RESUMEN
This review summarizes the current state of the art of statistical and (survey) methodological research on measurement (non)invariance, which is considered a core challenge for the comparative social sciences. After outlining the historical roots, conceptual details, and standard procedures for measurement invariance testing, the paper focuses in particular on the statistical developments that have been achieved in the last 10 years. These include Bayesian approximate measurement invariance, the alignment method, measurement invariance testing within the multilevel modeling framework, mixture multigroup factor analysis, the measurement invariance explorer, and the response shift-true change decomposition approach. Furthermore, the contribution of survey methodological research to the construction of invariant measurement instruments is explicitly addressed and highlighted, including the issues of design decisions, pretesting, scale adoption, and translation. The paper ends with an outlook on future research perspectives.
Asunto(s)
Proyectos de Investigación , Ciencias Sociales , Humanos , Teorema de Bayes , Encuestas y Cuestionarios , Análisis FactorialRESUMEN
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.
Asunto(s)
Aplicaciones Móviles , Humanos , Análisis de Clases Latentes , Funciones de Verosimilitud , Modelos Estadísticos , Método de Montecarlo , Proyectos de InvestigaciónRESUMEN
Meta-analytic structural equation modeling (MASEM) is increasingly applied to advance theories by synthesizing existing findings. MASEM essentially consists of two stages. In Stage 1, a pooled correlation matrix is estimated based on the reported correlation coefficients in the individual studies. In Stage 2, a structural model (such as a path model) is fitted to explain the pooled correlations. Frequently, the individual studies do not provide all the correlation coefficients between the research variables. In this study, we modify the currently optimal MASEM-method to deal with missing correlation coefficients, and compare its performance with existing methods. This study is the first to evaluate the performance of fixed-effects MASEM methods under different levels of missing correlation coefficients. We found that the often used univariate methods performed very poorly, while the multivariate methods performed well overall.
Asunto(s)
Algoritmos , Interpretación Estadística de Datos , Modelos Estadísticos , Humanos , Metaanálisis como AsuntoRESUMEN
Meta-analytic structural equation modeling (MASEM) is a statistical technique to pool correlation matrices and test structural equation models on the pooled correlation matrix. In Stage 1 of MASEM, correlation matrices from independent studies are combined to obtain a pooled correlation matrix, using fixed- or random-effects analysis. In Stage 2, a structural model is fitted to the pooled correlation matrix. Researchers applying MASEM may have hypotheses about how certain model parameters will differ across subgroups of studies. These moderator hypotheses are often addressed using suboptimal methods. The aim of the current article is to provide guidance and examples on how to test hypotheses about group differences in specific model parameters in MASEM. We illustrate the procedure using both fixed- and random-effects subgroup analysis with two real datasets. In addition, we present a small simulation study to evaluate the effect of the number of studies per subgroup on convergence problems. All data and the R-scripts for the examples are provided online.
Asunto(s)
Investigación Conductal , Análisis de Clases Latentes , Metaanálisis como Asunto , Investigación Conductal/métodos , Investigación Conductal/estadística & datos numéricos , Correlación de Datos , Humanos , Proyectos de InvestigaciónRESUMEN
BACKGROUND: Cross-national differences could affect the likelihood of endorsement of DSM cannabis abuse and dependence criteria. The present study examines whether cannabis abuse and dependence criteria function differently across U.S. and Dutch cannabis users. METHOD: Data on lifetime endorsement of DSM-IV cannabis abuse/dependence criteria were utilized from U.S. cannabis users who participated in the National Epidemiological Survey on Alcohol and Related Conditions (NESARC) and from Dutch cannabis users who participated in the Zuid-Holland study. In total, 1,568 cannabis users participated in the NESARC sample, and 359 cannabis users participated in the Zuid-Holland sample. The DSM-IV cannabis abuse/dependence criteria as well as cannabis withdrawal were determined using face-to-face computer-assisted personal interviews. RESULTS: Using Restricted Factor Analysis with Latent Moderated Structures, the cannabis abuse/dependence criteria legal problems (ß = -0.43), failed quit attempts (ß = -1.09), use despite problems (ß = -0.32), and withdrawal (ß = -0.53) showed measurement bias, and were more likely to be endorsed by U.S. than by Dutch cannabis users. Also, men were more likely than women to endorse the criteria hazardous use (ß = -0.27), legal problems (ß = -0.49) and tolerance (ß = -0.20). Findings on failed quit attempts and withdrawal were replicated in matched subsamples, while results on legal problems (country and gender) were partly replicated. CONCLUSIONS: Several CUD criteria showed measurement bias across two countries and between males and females. Therefore, differences between countries and gender in prevalence rates of CUD should be regarded with caution.
