Improved Insight into and Prediction of Network Dynamics by Combining VAR and Dimension Reduction

To understand within-person psychological processes, one may fit VAR(1) models (or continuous-time variants thereof) to multivariate time series and display the VAR(1) coefficients as a network. This approach has two major problems. First, the contemporaneous correlations between the variables will frequently be substantial, yielding multicollinearity issues. In addition, the shared effects of the variables are not included in the network. Consequently, VAR(1) networks can be hard to interpret. Second, crossvalidation results show that the highly parametrized VAR(1) model is prone to overfitting. In this article, we compare the pros and cons of two potential solutions to both problems. The first is to impose a lasso penalty on the VAR(1) coefficients, setting some of them to zero. The second, which has not yet been pursued in psychological network analysis, uses principal component VAR(1) (termed PC-VAR(1)). In this approach, the variables are first reduced to a few principal components, which are rotated toward simple structure; then VAR(1) analysis (or a continuous-time analog) is applied to the rotated components. Reanalyzing the data of a single participant of the COGITO study, we show that PC-VAR(1) has the better predictive performance and that networks based on PC-VAR(1) clearly represent both the lagged and the contemporaneous variable relations.     

The Gaussian Graphical Model in Cross-Sectional and Time-Series Data

We discuss the Gaussian graphical model (GGM; an undirected network of partial correlation coefficients) and detail its utility as an exploratory data analysis tool. The GGM shows which variables predict one-another, allows for sparse modeling of covariance structures, and may highlight potential causal relationships between observed variables. We describe the utility in three kinds of psychological data sets: data sets in which consecutive cases are assumed independent (e.g., cross-sectional data), temporally ordered data sets (e.g., n = 1 time series), and a mixture of the 2 (e.g., n > 1 time series). In time-series analysis, the GGM can be used to model the residual structure of a vector-autoregression analysis (VAR), also termed graphical VAR. Two network models can then be obtained: a temporal network and a contemporaneous network. When analyzing data from multiple subjects, a GGM can also be formed on the covariance structure of stationary means—the between-subjects network. We discuss the interpretation of these models and propose estimation methods to obtain these networks, which we implement in the R packages graphicalVAR and mlVAR. The methods are showcased in two empirical examples, and simulation studies on these methods are included in the supplementary materials.     

The Advantages of Using Standardized Scores in Causal Analysis

This article compares the usefulness of standard score results (such as correlations and standardized regression coefficients) to that of raw scores results (such as covariances and raw score regression coefficients). The 2 main advantages of presenting standardized results is then examined. First, reporting standardized coefficients allows recognition of causal relationships that replicate across studies, facilitating meta‐analysis. Second, an analysis with standardized coefficients allows the researcher to determine which of the predictor variables has the largest relative impact on the criterion variable. The implications of these 2 standardized score advantages are demonstrated with examples.     

Feature extraction of event-related potentials using wavelets: an application to human performance monitoring

This report describes the development and evaluation of mathematical models for predicting human performance from discrete wavelet transforms (DWT) of event-related potentials (ERP) elicited by task-relevant stimuli. The DWT was compared to principal components analysis (PCA) for representation of ERPs in linear regression and neural network models developed to predict a composite measure of human signal detection performance. Linear regression models based on coefficients of the decimated DWT predicted signal detection performance with half as many free parameters as comparable models based on PCA scores. In addition, the DWT-based models were more resistant to model degradation due to over-fitting than PCA-based models. Feed-forward neural networks were trained using the backpropagation algorithm to predict signal detection performance based on raw ERPs, PCA scores, or high-power coefficients of the DWT. Neural networks based on high-power DWT coefficients trained with fewer iterations, generalized to new data better, and were more resistant to overfitting than networks based on raw ERPs. Networks based on PCA scores did not generalize to new data as well as either the DWT network or the raw ERP network. The results show that wavelet expansions represent the ERP efficiently and extract behaviorally important features for use in linear regression or neural network models of human performance. The efficiency of the DWT is discussed in terms of its decorrelation and energy compaction properties. In addition, the DWT models provided evidence that a pattern of low-frequency activity (1 to 3.5 Hz) occurring at specific times and scalp locations is a reliable correlate of human signal detection performance.     

