类别变量的中介效应分析

在心理学和其他社科研究领域,研究者能熟练地进行连续变量的中介效应分析,但面对自变量、中介变量或(和)因变量为类别变量的中介效应分析,研究者往往束手无策。在阐述类别自变量中介分析方法的基础上,我们建议使用整体和相对中介相结合的类别自变量中介分析方法,并给出了分析流程。以二分因变量为例,讨论了中介变量或(和)因变量为类别变量的中介分析方法的发展过程(即尺度统一的过程),建议通过检验Za×Zb的显著性来判断中介效应的显著性。用二个实际例子演示如何进行类别变量的中介效应分析。最后展望了类别变量的中介效应分析研究的拓展方向。     

The EffectLiteR Approach for Analyzing Average and Conditional Effects

We present a framework for estimating average and conditional effects of a discrete treatment variable on a continuous outcome variable, conditioning on categorical and continuous covariates. Using the new approach, termed the EffectLiteR approach, researchers can consider conditional treatment effects given values of all covariates in the analysis and various aggregates of these conditional treatment effects such as average effects, effects on the treated, or aggregated conditional effects given values of a subset of covariates. Building on structural equation modeling, key advantages of the new approach are (1) It allows for latent covariates and outcome variables; (2) it permits (higher order) interactions between the treatment variable and categorical and (latent) continuous covariates; and (3) covariates can be treated as stochastic or fixed. The approach is illustrated by an example, and open source software EffectLiteR is provided, which makes a detailed analysis of effects conveniently accessible for applied researchers.     

基于多元回归的调节效应分析

在心理学和其他社科研究领域,大量实证研究建立调节模型,以分析自变量对因变量关系的影响机制,但在基于多元回归的调节效应分析实践中仍存在不足。我们回顾了均值中心化在基于多元回归的调节效应分析中的作用,均值中心化不影响乘积项(即调节效应)的检验,仅对一阶项(即主效应)的检验有影响。讨论了简单斜率的检验方法,建议在调节变量为连续变量时,使用Johnson-Neyman法进行简单斜率检验;在调节变量为类别变量或研究者对某个调节变量值感兴趣时,使用选点法。并用一个实际例子演示如何进行调节效应分析。随后展望了调节效应检验的拓展方向。     

A Systematic Investigation of Gesture Kinematics in Evolving Manual Languages in the Lab

Silent gestures consist of complex multi-articulatory movements but are now primarily studied through categorical coding of the referential gesture content. The relation of categorical linguistic content with continuous kinematics is therefore poorly understood. Here, we reanalyzed the video data from a gestural evolution experiment (Motamedi, Schouwstra, Smith, Culbertson, & Kirby, 2019), which showed increases in the systematicity of gesture content over time. We applied computer vision techniques to quantify the kinematics of the original data. Our kinematic analyses demonstrated that gestures become more efficient and less complex in their kinematics over generations of learners. We further detect the systematicity of gesture form on the level of thegesture kinematic interrelations, which directly scales with the systematicity obtained on semantic coding of the gestures. Thus, from continuous kinematics alone, we can tap into linguistic aspects that were previously only approachable through categorical coding of meaning. Finally, going beyond issues of systematicity, we show how unique gesture kinematic dialects emerged over generations as isolated chains of participants gradually diverged over iterations from other chains. We, thereby, conclude that gestures can come to embody the linguistic system at the level of interrelationships between communicative tokens, which should calibrate our theories about form and linguistic content.     

Predicting changes in adjustment using repeated measures of sociometric status

The authors' goals in the study were to investigate the possible gains made by including multiple assessments of status in the prediction of change in psychosocial adjustment and to compare the effectiveness of continuous and categorical measures of peer status in predicting adjustment. The authors obtained continuous and categorical measures of status (social preference and rejected status) for 644 Grade 4 students at 3 points within 1 school year (fall, winter, and spring). The authors measured peer, teacher, and self-report indexes of social adjustment (including aggression, anxiety, and sociability) in Grades 4 and 5. Both measures of peer status at all 3 time points in Grade 4 were significant predictors of adjustment in Grade 5, controlling for Grade 4 levels, with the midyear (i.e., winter) assessment showing a slight predictive advantage over the fall and spring assessments. Children who were classified as peer rejected over multiple assessments had more social adjustment problems in the next school year than did children who were classified as peer rejected at 1 time point only. The authors discuss these findings in terms of the utility of multiple assessments of both continuous and categorical measures of peer status for predicting later outcomes.     