Asunto(s)
Abuso de Marihuana/diagnóstico , Fumar Marihuana , Modelos Psicológicos , Adolescente , Adulto , Manual Diagnóstico y Estadístico de los Trastornos Mentales , Femenino , Humanos , Masculino , Países Bajos , Prevalencia , Estados Unidos , Adulto JovenRESUMEN
The test for cluster bias is a test of measurement invariance across clusters in 2-level data. This article examines the true positive rates (empirical power) and false positive rates of the test for cluster bias using the likelihood ratio test (LRT) and the Wald test with ordinal data. A simulation study indicates that the scaled version of the LRT that accounts for nonnormality of the data gives untrustworthy results, whereas the unscaled LRT and the Wald test have acceptable false positive rates and perform well in terms of empirical power rate if the amount of cluster bias is large. The test for cluster bias is illustrated with data from research on teacher-student relations.
RESUMEN
Researchers may have at their disposal the raw data of the studies they wish to meta-analyze. The goal of this study is to identify, illustrate, and compare a range of possible analysis options for researchers to whom raw data are available, wanting to fit a structural equation model (SEM) to these data. This study illustrates techniques that directly analyze the raw data, such as multilevel and multigroup SEM, and techniques based on summary statistics, such as correlation-based meta-analytical structural equation modeling (MASEM), discussing differences in procedures, capabilities, and outcomes. This is done by analyzing a previously published collection of datasets using open source software. A path model reflecting the theory of planned behavior is fitted to these datasets using different techniques involving SEM. Apart from differences in handling of missing data, the ability to include study-level moderators, and conceptualization of heterogeneity, results show differences in parameter estimates and standard errors across methods. Further research is needed to properly formulate guidelines for applied researchers looking to conduct individual participant data MASEM.
RESUMEN
Background: The world's population is aging, but life expectancy has risen more than healthy life expectancy (HALE). With respect to brain and cognition, the prevalence of neurodegenerative disorders increases with age, affecting health and quality of life, and imposing significant healthcare costs. Although the effects of physical exercise on cognition in advanced age have been widely explored, in-depth fundamental knowledge of the underlying mechanisms of the exercise-induced cognitive improvements is lacking. Recent research suggests that myokines, factors released into the blood circulation by contracting skeletal muscle, may play a role in mediating the beneficial effect of exercise on cognition. Our goal in this ongoing (living) review is to continuously map the rapidly accumulating knowledge on pathways between acute or chronic exercise-induced myokines and cognitive domains enhanced by exercise. Method: Randomized controlled studies will be systematically collected at baseline and every 6 months for at least 5 years. Literature search will be performed online in PubMed, EMBASE, PsycINFO, Web of Science, SportDiscus, LILACS, IBECS, CINAHL, SCOPUS, ICTRP, and ClinicalTrials.gov. Risk of bias will be assessed using the Revised Cochrane Risk of Bias tool (ROB 2). A random effects meta-analysis with mediation analysis using meta-analytic structural equation modeling (MASEM) will be performed. The primary research question is to what extent exercise-induced myokines serve as mediators of cognitive function. Secondarily, the pooled effect size of specific exercise characteristics (e.g., mode of exercise) or specific older adults' populations (e.g., cognitively impaired) on the relationship between exercise, myokines, and cognition will be assessed. The review protocol was registered in PROSPERO (CRD42023416996). Discussion: Understanding the triad relationship between exercise, myokines and cognition will expand the knowledge on multiple integrated network systems communicating between skeletal muscles and other organs such as the brain, thus mediating the beneficial effects of exercise on health and performance. It may also have practical implications, e.g., if a certain myokine is found to be a mediator between exercise and cognition, the optimal exercise characteristics for inducing this myokine can be prescribed. The living review is expected to improve our state of knowledge and refine exercise regimes for enhancing cognitive functioning in diverse older adults' populations. Registration: Systematic review and meta-analysis protocol was registered with the International Prospective Register of Systematic Reviews (PROSPERO) on the 24th of April 2023 (registration number CRD42023416996).