The Detection and Interpretation of Interaction Effects Between Continuous Variables in Multiple Regression

issues in the detection and interpretation of interaction effects between quantitative variables in multiple regression analysis are discussed. Recent articles by Cronbach (1987) and Dunlap and Kemery (1987) suggested the use of two transformations to reduce "problems" of multicollinearity. These transformations are discussed in the context of the conditional nature of multiple regression with product terms. It is argued that although additive transformations do not affect the overall test of statistical interaction, they do affect the interpretational value of regression coefficients. Factors other than multicollinearity that may account for failures to observe interaction effects are noted.     

A Computationally More Efficient and More Accurate Stepwise Approach for Correcting for Sampling Error and Measurement Error

Over the last decade or two, multilevel structural equation modeling (ML-SEM) has become a prominent modeling approach in the social sciences because it allows researchers to correct for sampling and measurement errors and thus to estimate the effects of Level 2 (L2) constructs without bias. Because the latent variable modeling software Mplus uses maximum likelihood (ML) by default, many researchers in the social sciences have applied ML to obtain estimates of L2 regression coefficients. However, one drawback of ML is that covariance matrices of the predictor variables at L2 tend to be degenerate, and thus, estimates of L2 regression coefficients tend to be rather inaccurate when sample sizes are small. In this article, I show how an approach for stabilizing covariance matrices at L2 can be used to obtain more accurate estimates of L2 regression coefficients. A simulation study is conducted to compare the proposed approach with ML, and I illustrate its application with an example from organizational research.     

Dyadic measurement invariance and its importance for replicability in romantic relationship science

Comparisons of group means, variances, correlations, and/or regression slopes involving psychological variables rely on an assumption of measurement invariance—that the latent variables under investigation have equivalent meaning and measurement across group. When measures are noninvariant, replicability suffers, as comparisons are either conceptually meaningless, or hindered by inflated Type I error rates. We propose that the failure to account for interdependence among dyad members when testing measurement invariance may be a potential source of unreplicable findings in relationship research. We developed fully dyadic versions of invariance models, created an R package (dySEM) to make specifying dyadic invariance models easier and reporting more reproducible, and executed a Registered Report for gauging the extent of dyadic (non)invariance in romantic relationship research across measures of relationship well‐being, personality, and sexuality in a sample of 282 heterosexual couples. We found that although a number of popular measures display good evidence of dyadic invariance, a few display concerning levels and interesting patterns of noninvariance, while others appeared either noninvariant or poorly fitting for both men and women. We discuss our findings in terms of their meaning for the replicability dyadic close relationship research. We close by arguing that increased theorizing and research on dyadic invariance, and inclusive methods for analyzing invariance with indistinguishable dyads, are needed to capitalize on the opportunity to advance our field's understanding of dyadic constructions of relational concepts.     

Individual differences in cognitive arithmetic

Unities in the processes involved in solving arithmetic problems of varying operations have been suggested by studies that have used both factor-analytic and information-processing methods. We designed the present study to investigate the convergence of mental processes assessed by paper-and-pencil measures defining the Numerical Facility factor and component processes for cognitive arithmetic identified by using chronometric techniques. A sample of 100 undergraduate students responded to 320 arithmetic problems in a true-false reaction-time (RT) verification paradigm and were administered a battery of ability measures spanning Numerical Facility, Perceptual Speed, and Spatial Relations factors. The 320 cognitive arithmetic problems comprised 80 problems of each of four types: simple addition, complex addition, simple multiplication, and complex multiplication. The information-processing results indicated that regression models that included a structural variable consistent with memory network retrieval of arithmetic facts were the best predictors of RT to each of the four types of arithmetic problems. The results also verified the effects of other elementary processes that are involved in the mental solving of arithmetic problems, including encoding of single digits and carrying to the next column for complex problems. The relation between process components and ability measures was examined by means of structural equation modeling. The final structural model revealed a strong direct relation between a factor subsuming efficiency of retrieval of arithmetic facts and of executing the carry operation and the traditional Numerical Facility factor. Furthermore, a moderate direct relation between a factor subsuming speed of encoding digits and decision and response times and the traditional Perceptual Speed factor was also found. No relation between structural variables representing cognitive arithmetic component processes and ability measures spanning the Spatial Relations factor was found. Results of the structural modeling support the conclusion that information retrieval from a network of arithmetic facts and execution of the carry operation are elementary component processes involved uniquely in the mental solving of arithmetic problems. Furthermore, individual differences in the speed of executing these two elementary component processes appear to underlie individual differences on ability measures that traditionally span the Numerical Facility factor.(ABSTRACT TRUNCATED AT 400 WORDS)     

