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。     

潜变量交互效应建模: 告别均值结构

潜变量交互效应建模研究近年来有了长足的发展, 但模型中被认为不可缺少的均值结构往往让实际应用工作者却步。本文首先分析了潜变量交互效应模型中均值结构产生的根源; 然后讨论了指标变换与均值结构的关系; 接着提出了一个均值为零的潜变量交互结构, 所建立的模型不需要均值结构, 却不会改变主效应和交互效应等参数; 最后用模拟例子对无均值结构和有均值结构的两种模型的参数估计进行了比较, 结果符合理论预期, 困扰人们多年的均值结构问题从此可以终结。     

潜变量交互效应分析方法

简要回顾了分析显变量交互效应的常用方法。详细讨论了目前分析潜变量交互效应的主要方法,包括用潜变量的因子得分做回归分析、分组线性结构方程模型分析、加入乘积项的结构方程模型分析和两步最小二乘回归分析,并比较和评价了这些方法的优缺点。最后归纳了潜变量交互效应分析方法的研究趋势,并介绍了新近进展(包括LMS方法和GAPI方法)。     

潜变量交互效应标准化估计：方法比较与选用策略

标准化估计对模型的解释和效应大小的比较有重要作用。虽然潜变量交互效应的恰当标准化估计公式已经面世超过10年, 国内外都在使用和引用, 但至今未见到关于不同估计方法得到的恰当标准化估计的系统比较。通过模拟实验, 比较了乘积指标法、潜调节结构方程(LMS)、无先验信息和有先验信息的贝叶斯法的潜变量交互效应标准化估计在不同条件下的表现。结果发现, 在正态条件下, LMS和有信息贝叶斯法表现较好; 而在非正态条件下, 乘积指标法比较稳健, 但需要较大的样本(不小于500), 小样本且外生潜变量之间相关很低时可使用无信息贝叶斯法。     

潜变量交互效应结构方程：分布分析方法

对于潜变量交互效应结构方程分析,目前应用较多的是乘积指标方法.分布分析方法国内还罕有应用,包括潜调节结构方程（LMS）方法和准极大似然（QML）方法.该研究以乘积指标方法的模型假设为参照,介绍了分布分析方法的模型假设.并简要叙述了LMS方法及其Mplus程序,QML方法及其QML程序.综合现有研究结果,总结出LMS和QML方法、无约束和约束方法的特点,从中可以看出各方法的优缺点,推荐了不同条件下合适的分析方法.     

基于多层结构方程模型的情境效应分析—— 兼与多层线性模型比较

多层(嵌套)数据的变量关系研究, 必须借助多层模型来实现。两层模型中, 层一自变量Xij按组均值中心化, 并将组均值 置于层2截距方程式中, 可将Xij对因变量Yij的效应分解为组间和组内部分, 二者之差被称为情境效应, 称为情境变量。多层结构方程模型(MSEM)将多层线性模型(MLM)和结构方程模型(SEM)相结合, 通过设置潜变量和多指标的方法校正了MLM在情境效应分析中出现的抽样误差和测量误差, 同时解决了数据的多层(嵌套)结构和潜变量的估计问题。除了分析原理的说明, 还以班级平均竞争氛围对学生竞争表现的情境效应为例进行分析方法的示范, 并比较MSEM和MLM的异同, 随后展望了MSEM情境效应模型、情境效应无偏估计方法和情境变量研究的拓展方向。     

基于结构方程模型的有调节的中介效应分析

有调节的中介模型是中介过程受到调节变量影响的模型。指出了目前有调节的中介效应分析普遍存在的问题：当前有调节的中介效应检验大多使用多元线性回归分析,忽略了测量误差;而基于结构方程模型(SEM)的有调节的中介效应分析需要产生乘积指标,又会面临乘积指标生成和乘积项非正态分布的问题。在简介潜调节结构方程(LMS)方法后,建议使用LMS方法得到偏差校正的bootstrap置信区间来进行基于SEM的有调节的中介效应分析。总结出一个有调节的中介SEM分析流程,并有示例和相应的Mplus程序。文末展望了LMS和有调节的中介模型的发展方向。     

结构方程模型中调节效应的标准化估计

回归分析和结构方程分析中,标准化估计对解释模型和比较效应大小有重要作用。对于调节效应模型（或交互效应模型）,通常的标准化估计没有意义。虽然显变量的调节效应模型标准化估计问题已经解决,但潜变量的调节效应模型标准化估计问题复杂得多。本文先介绍回归分析中显变量调节效应模型的标准化估计,然后提出了一种通过参数的原始估计和通常标准化估计来计算潜变量调节效应模型的“标准化”估计的方法,得到的“标准化”估计是尺度不变的,说明可以用“标准化”估计来解释和比较主效应和调节效应     