RESUMEN
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).
RESUMEN
This mixed methods systematic review and meta-analysis sheds more light on the role parenting practices play in children's adjustment after war exposure. Specifically, we quantitatively examined whether parenting behavior explained some of the well-known associations between war exposure and children's adjustment. In addition, we meta-synthesized qualitative evidence answering when and why parenting practices might change for war-affected families. We searched nine electronic databases and contacted experts in the field for relevant studies published until March 2018, identifying 4,147 unique publications that were further screened by title and abstract, resulting in 158 publications being fully screened. By running a meta-analytic structural equation model with 38 quantitative studies (N = 54,372, Mage = 12.00, SDage = 3.54), we found that more war-exposed parents showed less warmth and more harshness toward their children, which partly mediated the association between war exposure and child adjustment, that is, post-traumatic stress symptoms, depression and anxiety, social problems, externalizing behavior, and lower positive outcomes (e.g., quality of life). War exposure was not associated with parents' exercise of behavioral control. By meta-synthesizing 10 qualitative studies (N = 1,042; age range = 0-18), we found that the nature of war-related trauma affected parenting differently. That is, parents showed harshness, hostility, inconsistency, and less warmth in highly dangerous settings and more warmth and overprotection when only living under threat. We conclude that it is both how much and what families have seen that shapes parenting in times of war.
Asunto(s)
Responsabilidad Parental , Adolescente , Niño , Crianza del Niño , Preescolar , Humanos , Lactante , Recién Nacido , Padres , Calidad de Vida , Exposición a la GuerraRESUMEN
Meta-analytic structural equation modeling (MASEM) refers to fitting structural equation models (SEMs) (such as path models or factor models) to meta-analytic data. Currently, fitting MASEMs may be challenging for researchers that are not accustomed to working with R software and packages. Therefore, we developed webMASEM; a web application for MASEM. This app implements the one-stage MASEM approach, and allows users to apply MASEM in a user-friendly way. The aim of this article is to provide a tutorial on one-stage MASEM and a practical guide to webMASEM. We will pay specific attention to how the data should be structured and prepared for webMASEM, because mistakes in this step may lead to faulty results without receiving an error message. The use of webMASEM is illustrated with an analysis of a meta-analytic path model in which the path coefficients are moderated by a study-level variable, a meta-analytic factor model in which the factor loadings are moderated by a study-level variable, and a meta-analytic panel model in which the effects are moderated by a study-level variable. All used datafiles and R scripts are available online.
Asunto(s)
Modelos Estadísticos , Modelos Teóricos , Humanos , Análisis de Clases Latentes , Investigadores , Programas InformáticosRESUMEN
Meta-analytic structural equation modeling (MASEM) is an increasingly popular meta-analytic technique that combines the strengths of meta-analysis and structural equation modeling. MASEM facilitates the evaluation of complete theoretical models (e.g., path models or factor analytic models), accounts for sampling covariance between effect sizes, and provides measures of overall fit of the hypothesized model on meta-analytic data. We propose a novel MASEM method, one-stage MASEM, which is better suitable to explain study-level heterogeneity than existing methods. One-stage MASEM allows researchers to incorporate continuous or categorical moderators into the MASEM, in which any parameter in the structural equation model (e.g., path coefficients and factor loadings) can be modeled by the moderator variable, while the method does not require complete data for the primary studies included in the meta-analysis. We illustrate the new method on two real data sets, evaluate its empirical performance via a computer simulation study, and provide user-friendly R-functions and annotated syntax to assist researchers in applying one-stage MASEM. We close the article by presenting several future research directions. (PsycInfo Database Record (c) 2020 APA, all rights reserved).