Masculinity, femininity, type A behavior, and psychosocial adjustment in medical students

The freshman class of a midwestern medical school completed measures of masculinity and femininity, Type A behavior, and a variety of dependent variables concerning psychological well-being, adjustment, and interpersonal satisfaction. Appropriate statistical treatment of the data revealed strong and consistent masculinity effects on neuroticism, depression, self-esteem, confidence, hedonic capacity, locus of control, and relationship satisfaction. Femininity main effects varied in number as a function of the statistical method employed and involved a more diverse group of variables than is typically reported. Additive androgyny formulations of mental health were supported; balance androgyny formulations were not. No evidence for a Type A X Masculinity effect on adjustment was found. Discussion focuses on the correct interpretation of masculinity and femininity scales, comparability of analysis of variance (ANOVA) and multiple regression statistical analyses, and the viability of the concept of androgyny.     

Assessment of stress-buffering effects of social networks on psychological symptoms in an inner-city elderly population

Social network researchers have been divided into two camps: those who propose that social networks have a direct effect on subsequent psychological symptoms and those who posit a stress-buffering effect as well. Previous research has been limited by rudimentary measures of social interaction and the absence of longitudinal data as well as by different approaches to the assessment of possible buffering effects. In the present study, using 19 social network variables, the authors followed 133 elderly residents of mid-Manhattan SRO hotels for 1 year. Three different methods of determining buffering effects were examined: Dividing the sample into high- and low-stress groups and contrasting differences in percentage variance accounted for by social networks between the two groups; Examining the group as a whole to assess if any Network Variable X Stress interactional terms are significant; Examining the group as a whole to assess whether there is a reduction in the beta value of stress with respect to psychological symptoms when network variables are added to the analysis. Method 1 indicated a direct network effect, but none of the methods indicated a buffering effect. Of clinical relevance was the nonlinearity of the network effects, that is, depending upon a person's stressor level, different network dimensions must be emphasized and strengthened.     

On Enhancing the Cross‐Cultural Comparability of Likert‐Scale Personality and Value Measures: A Comparison of Common Procedures

This study aims to evaluate a number of procedures that have been proposed to enhance cross‐cultural comparability of personality and value data. A priori procedures (anchoring vignettes and direct measures of response styles (i.e. acquiescence, extremity, midpoint responding, and social desirability), a posteriori procedures focusing on data transformations prior to analysis (ipsatization and item parcelling), and two data modelling procedures (treating data as continuous vs as ordered categories) were compared using data collected from university students in 16 countries. We found that (i) anchoring vignettes showed lack of invariance, so they were not bias‐free; (ii) anchoring vignettes showed higher internal consistencies than raw scores where all other correction procedures, notably ipsatization, showed lower internal consistencies; (iii) in measurement invariance testing, no procedure yielded scalar invariance; anchoring vignettes and item parcelling slightly improved comparability, response style correction did not affect it, and ipsatization resulted in lower comparability; (iv) treating Likert‐scale data as categorical resulted in higher levels of comparability; (v) factor scores of scales extracted from different procedures showed similar correlational patterning; and (vi) response style correction was the only procedure that suggested improvement in external validity of country‐level conscientiousness. We conclude that, although no procedure resolves all comparability issues, anchoring vignettes, parcelling, and treating data as ordered categories seem promising to alleviate incomparability. We advise caution in uncritically applying any of these procedures. Copyright © 2017 European Association of Personality Psychology     

