Reducing Attenuation Bias in Regression Analyses Involving Rating Scale Data via Psychometric Modeling.

Many studies in fields such as psychology and educational sciences obtain information about attributes of subjects through observational studies, in which raters score subjects using multiple-item rating scales. Error variance due to measurement effects, such as items and raters, attenuate the regression coefficients and lower the power of (hierarchical) linear models. A modeling procedure is discussed to reduce the attenuation. The procedure consists of (1) an item response theory (IRT) model to map the discrete item responses to a continuous latent scale and (2) a generalizability theory (GT) model to separate the variance in the latent measurement into variance components of interest and nuisance variance components. It will be shown how measurements obtained from this mixture of IRT and GT models can be embedded in (hierarchical) linear models, both as predictor or criterion variables, such that error variance due to nuisance effects are partialled out. Using examples from the field of educational measurement, it is shown how general-purpose software can be used to implement the modeling procedure.

Intervention-induced temperament changes in children: Evidence from a randomized controlled trial of the Incredible Years parent program.

Child temperament has long been viewed as a potential susceptibility factor in the link between parenting and child disruptive behavior (CDB). Specifically, the idea is that children with higher negative emotionality, surgency, and lower effortful control are more affected by their received parenting, but experimental evidence is scarce. Also, others have argued that child temperament might not be a susceptibility factor but a factor that can change through parents' participation in a parenting intervention. To test both hypotheses, we analyzed pretest, posttest, and 4-month follow-up data from 386 mostly Dutch parents, mainly mothers (92%; Mage = 38.1, SD = 4.8) with children (Mage = 6.31, SD = 1.33; 54.2% boys). The children had above-average disruptive behavior (i.e., ≥75th percentile Eyberg Child Behavior Inventory questionnaire; Eyberg & Pincus, 1999). The families participated in a randomized controlled trial of the Incredible Years (IY) parenting program. Hierarchical regression analyses showed that child temperament did not moderate IY intervention effects on CDB. Furthermore, parallel process analyses showed that the IY intervention led to direct, simultaneous decreases in both negative emotionality and CDB. These findings counter the widely held belief that temperament traits are static, unchangeable modulators of the links between parenting and CDB. Instead, child temperament (negative emotionality) can at least partly be influenced by parents' participation in a parenting program. (PsycInfo Database Record (c) 2023 APA, all rights reserved).

Parental socialization of guilt and shame in early childhood.

Self-conscious emotions emerge early in human development and they help children navigate social relationships. Little is known about the socialization of self-conscious emotions in early childhood. We theorized that parental mental state language use and warmth would be important for young children's self-conscious emotions and their consequent prosocial behaviors. Ninety-eight children residing in the Netherlands (52% girls) aged 2-5 (M = 48.66 months, SD = 13.50 months) visited the research lab with one parent. First, we observed parental mental state language and warmth. Afterward, children were led to believe that they caused a mishap (i.e., accidentally breaking the experimenter's favorite toy) to evoke their guilt and shame, which we micro-coded. In subsequent tasks, we observed children's helping behaviors toward the experimenter. We found that the combination of frequent parental mental state language and high warmth was associated with children's quicker helping to the previously harmed experimenter across toddlerhood and early childhood. More guilt was related to more helping whereas more shame-like avoidance was related to less helping. Our findings based on the sample of Dutch parents and children suggest that, parental frequent mental state talk, in combination with high warmth, may promote children's ability to repair social relationships and behave prosocially after mishaps.

Autistic traits and self-conscious emotions in early childhood.

Self-conscious emotions arise from evaluating the self through the eyes of others. Given that children with autistic traits may experience difficulties with understanding others' minds, they might show less attuned self-conscious emotions. Two-to-five-year-old children's (N = 98, Mage  = 48.54 months, 50% girls, 92% White) self-conscious emotions (guilt, embarrassment, and shame-like avoidance) were observed after children "broke" the experimenter's favorite toy. Data were collected from March 2018 till June 2019. Children with more autistic traits showed less theory of mind (ToM), and more shame-like avoidance, but associations were not mediated by ToM. This provides initial evidence that children with more autistic traits may show disturbances in some but not all self-conscious emotions, which could hinder their social functioning.