Confusing developmental and individual differences

T. Globerson (1985, Developmental Review, 5, 261–273) compared the relative effects of an individual difference variable field dependency (FD/I) with a developmental variable (age) on a measure of M-capacity, in an analysis of variance (ANOVA) design. Age is a highly significant effect and the sum of squares associated with this factor accounts for 66% of the variance in M-capacity. FD/I is not significant and accounts for less than 1% of the variance in M-capacity. This is interpreted as indicating that M-capacity is a purely developmental phenomenon showing negligible individual differences. This paper argues that this result is an inevitable artifact of the experimental design. Further, the design is a common one in developmental psychology, and its dangers are largely unrecognized. The main problems concern (1) dichotomizing a continuous variable and treating the dichotomy as two levels of a factorial variable, (2) drawing inferences about the size of relationship between the continuous variable and the dependent variable on the basis of the variance accounted for by the factorial variable in the ANOVA, (3) differential reliability of measures of the independent variables, and (4) selective control of the sampling of the independent variables. These issues are explored in a simulation of Globerson's data and in an analogous analysis of data relating age and height, as independent variables, with weight as a dependent variable.     

Motor knowledge is one dimension for concept organization: Further evidence from a Chinese semantic dementia case

Neuropsychological and neuroimaging studies have indicated that motor knowledge is one potential dimension along which concepts are organized. Here we present further direct evidence for the effects of motor knowledge in accounting for categorical patterns across object domains (living vs. nonliving) and grammatical domains (nouns vs. verbs), as well as the integrity of other modality-specific knowledge (e.g., visual). We present a Chinese case, XRK, who suffered from semantic dementia with left temporal lobe atrophy. In naming and comprehension tasks, he performed better at nonliving items than at living items, and better at verbs than at nouns. Critically, multiple regression method revealed that these two categorical effects could be both accounted for by the charade rating, a continuous measurement of the significance of motor knowledge for a concept or a semantic feature. Furthermore, charade rating also predicted his performances on the generation frequency of semantic features of various modalities. These findings consolidate the significance of motor knowledge in conceptual organization and further highlights the interactions between different types of semantic knowledge.     

The Latent Structure of Substance Use Among American Indian Adolescents: An Example Using Categorical Variables

Researchers frequently have only categorical data to analyze and cannot, for theoretical or methodological reasons, assume that the observed variables are discrete representations of an underlying continuous variable. We present latent class analysis as an alternative method of measuring latent variables in these circumstances. Latent class analysis does not require the assumptions of factor analyses about the nature of manifest and latent variables, but does allow the use of more precise model selection than techniques such as cluster analysis. We modeled the lifetime substance use of American Indian youth. The latent class model of American Indian teenagers' substance use had four classes: Abstaining, Predominantly Alcohol, Predominantly Alcohol and Marijuana, and Plural Substance. We then demonstrated the usefulness of this latent variable by using it to differentiate levels of several variables in a manner consistent with Social Cognitive Theory.     

A technique for continuous measurement of body movement from video

The movements that we make with our body vary continuously along multiple dimensions. However, many of the tools and techniques presently used for coding and analyzing hand gestures and other body movements yield categorical outcome variables. Focusing on categorical variables as the primary quantitative outcomes may mislead researchers or distort conclusions. Moreover, categorical systems may fail to capture the richness present in movement. Variations in body movement may be informative in multiple dimensions. For example, a single hand gesture has a unique size, height of production, trajectory, speed, and handshape. Slight variations in any of these features may alter how both the speaker and the listener are affected by gesture. In this paper, we describe a new method for measuring and visualizing the physical trajectory of movement using video. This method is generally accessible, requiring only video data and freely available computer software. This method allows researchers to examine features of hand gestures, body movement, and other motion, including size, height, curvature, and speed. We offer a detailed account of how to implement this approach, and we also offer some guidelines for situations where this approach may be fruitful in revealing how the body expresses information. Finally, we provide data from a small study on how speakers alter their hand gestures in response to different characteristics of a stimulus to demonstrate the utility of analyzing continuous dimensions of motion. By creating shared methods, we hope to facilitate communication between researchers from varying methodological traditions.     

Medical burden, cerebrovascular disease, and cognitive impairment in geriatric depression: modeling the relationships with the CART analysis

Herein, the background information reflecting roles of medical burden, cerebrovascular disease and risk factors, and cognitive impairment in geriatric depression are reviewed. The authors then propose a nonparametric statistical approach to the data analysis of multiple putative causal variables for late-life depression, the Classification and Regression Tree Analysis. This analysis presents a useful approach to modeling nonlinear relationships and interactions among variables measuring physical and mental health, as well as magnetic resonance imaging and cognitive measures in depressed elderly. This method uncovers the existing interactions among multiple predictor variables, and provide thresholds for each variable, at which its predictive power becomes statistically significant. It presents a "hierarchy" of the predictors in a form of a decision tree by finding the best combination of predictors of an outcome. The authors present two models based on demographic variables, measures of vascular and nonvascular medical burden, neuroimaging indices, the Mini-Mental State Examination score, and neuropsychological test scores of 81 elderly depressed subjects. Cognitive tests of verbal fluency and executive function are identified as the best predictors of depression, followed by the frontal lobe volume and Mini-Mental State Examination. The authors observed that an interaction between frontal lobe volume, total lesion volume, and medical burden was predictive of depression.     