A Cross-Cultural Study of Response Biases in Personality Measures

The purposes of this study were to (a) compare the response bias tendencies of U.S. and Philippine college students and men and women in each culture when responding to personality measures, (b) examine the comparability of different measures of the same response biases, (c) examine the stability or consistency of response biases across instruments, (d) examine the extent to which controlling for response biases affects cultural mean comparisons in personality variables, and (e) test hypothesized personality correlates of response biases. The results did not support the presence of large cross-cultural or gender differences in response biases. Moderate to high agreement was found between different indexes of the same biases. Participants' response bias tendencies were moderately stable across instruments. Controlling for response biases led to trivial changes in effect sizes; in most cases, conclusions about cultural differences in personality constructs did not change. Most hypotheses relating personality variables to response biases were not supported.     

Stability of individual differences in personality characteristics from 3 to 15 years

Ratings of 23 personality variables on a longitudinal sample of 97 children at the ages of 3, 7, 11 and 15 yr had previously been shown to yield orthogonal factors of high continuity of structure across ages. Subjects' scores on these factors were employed to derive stability coefficients over these ages. The general level of obtained r's is modest, but is higher for measures of Extraversion-Introversion (EI) and other broad factors (Neuroticism (N), Melancholic-Sanguine (MS), and Choleric-Phlegmatic (CP)) than for narrower Varimax factors (EI, Melancholic and Activity), except in the case of Choleric. Complete regression factor scores proved more satisfactory than an inexact method. Global ratings of EI and N yielded similar r's to the factor measures.The effects of correcting for rater unreliability and possible values of ‘protocol’ unreliability are examined. Even in the case of a maximal correction, stability tends to be lower than is found for IQ data, suggesting greater instability of the underlying personality variables.Lawley's test of heterogeneity of correlation matrices provides only limited evidence of a simplex pattern across ages, and virtually none of a ‘sleeper effect’. Jennrich's test of differences in matrices provides no evidence of sex differences in stability.Results on stability of the specific variables are also reported. They show, with few exceptions, lower stability than factor variables. Jennrich's test yielded significant sex differences on only three specific variables.     

The Stability of Child Problem Behaviors: A Longitudinal Analysis of Inner-City African American Children

We focused on the stability of child problem behaviors in a sample of 124 low-income inner-city African American families. Internalizing and externalizing problems were assessed longitudinally across four years. Test-retest correlation coefficients indicated that the relative stability of both internalizing and externalizing problems over the four-year assessment was high for both child and mother reported variables. Partial support was obtained for absolute stability of child problem behavior as analyses of variance revealed that two of four variables of interest did not change significantly over time. Mother report of child problem behavior was more stable than child report, but gender of child or type of problem behavior (internalizing vs. externalizing) was not related to stability. Hierarchical multiple regression analyses revealed that the historical context of child problem behaviors is important to consider, as earlier problem behaviors accounted for unique variance in later problem behaviors, beyond that accounted for the most recent assessment.     

Direct,mediated and moderated impacts of personality variables on smoking initiation in adolescents

Role of ‘Big Five’ personality traits as predictors of smoking and moderators of the intention-smoking relationship was tested. Five hundred and fifty-three adolescents (aged 11–12) completed measures of self-reported past smoking, gender, intentions to smoke, perceived behavioural control, family smoking, friends smoking at times 1 and 2 (6 months apart). At time 3, 2 years later, the same adolescents completed measures of the Big Five and self-reported smoking (a subset of 300 also provided an objective smoking measure). At time 4, two years after time 3, a sub-sample of 122 adolescents provided a self-report measure of recent smoking. Simple correlations indicated significant direct effects of conscientiousness (self-reported smoking, times 3 and 4), extraversion (time 4 smoking) and neuroticism (all smoking measures) on smoking. Logistic regression showed intention, and the interaction between conscientiousness and intention to significantly predict both self-reported and objectively assessed smoking (both at time 3) after controlling for other variables. Multiple regression showed intentions, family smoking and the interaction between conscientiousness and intention to significantly predict self-reported smoking at time 4 after controlling for other variables. The findings indicate that the impact of personality variables on smoking is through mediated (through cognitions) and moderated (conscientiousness by intention interaction) pathways.     