Power analysis for conditional indirect effects: A tutorial for conducting Monte Carlo simulations with categorical exogenous variables.

Conceptual and statistical models that include conditional indirect effects (i.e., so-called "moderated mediation" models) are increasingly popular in the behavioral sciences. Although there is ample guidance in the literature for how to specify and test such models, there is scant advice regarding how to best design studies for such purposes, and this especially includes techniques for sample size planning (i.e., "power analysis"). In this paper, we discuss challenges in sample size planning for moderated mediation models and offer a tutorial for conducting Monte Carlo simulations in the specific case where one has categorical exogenous variables. Such a scenario is commonly faced when one is considering testing conditional indirect effects in experimental research, wherein the (assumed) predictor and moderator variables are manipulated factors and the (assumed) mediator and outcome variables are observed/measured variables. To support this effort, we offer example data and reproducible R code that constitutes a "toolkit" to make up for limitations in other software and aid researchers in the design of research to test moderated mediation models.

Updated guidelines on selecting an intraclass correlation coefficient for interrater reliability, with applications to incomplete observational designs.

Several intraclass correlation coefficients (ICCs) are available to assess the interrater reliability (IRR) of observational measurements. Selecting an ICC is complicated, and existing guidelines have three major limitations. First, they do not discuss incomplete designs, in which raters partially vary across subjects. Second, they provide no coherent perspective on the error variance in an ICC, clouding the choice between the available coefficients. Third, the distinction between fixed or random raters is often misunderstood. Based on generalizability theory (GT), we provide updated guidelines on selecting an ICC for IRR, which are applicable to both complete and incomplete observational designs. We challenge conventional wisdom about ICCs for IRR by claiming that raters should seldom (if ever) be considered fixed. Also, we clarify how to interpret ICCs in the case of unbalanced and incomplete designs. We explain four choices a researcher needs to make when selecting an ICC for IRR, and guide researchers through these choices by means of a flowchart, which we apply to three empirical examples from clinical and developmental domains. In the Discussion, we provide guidance in reporting, interpreting, and estimating ICCs, and propose future directions for research into the ICCs for IRR. (PsycInfo Database Record (c) 2023 APA, all rights reserved).

Assessing measurement invariance with moderated nonlinear factor analysis using the R package OpenMx.

Assessing measurement invariance is an important step in establishing a meaningful comparison of measurements of a latent construct across individuals or groups. Most recently, moderated nonlinear factor analysis (MNLFA) has been proposed as a method to assess measurement invariance. In MNLFA models, measurement invariance is examined in a single-group confirmatory factor analysis model by means of parameter moderation. The advantages of MNLFA over other methods is that it (a) accommodates the assessment of measurement invariance across multiple continuous and categorical background variables and (b) accounts for heteroskedasticity by allowing the factor and residual variances to differ as a function of the background variables. In this article, we aim to make MNLFA more accessible to researchers without access to commercial structural equation modeling software by demonstrating how this method can be applied with the open-source R package OpenMx. (PsycInfo Database Record (c) 2023 APA, all rights reserved).

Interrater reliability for multilevel data: A generalizability theory approach.

Current interrater reliability (IRR) coefficients ignore the nested structure of multilevel observational data, resulting in biased estimates of both subject- and cluster-level IRR. We used generalizability theory to provide a conceptualization and estimation method for IRR of continuous multilevel observational data. We explain how generalizability theory decomposes the variance of multilevel observational data into subject-, cluster-, and rater-related components, which can be estimated using Markov chain Monte Carlo (MCMC) estimation. We explain how IRR coefficients for each level can be derived from these variance components, and how they can be estimated as intraclass correlation coefficients (ICC). We assessed the quality of MCMC point and interval estimates with a simulation study, and showed that small numbers of raters were the main source of bias and inefficiency of the ICCs. In a follow-up simulation, we showed that a planned missing data design can diminish most estimation difficulties in these conditions, yielding a useful approach to estimating multilevel interrater reliability for most social and behavioral research. We illustrated the method using data on student-teacher relationships. All software code and data used for this article is available on the Open Science Framework: https://osf.io/bwk5t/. (PsycInfo Database Record (c) 2022 APA, all rights reserved).

