107 students who attend a church-sponsored university. A number of personality measures were given and gender differences were analyzed. The role of externalization was similar for males in this sample to that of earlier studies. However, it was found that females showed positive correlations between externalization and both shame and guilt. For females, more fundamentalistic religious training may help to contribute to an external orientation at the expense of identifying with a personal religion, and appropriate guilt.     

无均值结构的潜变量交互效应模型的标准化估计

潜变量交互效应建模研究近年来有两项重要进展, 一是提出了潜变量交互效应模型的标准化估计及其计算公式; 二是发现无均值结构模型可以取代传统的有均值结构模型, 建模大为简化。但标准化估计是在传统的有均值结构模型中建立的, 在简化的模型中同样适用吗？本文在无均值结构模型的框架内, 给出了潜变量交互效应模型的标准化形式、计算公式和建模步骤, 并通过模拟研究比较了极大似然和广义最小二乘两种估计方法、配对乘积指标和全部乘积指标两种指标类型, 结果表明, 在计算交互效应的标准化估计时, 应当使用配对乘积指标建模, 并且首选极大似然估计。     

Women helping chewers: partner support and smokeless tobacco cessation.

Data from 363 male smokeless tobacco users and their romantic partners were analyzed to discern the role of support in cessation. Women reported playing a part in enrollment (71%), and more than half examined program materials or discussed cessation activities with the chewers. Women's reports of delivered support correlated substantially with men's experience of received support. Men's received positive support predicted abstinence at 6-month follow-up (odds ratio = 1.29, confidence interval = 1.03-1.61) and more than 24 hr of abstinence for those still using tobacco at 6 months (odds ratio = 1.75, confidence interval = 1.30-2.36) and moderated the effect of baseline depression and addiction on abstinence. Women played a major role through all stages of cessation.     

Navigating joint projects with dialogue

Dialogue has its origins in joint activities, which it serves to coordinate. Joint activities, in turn, usually emerge in hierarchically nested projects and subprojects. We propose that participants use dialogue to coordinate two kinds of transitions in these joint projects: vertical transitions, or entering and exiting joint projects; and horizontal transitions, or continuing within joint projects. The participants help signal these transitions with project markers, words such as uh‐huh, m‐hm, yeah, okay, or all right. These words have been studied mainly as signals of listener feedback (back‐channel signals) or turn‐taking devices (acknowledgment tokens). We present evidence from several types of well‐defined tasks that they are also part of a system of contrasts specialized for navigating joint projects. Uh‐huh, m‐hm and yeah are used for horizontal transitions, and okay and all right for vertical transitions.     

The troublesome semantics of conflict of interest

The sensible response to conflicts of interest is impaired by misconceptions and sloppy usage of terminology. Apparent and potential are widely misused modifiers for conflicts. Excessive legislative focus on financial interests limits understanding of the scope and significance of researchers' conflicts of interest. There is no moral or ethical failing in having a conflict of interest; the problem occurs when conflicts are not disclosed appropriately and when conflicts are allowed to bias research, teaching, or practice. Avoidance and prevention should be applied to bias, not conflicts.     

An introduction to research ethics

Practical issues throughout scientific research can be found to have an ethical aspect. There is a gray area in which scientific error (“honest error”) may be difficult to distinguish from unacceptably poor research practice or an unethical failure to follow scientific norms. Further, there is no clear margin between deceptive practices which are widely accepted and those which must be considered fraudulent. Practical problems arise in matters of data management and presentation, authorship, publication practices, “grantsmanship”, and rights of research trainees, as well as the well-recognized areas of human and animal experimentation. Beyond the gray areas, the legal definition of research misconduct is discussed in relation to research fraud, and the latest proposed definition of the Commission on Research Integrity is briefly reviewed. It is noted that the standards of ethical research are changing. Finally, there is a comment on the idea of institutional integrity in research, and the critical role of the mentor in transmitting research standards to the next generation.