二阶潜增长模型标度方法及其可比的一阶潜增长模型

潜增长模型(LGM)是分析纵向数据的一种强有力工具,在心理学和其他社会科学研究领域受到重视。多指标测量的变量,既可以用合成分数建立单变量LGM(一阶LGM),也可以用指标建立潜变量LGM(二阶LGM)。简述了二阶LGM标度方法(包括尺度指标法和效应编码),提出了有可操作性的潜变量标准化标度方法和合成分数的一阶LGM标准化模型。系统总结了二阶LGM标度方法及其可比的一阶LGM建模,并用多指标变量的实测数据进行示例。推荐使用效应编码法对二阶LGM进行标度和标准化。     

基于结构方程模型的多层调节效应

使用多层线性模型进行调节效应分析在社科领域已常有应用。尽管多层线性模型区分了层1自变量的组间和组内效应、实现了多层调节效应的分解, 仍然存在抽样误差和测量误差。建议在多层结构方程模型框架下, 设置潜变量和多指标来有效校正抽样误差和测量误差。在介绍多层调节SEM分析的随机系数预测法和潜调节结构方程法后, 总结出一套多层调节的SEM分析流程, 通过一个例子来演示如何用Mplus软件进行多层调节SEM分析。随后评述了多层调节效应分析方法在国内心理学的应用现状, 并展望了多层结构方程和多层调节研究的拓展方向。     

Multilevel covariance structure analysis of sibling substance use and intrafamily conflict

Conventional covariance structure analysis, such as factor analysis, is often applied to data that are obtained in a hierarchical fashion, such as siblings observed within families. A more appropriate specification is demonstrated which explicitly models the within-level and between-level covariance matrices of sibling substance use and intrafamily conflict. Participants were 267 target adolescents (mean age=13.11 years) and 318 siblings (mean age=15.03 years). The level of homogeneity within sibling clusters, and heterogeneity among families, was sufficient to conduct a multilevel covariance structure analysis (MCA). Parental and family-level variables consisting of marital status, socioeconomic status, marital discord, parent use, and modeling of substances were used to explain heterogeneity among families. Marital discord predicted intrafamily conflict, and parent marital status and modeling of substances predicted sibling substance use. Advantages and uses of hierarchical designs and conventional covariance structure software for multilevel data are discussed.     

The Strengths and Difficulties Questionnaire: the self-report version in American urban and suburban youth

This study examined the factor structure of the Strengths and Difficulties Questionnaire (SDQ) in urban inner-city and suburban general population samples of American youth. The SDQ was administered to 4,661 predominantly minority urban youth (mean age = 13.0 years, SD = 2.02) and 937 predominantly Caucasian suburban youth (mean age = 14.0 years, SD = 1.56). The authors examined SDQ factor structure using structural equation modeling techniques. The fit for the 5-factor models in both samples was very good (comparative fit index = .96-.97). However, the factor loadings and, correspondingly, the scale reliabilities for most of the SDQ scales (except for Emotional Problems) were low, which suggests poor specificity of the items. Subsequent exploratory and confirmatory factor analyses suggested that a 3-factor structure had a comparable model fit and better internal consistency of the scales. The study concluded that, although the SDQ scales do conform reasonably well to a 5-factor model, the scales are unsatisfactory in other respects and that, in its present form, the instrument has inadequate psychometric characteristics. Future research is needed for further scale development.     

Exploratory structural equation modeling,bifactor models,and standard confirmatory factor analysis models: Application to the BASC-2 Behavioral and Emotional Screening System Teacher Form

Several psychological assessment instruments are based on the assumption of a general construct that is composed of multiple interrelated domains. Standard confirmatory factor analysis is often not well suited for examining the factor structure of such scales. This study used data from 1885 elementary school students (mean age = 8.77 years, SD = 1.47 years) to examine the factor structure of the Behavioral Assessment System for Children, Second Edition (BASC-2) Behavioral and Emotional Screening System (BESS) Teacher Form that was designed to assess general risk for emotional/behavioral difficulty among children. The modeling sequence included the relatively new exploratory structural equation modeling (ESEM) approach and bifactor models in addition to more standard techniques. Findings revealed that the factor structure of the BASC-2 BESS Teacher Form is multidimensional. Both ESEM and bifactor models showed good fit to the data. Bifactor models were preferred on conceptual grounds. Findings illuminate the hypothesis-generating power of ESEM and suggest that it might not be optimal for instruments designed to assess a predominant general factor underlying the data.     