Analytical power calculations for structural equation modeling: A tutorial and Shiny app.

Conducting a power analysis can be challenging for researchers who plan to analyze their data using structural equation models (SEMs), particularly when Monte Carlo methods are used to obtain power. In this tutorial, we explain how power calculations without Monte Carlo methods for the χ2 test and the RMSEA tests of (not-)close fit can be conducted using the Shiny app "power4SEM". power4SEM facilitates power calculations for SEM using two methods that are not computationally intensive and that focus on model fit instead of the statistical significance of (functions of) parameters. These are the method proposed by Satorra and Saris (Psychometrika 50(1), 83-90, 1985) for power calculations of the likelihood ratio test, and that described by MacCallum, Browne, and Sugawara (Psychol Methods 1(2) 130-149, 1996) for RMSEA-based power calculations. We illustrate the use of power4SEM with examples of power analyses for path models, factor models, and a latent growth model.

Corrigendum: Determinants of Non-paid Task Division in Gay-, Lesbian-, and Heterosexual-Parent Families With Infants Conceived Using Artificial Reproductive Techniques.

[This corrects the article DOI: 10.3389/fpsyg.2020.00914.].

Dopamine Functioning and Child Externalizing Behavior: A Longitudinal Analysis of Polygenic Susceptibility to Parenting.

OBJECTIVE: This study investigated how genetic susceptibility may affect children's sensitivity to parenting practices in their development of externalizing behavior. We created a continuous polygenic index composed of 5 dopamine polymorphisms to investigate the moderating role of dopamine-related genes in shaping parent-child gene-by-environment (Gc×E) interactions. Accumulating research supports that differences in children's dopamine neurotransmission make certain children more susceptible to both negative and positive parenting practices, a "for-better and for-worse" effect. METHODS: Data from a 3-wave longitudinal study (4 months between waves) on 190 at-risk families with children aged 4 to 8 were used to investigate whether a heightened polygenic index score amplified the associations between negative and positive parenting and later children's externalizing behavior. Parenting practices and child externalizing behavior were assessed by parent-reported questionnaires. RESULTS: Findings were not in line with the expectation that there was a stronger association between positive and negative parenting and later externalizing behavior for children with higher scores on the polygenic susceptibility index. Rather, children with a lower score on the polygenic susceptibility index showed more later externalizing behavior in response to positive parenting behavior, whereas for children with a higher score on the polygenic index, positive parenting was predictive of relatively lower levels of later child externalizing behavior. CONCLUSION: The results indicate that not only are children with higher but also lower scores on the polygenic index sensitive to parenting, they suggest that different phenotypical characteristics related to reward processing might underlie these genetic susceptibilities to parenting practices.

Determinants of Non-paid Task Division in Gay-, Lesbian-, and Heterosexual-Parent Families With Infants Conceived Using Artificial Reproductive Techniques.

Background: The division of non-paid labor in heterosexual parents in the West is usually still gender-based, with mothers taking on the majority of direct caregiving responsibilities. However, in same-sex couples, gender cannot be the deciding factor. Inspired by Feinberg's ecological model of co-parenting, this study investigated whether infant temperament, parent factors (biological relatedness to child, psychological adjustment, parenting stress, and work status), and partner relationship quality explained how first-time gay, lesbian, and heterosexual parents divided labor (childcare and family decision-making) when their infants were 4 and 12 months old. We also tested whether family type acted as a moderator. Method: Participants were drawn from the new parents study. Only those who provided information about their biological relatedness to their child (N = 263 parents) were included. When infants were 4 months (T1), parents completed a password-protected online questionnaire exploring their demographic characteristics including work status and standardized online-questionnaires on task division (childcare and family decision-making), infant temperament, parental anxiety, parental depression, parental stress, and partner relationship satisfaction. When infants were 12-months-old (T2), parents provided information about task division and their biological relatedness to their children. Results: Linear mixed models showed that no factor explained the division of family decision making at T1 and T2. For relative time spent on childcare tasks at T1, biological relatedness mattered for lesbian mothers only: biologically related mothers appeared to spend more time on childcare tasks than did non-related mothers. Results showed that, regardless of family type, parents who were not working or were working part-time at T1 performed more childcare tasks at T1. This was still true at T2. The other factors did not significantly contribute to relative time spent on childcare tasks at T2. Conclusion: We had the opportunity to analyze the division of non-paid tasks in families where parenting was necessarily planned and in which gender could not affect that division. Although Feinberg's model of co-parenting suggests that various factors are related to task division, we found that paid work outside the home was most important during the first year of parenthood in determining caregiving roles.