Asunto(s)
Análisis de Clases Latentes , Metaanálisis como Asunto , Modelos Estadísticos , Psicología/métodos , Simulación por Computador , HumanosRESUMEN
Data often have a nested, multilevel structure, for example when data are collected from children in classrooms. This kind of data complicate the evaluation of reliability and measurement invariance, because several properties can be evaluated at both the individual level and the cluster level, as well as across levels. For example, cross-level invariance implies equal factor loadings across levels, which is needed to give latent variables at the two levels a similar interpretation. Reliability at a specific level refers to the ratio of true score variance over total variance at that level. This paper aims to shine light on the relation between reliability, cross-level invariance, and strong factorial invariance across clusters in multilevel data. Specifically, we will illustrate how strong factorial invariance across clusters implies cross-level invariance and perfect reliability at the between level in multilevel factor models.
RESUMEN
Big data is a field that has traditionally been dominated by disciplines such as computer science and business, where mainly data-driven analyses have been performed. Psychology, a discipline in which a strong emphasis is placed on behavioral theories and empirical research, has the potential to contribute greatly to the big data movement. However, one challenge to psychologists-and probably the most crucial one-is that most researchers may not have the necessary programming and computational skills to analyze big data. In this study we argue that psychologists can also conduct big data research and that, rather than trying to acquire new programming and computational skills, they should focus on their strengths, such as performing psychometric analyses and testing theories using multivariate analyses to explain phenomena. We propose a split/analyze/meta-analyze approach that allows psychologists to easily analyze big data. Two real datasets are used to demonstrate the proposed procedures in R. A new research agenda related to the analysis of big data in psychology is outlined at the end of the study.
RESUMEN
Meta-analytic structural equation modeling (MASEM) involves fitting models to a common population correlation matrix that is estimated on the basis of correlation coefficients that are reported by a number of independent studies. MASEM typically consist of two stages. The method that has been found to perform best in terms of statistical properties is the two-stage structural equation modeling, in which maximum likelihood analysis is used to estimate the common correlation matrix in the first stage, and weighted least squares analysis is used to fit structural equation models to the common correlation matrix in the second stage. In the present paper, we propose an alternative method, ML MASEM, that uses ML estimation throughout. In a simulation study, we use both methods and compare chi-square distributions, bias in parameter estimates, false positive rates, and true positive rates. Both methods appear to yield unbiased parameter estimates and false and true positive rates that are close to the expected values. ML MASEM parameter estimates are found to be significantly less bias than two-stage structural equation modeling estimates, but the differences are very small. The choice between the two methods may therefore be based on other fundamental or practical arguments. Copyright © 2016 John Wiley & Sons, Ltd.
Asunto(s)
Funciones de Verosimilitud , Metaanálisis como Asunto , Estadística como Asunto , Algoritmos , Sesgo , Simulación por Computador , Reacciones Falso Positivas , Análisis de los Mínimos Cuadrados , Modelos Estadísticos , Reproducibilidad de los Resultados , Proyectos de InvestigaciónRESUMEN
Participation in motor activities is essential for social interaction and life satisfaction in children. Self-perceptions and task values have a central position in why children do or do not participate in (motor) activities. Investigating developmental changes in motor self-perceptions and motor task values in elementary school children would provide vital information about their participation in motor activities. We therefore examined the change in, and associations between, self-perceptions and task values of fine motor competence, ball competence, and athletic competence in 292 children from kindergarten to grade 4. We also investigated differences between boys and girls, and between children with motor problems and typically developing children. Results indicated that self-perceptions and task values are domain specific and differ between boys and girls, but not between children with motor problems and typically developing children. Self-perceptions were not associated with task values. Educators should address specific self-perceptions to enhance participation into the corresponding motor activities in children between kindergarten and grade 4, and differences in self-perceptions and task values between boys and girls should be taken into account.