Callous-Unemotional Traits are Uniquely Associated with Poorer Peer Functioning in School-Aged Children

This study examines externalizing symptoms (attention-deficit/hyperactivity disorder [ADHD], conduct problems, and callous-unemotional [CU] traits) in relation to domains of peer functioning (social competence, loneliness, and close friendship quality), with a specific focus on the role of CU traits. One hundred twenty-four elementary students (grades 3–6; 45% boys) completed multiple measures of peer functioning, and teachers completed measures of externalizing symptoms and social competence. After controlling for demographic variables and other externalizing symptoms, CU traits were significantly associated with poorer peer functioning across all variables except for demands of exclusivity in close friendships. ADHD symptoms were also uniquely associated with poorer social functioning across a number of variables. In contrast, conduct problems were at times associated with better social functioning after controlling for the effects of other externalizing problems. These findings bolster the importance of developing and evaluating social skills interventions for children displaying elevated CU traits.     

A computer program for automated scoring of exteroceptive suppression periods

Exteroceptive suppression (ES) of temporalis muscle activity has received recent attention as a method of investigating central mechanisms in the pathogenesis of chronic headache disorders. Unfortunately, the resolution of conflicting findings has been hampered by the use of different assessment and scoring methodologies across laboratories, which has limited the direct comparison of findings across studies. A computer program was developed with the goal of providing an automated and reliable method of scoring ES responses. Processing an ES response with this software provides both visual and quantitative information relating to various response parameters. Multiple levels of smoothing the raw waveform data are available, and the software offers scoring on the basis of multiple depth-of-suppression criteria. The program also generates values for a variety of ES parameters: latency, duration, area, maximum level of suppression with its corresponding time (msec), and baseline level of muscle activity. It is likely that a reliable and standardized scoring protocol would enhance comparability of ES data generated across different settings.     

Multicollinearity and the use of regression analyses in discrimination litigation

Multicollinearity is a problem that can adversely affect the estimation of coefficients in regression equations. The types of regression models used as evidence in employment discrimination cases may be particularly susceptible to estimation problems resulting from multicollinearity, yet courts have in most instances failed to address this difficulty. This paper discusses the problems associated with multi- collinearity and suggests possible sources of the multicolli- nearity, particularly in the context of employment discrimination litigation. An academic example demonstrates how multicollinearity can be identified and corrected. Other possible techniques for curing the multicollinearity or reducing its effects are also presented. The related problem of “tainted” variables, often discussed by courts in employment discrimination litigation, is differentiated from the multicollinearity problem. For regression equations to be reliable and for inferences regarding discrimination to be accurately made, courts should concern themselves with multicollinearity, and not so-called “tainting.”     

Comparing the Characteristics of Homeless Adults in Poland and the United States

This study compared the characteristics of probability samples of homeless adults in Poland (N = 200 from two cities) and the United States (N = 219 from one city), using measures with established reliability and validity in homeless populations. The same measures were used across nations and a systemic translation procedure assured comparability of measurement. The two samples were similar on some measures: In both nations, most homeless adults were male, many reported having dependent children and experiencing out-of-home placements when they themselves were children, and high levels of physical health problems were observed. Significant national differences were also found: Those in Poland were older, had been homeless for longer, showed lower rates on all psychiatric diagnoses assessed (including severe mental and substance abuse disorders), reported less contact with family and supportive network members, were less satisfied when they sought support from their networks, and reported fewer recent stressful life events and fewer risky sexual behaviors. Culturally-informed interpretations of these findings and their implications are presented.