Assessing Omitted Confounder Bias in Multilevel Mediation Models

To draw valid inference about an indirect effect in a mediation model, there must be no omitted confounders. No omitted confounders means that there are no common causes of hypothesized causal relationships. When the no-omitted-confounder assumption is violated, inference about indirect effects can be severely biased and the results potentially misleading. Despite the increasing attention to address confounder bias in single-level mediation, this topic has received little attention in the growing area of multilevel mediation analysis. A formidable challenge is that the no-omitted-confounder assumption is untestable. To address this challenge, we first analytically examined the biasing effects of potential violations of this critical assumption in a two-level mediation model with random intercepts and slopes, in which all the variables are measured at Level 1. Our analytic results show that omitting a Level 1 confounder can yield misleading results about key quantities of interest, such as Level 1 and Level 2 indirect effects. Second, we proposed a sensitivity analysis technique to assess the extent to which potential violation of the no-omitted-confounder assumption might invalidate or alter the conclusions about the indirect effects observed. We illustrated the methods using an empirical study and provided computer code so that researchers can implement the methods discussed.     

第二类有中介的调节模型

心理学研究常用有中介的调节模型揭示调节作用通过中介变量间接实现的现象。介绍了第二类有中介的调节模型(meMO-II)的概念及优势; 将meMO-II与其它中介调节混合模型进行了辨析; 给出了meMO-II的建模方法和分析流程, 并用一个实例演示; 介绍了基于潜变量的meMO-II的分析方法、meMO-II建模方法的新进展以及meMO-II的变式。研究有助于推动调节机制研究的发展。     

国内中介效应的方法学研究

中介效应可以分析自变量对因变量的影响过程和作用机制, 已成为分析多个变量之间关系的一种重要统计方法。最近20年, 中介效应成了研究方法的一个热点。从中介效应的检验方法、效应量、类别变量的中介效应检验、纵向数据的中介效应检验和模型拓展(包括多重中介、多层中介、有调节的中介和有中介的调节模型)五个方面系统总结了国内中介效应的方法学研究的发展历程。最后对中介效应的国外方法学研究进展和中介效应的未来研究方向做了讨论和拓展。     

认知诊断框架下多级评分题目的DIF检测方法及其应用

本文对多级计分认知诊断测验的DIF概念进行了界定,并通过模拟实验以及实证研究对四种常见的多级计分DIF检验方法的适用性进行理论以及实践性的探索。研究结果表明：四种方法均能对多级计分认知诊断中的DIF进行有效的检验,且各方法的表现受模型的影响不大;相较于以总分为匹配变量,以KS为匹配变量时更利于DIF的检测;以KS为匹配变量的LDFA方法以及以KS为匹配变量的曼特尔检验方法在检测DIF题目时有着最高的检验力。     

COMPARING PLEASURE AND PAIN: THE FUNDAMENTAL MATHEMATICAL EQUIVALENCE OF REWARD GAIN AND SHOCK REDUCTION UNDER VARIABLE INTERVAL SCHEDULES

The relationship between positive and negative reinforcement and the symmetry of Thorndike's law of effect are unresolved issues in operant psychology. Here we show that, for a given pattern of responding on variable interval (VI) schedules with the same programmed rate of food rewards (positive reinforcement VI) or electric shocks (negative reinforcement VI), there is a fundamental mathematical equivalence between reward gain and shock reduction. We also provide the first normative account of how animals should respond on a negative VI schedule, showing that it is better to space responses evenly than to respond with a variable interresponse time (IRT). Published data from rats, however, indicate that these animals respond irregularly, often with a burst of activity immediately following a shock. While this is irrational in the experimental setting, it may represent an appropriate response to the heterogeneity of stimuli commonly encountered in natural environments. We discuss the broader implications of our analysis for understanding how animals evaluate appetitive and aversive stimuli.     

WHEN PERFORMANCE IMPROVEMENT IS THE GOAL: A NEW SET OF CRITERIA FOR CRITERIA

Weitz (1961), aware of the lure of tradition and expedience, urged the identification of evaluation standards for dependent variables, which he referred to as criteria for criteria. In this article, five criteria are proposed using the mnemonic SURF & C: the directness of sampling (S); the responsiveness of the target (i.e., making sure the dependent variable is under (U) the worker's control); the reliability (R) of observers; the frequent (F) assessment of the target during the intervention period; and the critical (C) nature of the target. Together the criteria provide guidelines for what and how targets should be assessed. Their necessity is illustrated in two year-long experiments designed to improve the preventive maintenance of heavy equipment in the U.S. Marine Corps. Although the criteria are limited to evaluating dependent variables in field experiments, they are recommended as the foundation for successful performance efforts in any applied setting.