Using SAS PROC CALIS to fit Level-1 error covariance structures of latent growth models

In the present article, we demonstrates the use of SAS PROC CALIS to fit various types of Level-1 error covariance structures of latent growth models (LGM). Advantages of the SEM approach, on which PROC CALIS is based, include the capabilities of modeling the change over time for latent constructs, measured by multiple indicators; embedding LGM into a larger latent variable model; incorporating measurement models for latent predictors; and better assessing model fit and the flexibility in specifying error covariance structures. The strength of PROC CALIS is always accompanied with technical coding work, which needs to be specifically addressed. We provide a tutorial on the SAS syntax for modeling the growth of a manifest variable and the growth of a latent construct, focusing the documentation on the specification of Level-1 error covariance structures. Illustrations are conducted with the data generated from two given latent growth models. The coding provided is helpful when the growth model has been well determined and the Level-1 error covariance structure is to be identified.     

Analysis of Variance of Multiply Imputed Data

As a procedure for handling missing data, Multiple imputation consists of estimating the missing data multiple times to create several complete versions of an incomplete data set. All these data sets are analyzed by the same statistical procedure, and the results are pooled for interpretation. So far, no explicit rules for pooling F tests of (repeated-measures) analysis of variance have been defined. In this article we outline the appropriate procedure for the results of analysis of variance (ANOVA) for multiply imputed data sets. It involves both reformulation of the ANOVA model as a regression model using effect coding of the predictors and applying already existing combination rules for regression models. The proposed procedure is illustrated using 3 example data sets. The pooled results of these 3 examples provide plausible F and p values.     

Building an identifiable latent class model with covariate effects on underlying and measured variables

In recent years, latent class models have proven useful for analyzing relationships between measured multiple indicators and covariates of interest. Such models summarize shared features of the multiple indicators as an underlying categorical variable, and the indicators' substantive associations with predictors are built directly and indirectly in unique model parameters. In this paper, we provide a detailed study on the theory and application of building models that allow mediated relationships between primary predictors and latent class membership, but that also allow direct effects of secondary covariates on the indicators themselves. Theory for model identification is developed. We detail an Expectation-Maximization algorithm for parameter estimation, standard error calculation, and convergent properties. Comparison of the proposed model with models underlying existing latent class modeling software is provided. A detailed analysis of how visual impairments affect older persons' functioning requiring distance vision is used for illustration.This work was supported by National Institute on Aging (NIA) Program Project P01-AG-10184-03 and National Institutes of Mental Health grant R01-MH-56639-01A1. Dr. Bandeen-Roche is a Brookdale National Fellow. The authors wish to thank Drs. Gary Rubin and Sheila West for kindly making the Salisbury Eye Evaluation data available. We also thank the Editor, the Associate Editor, and three referees for their valuable comments.     

Ignoring information in binary choice with continuous variables: When is less “more”?

When can a single variable be more accurate in binary choice than multiple sources of information? We derive analytically the probability that a single variable (SV) will correctly predict one of two choices when both criterion and predictor are continuous variables. We further provide analogous derivations for multiple regression (MR) and equal weighting (EW) and specify the conditions under which the models differ in expected predictive ability. Key factors include variability in cue validities, intercorrelation between predictors, and the ratio of predictors to observations in MR. Theory and simulations are used to illustrate the differential effects of these factors. Results directly address why and when “one-reason” decision making can be more effective than analyses that use more information. We thus provide analytical backing to intriguing empirical results that, to date, have lacked theoretical justification. There are predictable conditions for which one should expect “less to be more.”     

Relative Importance Analysis: A Useful Supplement to Regression Analysis

This article advocates for the wider use of relative importance indices as a supplement to multiple regression analyses. The goal of such analyses is to partition explained variance among multiple predictors to better understand the role played by each predictor in a regression equation. Unfortunately, when predictors are correlated, typically relied upon metrics are flawed indicators of variable importance. To that end, we highlight the key benefits of two relative importance analyses, dominance analysis and relative weight analysis, over estimates produced by multiple regression analysis. We also describe numerous situations where relative importance weights should be used, while simultaneously cautioning readers about the limitations and misconceptions regarding the use of these weights. Finally, we present step-by-step recommendations for researchers interested in incorporating these analyses in their own work and point them to available web resources to assist them in producing these weights.     

