基于结构方程模型的多重中介效应分析

多重中介模型是指存在多个中介变量的模型。多重中介模型可以分析特定中介效应、总的中介效应和对比中介效应。指出了目前多重中介模型分析普遍存在的问题,包括分析不完整、使用Sobel检验带来的局限。建议通过增加辅助变量的方法进行完整的多重中介效应分析,使用偏差校正的Bootstrap方法进行中介检验。总结出一个多重中介SEM分析流程,并有示例和相应的MPLUS程序。随后展望了辅助变量和中介效应检验方法的发展方向。     

BRIEF EXPERIMENTAL ANALYSIS OF STIMULUS PROMPTS FOR ACCURATE RESPONDING ON ACADEMIC TASKS IN AN OUTPATIENT CLINIC

Brief multielement designs were used to examine the effects of specific instructional strategies on accuracy of academic performance during outpatient evaluations of 4 children with learning disorders. Instructional strategies that improved accuracy on academic tasks were identified for all participants. These results suggest that the application of experimental analysis methodologies to instructional variables may facilitate the identification of stimulus prompts that are associated with enhanced academic performance.     

Foveal-to-peripheral priming in letter identification reflects response bias, not enhanced visibility

Several studies have suggested that it might be easier to identify a letter when its replication is simultaneously presented in the visual display. We have measured identification accuracy for a memorized target letter that was shown in the peripheral field in two conditions in which a letter either identical or different in shape was flashed simultaneously at fixation. The results of three forced-choice experiments showed that, contrary to Geiger and Lettvin's (1986) previous findings, an identity prime did not improve the visibility of a temporally backward masked target letter. However, the foveal prime letter could apparently bias the subject's decision so that he/she was much more ready to report the more visible prime letter as the peripheral target.     

An alternative two stage least squares (2SLS) estimator for latent variable equations

The Maximum-likelihood estimator dominates the estimation of general structural equation models. Noniterative, equation-by-equation estimators for factor analysis have received some attention, but little has been done on such estimators for latent variable equations. I propose an alternative 2SLS estimator of the parameters in LISREL type models and contrast it with the existing ones. The new 2SLS estimator allows observed and latent variables to originate from nonnormal distributions, is consistent, has a known asymptotic covariance matrix, and is estimable with standard statistical software. Diagnostics for evaluating instrumental variables are described. An empirical example illustrates the estimator. I gratefully acknowledge support for this research from the Sociology Program of the National Science Foundation (SES-9121564) and the Center for Advanced Study in the Behavioral Sciences, Stanford, California. This paper was presented at the Interdisciplinary Consortium for Statistical Applications at Indiana University at Bloomington (March 2, 1994) and at the RMD Conference on Causal Modeling at Purdue University, West Lafayette, Indiana (March 3-5, 1994).     

Family characteristics as mediators of the influence of problem drinking and multiple risk status on child mental health

Structural equation modeling was used to test a theoretical model in which family cohesion and family reframing coping were hypothesized as mediators between family drinking problems, multiple risk factors, negative life events, and child mental health (conduct disorder, depression, anxiety) in two-parent families. Family cohesion mediated the relationships of family drinking problems and negative life events to child conduct disorder and depression. Negative life events mediated the relationships of family drinking problems and family multiple risk to child conduct disorder. Family reframing coping did not function as a mediator nor was it related to child mental health when other factors were considered simultaneously. Results indicate that increasing family cohesion and reducing sources of stress within the family (negative life events) represent promising areas for interventions for children with problem-drinking parents. Work on this study was funded in part by the National Institute for Mental Health Grant 2-P50-MH39246-06 to support a Preventive Intervention Research Center. The authors gratefully acknowledge contributions made by Rita Shell, Marcia Michaels, Joanne Gersten, George Knight, and Carolyn Berg.     

The Effect of Non-conscious Perception of Frequent Stimuli on Credibility Judgement

The aim of the work presented here is to investigate the effects of mere-exposure to familiar and unfamiliar stimuli (as primes) on credibility judgement about sentences unrelated to the primes. These target sentences are presented just after the prime. In all, 124 students participated in 3 priming experiments. The nature of the primes (both infraliminary and supraliminary) is different in each experiment: we used public vs. unknown faces, exposed vs. unexposed faces, and objects vs. nonobjects. Primes were presented for 50msec or 400msec. After the presentation of each prime, subjects had to judge the credibility of an assertion; its ambiguity has previously been tested with 100 subjects. The results show that assertions which follow familiar primes (public faces, exposed faces, or objects) are granted more credibility that those which follow unfamiliar primes (unknown faces or nonobjects). This effect is observed especially when presentation time is 50msec.     

再认的两种机制及其与启动效应的关系

通过两个实验对两种再认所涉及的心理机制及其与启动效应的关系进行探讨。实验一用高频字和低频字作实验材料,分别测量集中注意和分散注意条件下,两种再认的成绩。结果表明：学习阶段的注意分散使以提取为基础的再认成绩下降,对以熟悉性为基础的再认成绩没有影响。低频字以熟悉性为基础的再认优于高频字,但字频对以提取为基础的再认没有影响。实验二分别测量两种再认项目的补笔正确率。结果表明：以熟悉性为基础的再认的补笔正确率显著高于以提取为基础的再认的补笔正确率。据此可以推论：两种再认可能有不同的心理机制。以提取为基础的再认属于有意识记忆的范畴,以熟悉性为基础的再认属于无意识记忆的范畴。后者和启动效应的心理基础可能是共同的。     

初中生考试成绩归因模式研究

在广泛调查初中生对考试成绩的归因和情感反应的基础上,根据Ｗｅｉｎｅｒ的动机和情绪的归因理论,提出了初中生对考试成绩归因的假设模式,然后运用协方差结构模型的统计方法对这个假设模型进行了检验．虽然最初模型经ｘ２拟合度检验被拒绝,但经设定探查产生的最终模型较合适地拟合了数据,并部分支持了Ｗｅｉｎｅｒ的归因理论．按照这个模型,努力归因一内源性维度对初中生的自信心、期望、成就动机产生积极的影响．     

字符的知觉表征及其启动

实验以英文单词,汉字和图片为材料,采用速示补笔测验测量了大脑左右半球的启动效应。结果是,如果学习与测验的字体和大小不变,则视觉启动右半球高于左半球。其他材料不产生这种效应。这表明,至少有一个分离的系统对构成启动的汉字的视觉表征进行编码,此系统在右半球表征形状特异形状特异信息更有效;对字符抽象特征的表征,两半球无显著差异,速示补笔测量的是前语义水平的知觉表征,是一种知觉（数据驱动）内隐记忆测验。