Factors Affecting the Self-Concepts of South African Students

Abstract

Relationships among family macrosocial structures, proximate family settings, attributions of responsibility, and African adolescents' self-concepts were examined. Data were collected from 460 South African high school students (234 girls, 226 boys; mean age = 18.6 years). On the basis of partial least squares path modeling, the results suggest that (a) family macrosocial structure, proximate family settings, and the individual's sense of responsibility for academic outcomes had modest to strong associations with different dimensions of self-concept; and (b) there were gender-related differences in the structure of the adolescents' social status backgrounds and self-concepts and in the relationships among social status, perceptions of parents' support for learning, personal responsibility, and self-concept.     

新世纪20年国内心理统计方法研究回顾

新世纪头20年, 国内心理学11本专业期刊一共发表了213篇统计方法研究论文。研究范围主要包括以下10类(按论文篇数排序)：结构方程模型、测验信度、中介效应、效应量与检验力、纵向研究、调节效应、探索性因子分析、潜在类别模型、共同方法偏差和多层线性模型。对各类做了简单的回顾与梳理。结果发现, 国内心理统计方法研究的广度和深度都不断增加, 研究热点在相互融合中共同发展; 但综述类论文比例较大, 原创性研究论文比例有待提高, 研究力量也有待加强。     

INTERVENTIONS TO REDUCE HIGH-VOLUME PORTABLE HEADSETS: “TURN DOWN THE SOUND!”

Two studies examined effects of interventions to reduce noise levels from portable stereo headphones. Study 1 examined the effectiveness of warning signs posted in and nearby public elevators with 567 passengers possessing a portable headphone (total N = 7,811). During a 9-day baseline, the mean percentage of headphones played at an observer-audible level was 85%. During a subsequent 6-day warning sign phase, the mean percentage of audible headphones declined to 59%, which increased to a mean of 76% during a second baseline phase (5 days). Study 2 assessed the impact of a student confederate who lowered his or her observer-audible headphone volume at the polite request of a second student confederate. Of the 4,069 elevator passengers, 433 possessed a portable headset. The mean percentage of observer-audible headphones during a 4-day baseline was 85%. Subsequently, a 5-day modeling intervention reduced audible volumes to a mean of 46%. During a second baseline phase of 4 days, the mean level was 77%, and during reintroduction of the modeling phase (9 days) the mean level was 42%. The modeling intervention was significantly more effective with women (53% compliance) than men (29% compliance).     

Fitting multivariage normal finite mixtures subject to structural equation modeling

This paper is about fitting multivariate normal mixture distributions subject to structural equation modeling. The general model comprises common factor and structural regression models. The introduction of covariance and mean structure models reduces the number of parameters to be estimated in fitting the mixture and enables one to investigate a variety of substantive hypotheses concerning the differences between the components in the mixture. Within the general model, individual parameters can be subjected to equality, nonlinear and simple bounds constraints. Confidence intervals are based on the inverse of the Hessian and on the likelihood profile. Several illustrations are given and results of a simulation study concerning the confidence intervals are reported.     

Observational modeling effects for movement dynamics and movement outcome measures across differing task constraints: a meta-analysis

The authors conducted a meta-analysis of the observational modeling literature to quantify overall between-participants treatment effects obtained when movement behaviors are acquired. Effects were obtained and reported separately for movement dynamics (MD) and movement outcome (MO) measures. The overall mean observational modeling treatment effects (delta(u)Bi) were 0.77 and 0.17, respectively, for MD and MO measures. The effects reflected a significant advantage of observational modeling over practice-only control conditions. Most important, the magnitude of the obtained effects was far stronger for MD than for MO measures, confirming a distinctive response to observational modeling during motor learning. The advantage for MD measures over observational modeling measures was replicated for different types of tasks. Observational modeling was particularly beneficial for serial tasks (delta(u)Bi = 1.62 and 0.61, respectively, for MD and MO). There were slightly reduced effects for continuous tasks (delta(u)Bi = 1.01 and 0.51, respectively, for MD and MO) and smaller to medium-sized effects for discrete tasks (delta(u)Bi = 0.56 and 0.10, respectively, for MD and MO). The authors discuss those findings with reference to the visual perception perspective on observational modeling, i.e., that demonstrations primarily convey relative motions required to approximate modeled movement behaviors.