Bayesian Estimation of Categorical Dynamic Factor Models

Dynamic factor models have been used to analyze continuous time series behavioral data. We extend 2 main dynamic factor model variations—the direct autoregressive factor score (DAFS) model and the white noise factor score (WNFS) model—to categorical DAFS and WNFS models in the framework of the underlying variable method and illustrate them with a categorical time series data set from an emotion study. To estimate the categorical dynamic factor models, a Bayesian method via Gibbs sampling is used. The results show that today's affect directly influences tomorrow's affect. The results are then validated by means of simulation studies. Differences between continuous and categorical dynamic factor models are examined.     

A better lemon squeezer? Maximum-likelihood regression with beta-distributed dependent variables

Uncorrectable skew and heteroscedasticity are among the "lemons" of psychological data, yet many important variables naturally exhibit these properties. For scales with a lower and upper bound, a suitable candidate for models is the beta distribution, which is very flexible and models skew quite well. The authors present maximum-likelihood regression models assuming that the dependent variable is conditionally beta distributed rather than Gaussian. The approach models both means (location) and variances (dispersion) with their own distinct sets of predictors (continuous and/or categorical), thereby modeling heteroscedasticity. The location sub-model link function is the logit and thereby analogous to logistic regression, whereas the dispersion sub-model is log linear. Real examples show that these models handle the independent observations case readily. The article discusses comparisons between beta regression and alternative techniques, model selection and interpretation, practical estimation, and software.     

Dysfunctional personality traits in adolescence: effects on alerting,orienting and executive control of attention

The present study examined attentional networks performance in 39 adolescents with dysfunctional personality traits, split into two group, Group < 10 and Group ≥ 10, according to the number of criteria they met at the Structured Clinical Interview for DSM-IV Axis II Personality Disorders. The attentional performance has been tested by means of a modified version of the Attentional Network Test (ANTI-V) which allows testing both phasic and tonic components of the alerting system, the exogenous aspect of the orienting system, the executive network and their interactions. Results showed that the orienting costs of having an invalid spatial cue were reduced in the Group ≥ 10 criteria compared to the Group < 10. Moreover, adolescents included in the Group ≥ 10 showed lower conflict when attention was cued to the target location (valid trials) but showed normal interference when there was no overpowering focus of attention (invalid trials). The results found with ANOVA after splitting the sample into two categorical groups were also observed in a complementary correlation analysis keeping intact the continuous nature of such variables. These findings are consistent with the notion that dysfunctional features of personality disorders may represent the psychological manifestations of a neuropsychological abnormality in attention and executive functioning. Finally, we discuss the implications of this attentional anomaly for dysfunctional personality traits and behaviour.     

A Unified Framework for the Comparison of Treatments with Ordinal Responses

Different latent variable models have been used to analyze ordinal categorical data which can be conceptualized as manifestations of an unobserved continuous variable. In this paper, we propose a unified framework based on a general latent variable model for the comparison of treatments with ordinal responses. The latent variable model is built upon the location-scale family and is rich enough to include many important existing models for analyzing ordinal categorical variables, including the proportional odds model, the ordered probit-type model, and the proportional hazards model. A flexible estimation procedure is proposed for the identification and estimation of the general latent variable model, which allows for the location and scale parameters to be freely estimated. The framework advances the existing methods by enabling many other popular models for analyzing continuous variables to be used to analyze ordinal categorical data, thus allowing for important statistical inferences such as location and/or dispersion comparisons among treatments to be conveniently drawn. Analysis on real data sets is used to illustrate the proposed methods.     

Analysis of Interactions and Nonlinear Effects with Missing Data: A Factored Regression Modeling Approach Using Maximum Likelihood Estimation

Abstract

When estimating multiple regression models with incomplete predictor variables, it is necessary to specify a joint distribution for the predictor variables. A convenient assumption is that this distribution is a multivariate normal distribution, which is also the default in many statistical software packages. This distribution will in general be misspecified if predictors with missing data have nonlinear effects (e.g., x²) or are included in interaction terms (e.g., x·z). In the present article, we introduce a factored regression modeling approach for estimating regression models with missing data that is based on maximum likelihood estimation. In this approach, the model likelihood is factorized into a part that is due to the model of interest and a part that is due to the model for the incomplete predictors. In three simulation studies, we showed that the factored regression modeling approach produced valid estimates of interaction and nonlinear effects in regression models with missing values on categorical or continuous predictor variables under a broad range of conditions. We developed the R package mdmb, which facilitates a user-friendly application of the factored regression modeling approach, and present a real-data example that illustrates the flexibility of the software.