Asunto(s)
Conducta Infantil/fisiología , Actividad Motora/fisiología , Trastornos de la Destreza Motora/fisiopatología , Destreza Motora/fisiología , Autoimagen , Niño , Preescolar , Femenino , Humanos , Estudios Longitudinales , Masculino , Factores SexualesRESUMEN
Cluster bias refers to measurement bias with respect to the clustering variable in multilevel data. The absence of cluster bias implies absence of bias with respect to any cluster-level (level 2) variable. The variables that possibly cause the bias do not have to be measured to test for cluster bias. Therefore, the test for cluster bias serves as a global test of measurement bias with respect to any level 2 variable. However, the validity of the global test depends on the Type I and Type II error rates of the test. We compare the performance of the test for cluster bias with the restricted factor analysis (RFA) test, which can be used if the variable that leads to measurement bias is measured. It appeared that the RFA test has considerably more power than the test for cluster bias. However, the false positive rates of the test for cluster bias were generally around the expected values, while the RFA test showed unacceptably high false positive rates in some conditions. We conclude that if no significant cluster bias is found, still significant bias with respect to a level 2 violator can be detected with an RFA model. Although the test for cluster bias is less powerful, an advantage of the test is that the cause of the bias does not need to be measured, or even known.
Asunto(s)
Algoritmos , Sesgo , Análisis por Conglomerados , Interpretación Estadística de Datos , Métodos Epidemiológicos , Modelos Estadísticos , Simulación por Computador , Reproducibilidad de los Resultados , Sensibilidad y EspecificidadRESUMEN
INTRODUCTION: Higher prevalence rates of cannabis abuse/dependence and abuse/dependence criteria in 18-24year old versus older cannabis users and in males versus females might reflect true differences in the prevalence of these disorders across age and gender or, alternatively, they could arise from age- and gender-related measurement bias. To understand differences in endorsement across important subgroups, we examined the influence of age and gender simultaneously on the likelihood of endorsement of the various abuse/dependence criteria. METHOD: The sample consisted of 1603 adult past year cannabis users participating in the National Epidemiological Survey on Alcohol and Related Conditions (NESARC), a U.S. population study (39.6% aged 18-24; 62.1% male). Past year DSM-IV cannabis abuse/dependence criteria and withdrawal were assessed with the AUDADIS-IV. A restricted factor analysis with latent moderated structures was used to detect measurement bias. RESULTS: Although cannabis abuse and dependence diagnoses and various individual abuse/dependence criteria showed different prevalence rates across younger and older male and female cannabis users, none of the items showed uniform or non-uniform measurement bias with respect to age or gender. CONCLUSION: The results indicate that, although prevalence rates of cannabis abuse/dependence criteria differ across age and gender, past year abuse/dependence criteria function similarly across these groups. It can thus be concluded that the criteria are applicable to younger and older, as well as male and female, adult cannabis users.
Asunto(s)
Abuso de Marihuana/epidemiología , Adolescente , Adulto , Factores de Edad , Anciano , Femenino , Encuestas Epidemiológicas , Humanos , Masculino , Persona de Mediana Edad , Prevalencia , Factores Sexuales , Estados Unidos/epidemiología , Adulto JovenRESUMEN
Within structural equation modeling, the most prevalent model to investigate measurement bias is the multigroup model. Equal factor loadings and intercepts across groups in a multigroup model represent strong factorial invariance (absence of measurement bias) across groups. Although this approach is possible in principle, it is hardly practical when the number of groups is large or when the group size is relatively small. Jak et al. (2013) showed how strong factorial invariance across large numbers of groups can be tested in a multilevel structural equation modeling framework, by treating group as a random instead of a fixed variable. In the present study, this model is extended for use with three-level data. The proposed method is illustrated with an investigation of strong factorial invariance across 156 school classes and 50 schools in a Dutch dyscalculia test, using three-level structural equation modeling.
RESUMEN
This paper intends to gain insight into the role of childhood relationships and experiences within the parental home for the formation and meaning of later family relationships and loneliness. Particularly, childhood attachment to mother and father and stressful childhood experiences were studied in their association with satisfaction in the romantic relationship, the quality of adult family ties and the perceived quality of the social network, i.e. loneliness in adulthood. Based on data from the Netherlands Kinship Panel Study (N=3980) structural equation models were estimated to predict adult relationships and loneliness with childhood experiences. Positive attachment experiences with parents, such as reliability, closeness and supportiveness during childhood were associated with greater satisfaction in the romantic relationship, stronger family ties and less loneliness, whereas stressful childhood experiences, such as conflicts and violence negatively predicted the quality of adult relationships. Life span theoretical perspectives, such as attachment theory are discussed as useful unifying framework to study social relationships, their interconnectedness and association with outcome during all phases of life.