Adapting fit indices for Bayesian structural equation modeling: Comparison to maximum likelihood.

In a frequentist framework, the exact fit of a structural equation model (SEM) is typically evaluated with the chi-square test and at least one index of approximate fit. Current Bayesian SEM (BSEM) software provides one measure of overall fit: the posterior predictive p value (PPP χ2 ). Because of the noted limitations of PPP χ2 , common practice for evaluating Bayesian model fit instead focuses on model comparison, using information criteria or Bayes factors. Fit indices developed under maximum-likelihood estimation have not been incorporated into software for BSEM. We propose adapting 7 chi-square-based approximate fit indices for BSEM, using a Bayesian analog of the chi-square model-fit statistic. Simulation results show that the sampling distributions of the posterior means of these fit indices are similar to their frequentist counterparts across sample sizes, model types, and levels of misspecification when BSEMs are estimated with noninformative priors. The proposed fit indices therefore allow overall model-fit evaluation using familiar metrics of the original indices, with an accompanying interval to quantify their uncertainty. Illustrative examples with real data raise some important issues about the proposed fit indices' application to models specified with informative priors, when Bayesian and frequentist estimation methods might not yield similar results. (PsycINFO Database Record (c) 2020 APA, all rights reserved).

A multiple imputation score test for model modification in structural equation models.

Structural equation modeling (SEM) applications routinely employ a trilogy of significance tests that includes the likelihood ratio test, Wald test, and score test or modification index. Researchers use these tests to assess global model fit, evaluate whether individual estimates differ from zero, and identify potential sources of local misfit, respectively. This full cadre of significance testing options is not yet available for multiply imputed data sets, as methodologists have yet to develop a general score test for this context. Thus, the goal of this article is to outline a new score test for multiply imputed data. Consistent with its complete-data counterpart, this imputation-based score test provides an estimate of the familiar expected parameter change statistic. The new procedure is available in the R package semTools and naturally suited for identifying local misfit in SEM applications (i.e., a model modification index). The article uses a simulation study to assess the performance (Type I error rate, power) of the proposed score test relative to the score test produced by full information maximum likelihood (FIML) estimation. Due to the two-stage nature of multiple imputation, the score test exhibited slightly lower power than the corresponding FIML statistic in some situations but was generally well calibrated. (PsycInfo Database Record (c) 2020 APA, all rights reserved).

Self-Reports and Caregivers' Proxy Reports of Unmet Needs of Persons With Dementia: Implications for Both Partners' Health-Related Quality of Life.

OBJECTIVE: This study examined in a large sample of dementia caregiving dyads the associations between both partners' reports of unmet needs in persons with dementia (PwDs) and both partners' health-related quality of life (HRQOL). METHODS: This was a cross-sectional self-report survey of 521 community-dwelling dyads in a pragmatic trial in the Netherlands. The Camberwell Needs Assessment was used to measure PwDs' unmet needs. Both partners' self-reported their HRQOL using the EuroQol-5. RESULTS: Controlling for covariates, PwDs' self-reported greater unmet needs were significantly associated with PwDs' and caregivers' lower self-reported HRQOL (actor effect; b = -0.044, ß = -0.226, z = -3.588, p <0.001 and partner effect; b = -0.021, ß = -0.131, z = -2.154, p = 0.031). Caregivers' proxy reports were greater than PwDs' self-reported unmet needs (Δ=0.66,χ2(1)=55.881,p<.0001). CONCLUSION: Clinicians should use caution in relying on caregiver proxy reports of PwDs' needs and HQOL alone regarding healthcare decision making.

Using restricted factor analysis to select anchor items and detect differential item functioning.

Restricted factor analysis (RFA) is a powerful method to test for uniform differential item functioning (DIF), but it may require empirically selecting anchor items to prevent inflated Type I error rates. We conducted a simulation study to compare two empirical anchor-selection strategies: a one-step rank-based strategy and an iterative selection procedure. Unlike the iterative procedure, the rank-based strategy had a low risk and degree of contamination within the empirically selected anchor set, even with small samples. To detect nonuniform DIF, RFA requires an interaction effect with the latent factor. The latent moderated structural equations (LMS) method has been applied to RFA and has revealed inflated Type I error rates. We propose using product indicators (PI) as a more widely available alternative to measure the latent interaction. A simulation study, involving several sample-size conditions and magnitudes of uniform and nonuniform DIF, revealed that PI obtained similar power but lower Type I error rates, as compared to LMS.

Exploring the gold-standard: Evidence for a two-factor model of the Clinician Administered PTSD Scale for the DSM-5.

OBJECTIVE: The latent factor structure of posttraumatic stress disorder (PTSD) remains a source of considerable variability. The current study compared several a priori factor structures, as well as a novel 2-factor structure of posttraumatic psychological distress as measured by the Clinician Administered PTSD scale for the DSM-5 (CAPS-5). In addition, variability in diagnostic rates according to the divergent DSM-5 and ICD-11 criteria were explored. METHOD: The setting for this study was a Level 1 trauma center in a U.S. metropolitan city. Data were pooled from 2 studies (N = 309) and participants were administered the CAPS-5 at 1 (n = 139) or 6 months postinjury (n = 170). Confirmatory factor analysis (CFA) was used to compare several factor models, and prevalence rates based on DSM-5 and ICD-11 criteria were compared via z tests and kappa. RESULTS: CFAs of 5 factor structures indicated good fit for all models. A novel 2-factor model based on competing models of PTSD symptoms and modification indices was then tested. The 2-factor model of the CAPS-5 performed as well or better on most indices compared to a 7-factor hybrid. Comparisons of PTSD prevalence rates found no significant differences, but agreement was variable. CONCLUSIONS: These findings indicate that the CAPS-5 can be seen as measuring 2 distinct phenomena: posttraumatic stress disorder and general posttraumatic dysphoria. This is an important contribution to the current debate on which latent factors constitute PTSD and may reduce discordance. (PsycINFO Database Record

Permutation randomization methods for testing measurement equivalence and detecting differential item functioning in multiple-group confirmatory factor analysis.

In multigroup factor analysis, different levels of measurement invariance are accepted as tenable when researchers observe a nonsignificant (Δ)χ2 test after imposing certain equality constraints across groups. Large samples yield high power to detect negligible misspecifications, so many researchers prefer alternative fit indices (AFIs). Fixed cutoffs have been proposed for evaluating the effect of invariance constraints on change in AFIs (e.g., Chen, 2007; Cheung & Rensvold, 2002; Meade, Johnson, & Braddy, 2008). We demonstrate that all of these cutoffs have inconsistent Type I error rates. As a solution, we propose replacing χ2 and fixed AFI cutoffs with permutation tests. Randomly permuting group assignment results in average between-groups differences of zero, so iterative permutation yields an empirical distribution of any fit measure under the null hypothesis of invariance across groups. Our simulations show that the permutation test of configural invariance controls Type I error rates better than χ2 or AFIs when the model contains parsimony error (i.e., negligible misspecification) but the factor structure is equivalent across groups (i.e., the null hypothesis is true). For testing metric and scalar invariance, Δχ2 and permutation yield similar power and nominal Type I error rates, whereas ΔAFIs yield inflated errors in smaller samples. Permuting the maximum modification index among equality constraints control familywise Type I error rates when testing multiple indicators for lack of invariance, but provide similar power as using a Bonferroni adjustment. An applied example and syntax for software are provided. (PsycINFO Database Record (c) 2018 APA, all rights reserved).