Augustin Kelava

Risk for Permanent Pacemaker After Transcatheter Aortic Valve Implantation: A Comprehensive Analysis of the Literature

Risk for Permanent Pacemaker After Transcatheter Aortic Valve Implantation.

Advanced nonlinear latent variable modeling: Distribution analytic LMS and QML estimators of interaction and quadratic effects

Interaction and quadratic effects in latent variable models have to date only rarely been tested in practice. Traditional product indicator approaches need to create product indicators (e.g., x 1 2, x 1 x 4) to serve as indicators of each nonlinear latent construct. These approaches require the use of complex nonlinear constraints and additional model specifications and do not directly address the nonnormal distribution of the product terms. In contrast, recently developed, easy-to-use distribution analytic approaches do not use product indicators, but rather directly model the nonlinear multivariate distribution of the measured indicators. This article outlines the theoretical properties of the distribution analytic Latent Moderated Structural Equations (LMS; Klein & Moosbrugger, 2000) and Quasi-Maximum Likelihood (QML; Klein & Muthén, 2007) estimators. It compares the properties of LMS and QML to those of the product indicator approaches. A small simulation study compares the two approaches and illustrates the advantages of the distribution analytic approaches as multicollinearity increases, particularly in complex models with multiple nonlinear terms. An empirical example from the field of work stress applies LMS and QML to a model with an interaction and 2 quadratic effects. Example syntax for the analyses with both approaches is provided.

Probing the Unique Contributions of Self-Concept, Task Values, and Their Interactions Using Multiple Value Facets and Multiple Academic Outcomes

Drawing on expectancy-value theory, the present study examined the unique contributions of the four major value beliefs and self-concept on achievement, self-reported effort, and teacher-rated behavioral engagement in mathematics. In particular, we examined the multiplicative effects of self-concept and task values on educational outcomes using the latent moderated structural equation approach. Participants were 1,868 German ninth-grade students. The data analyses relied on a higher-order structure of value beliefs, which is suited to parsing the differential patterns of predictive relations for different value beliefs. The findings revealed that (a) self-concept was more predictive of achievement, whereas value beliefs were more predictive of self-rated effort; (b) self-concept and value beliefs emerged as equally important predictors of teacher-reported engagement; (c) among the four value beliefs, achievement was more associated with low cost, whereas effort was more associated with attainment value; and (d) latent interactions between self-concept and value beliefs predicted the three outcomes synergistically.

Deskriptivstatistische Evaluation von Items (Itemanalyse) und Testwertverteilungen

Nachdem die Planungs- und Entwicklungsphase eines psychologischen Tests oder Fragebogens (vgl. Jonkisz, Moosbrugger & Brandt 2011, ► Kap. 3 in diesem Band) abgeschlossen ist, besteht der nachste Schritt darin, die Items an einer fur die Zielgruppe moglichst reprasentativen Stichprobe einer deskriptivstatistischen Evaluation zu unterziehen. Erst nach diesen unter dem Namen »Itemanalyse« zusammengefassten Untersuchungsschritten konnen wir eine tragfahige Testfassung erstellen.

Synergistic effects of expectancy and value on homework engagement: The case for a within-person perspective.

Historically, expectancy–value models of motivation assumed a synergistic relation between expectancy and value: motivation is high only when both expectancy and value are high. Motivational processes were studied from a within-person perspective, with expectancies and values being assessed or experimentally manipulated across multiple domains and the focus being placed on intraindividual differences. In contrast, contemporary expectancy–value models in educational psychology concentrate almost exclusively on linear effects of expectancy and value on motivational outcomes, with a focus on between-person differences. Recent advances in latent variable methodology allow both issues to be addressed in observational studies. Using the expectancy–value model of homework motivation as a theoretical framework, this study estimated multilevel structural equation models with latent interactions in a sample of 511 secondary school students and found synergistic effects between domain-specific homework expectancy and homework value in predicting homework engagement in 6 subjects. This approach not only brings the “×” back into expectancy–value theory but also reestablishes the within-person perspective as the appropriate level of analysis for latent expectancy–value models.

A Nonlinear Structural Equation Mixture Modeling Approach for Nonnormally Distributed Latent Predictor Variables

Structural equation models with interaction and quadratic effects have become a standard tool for testing nonlinear hypotheses in the social sciences. Most of the current approaches assume normally distributed latent predictor variables. In this article, we describe a nonlinear structural equation mixture approach that integrates the strength of parametric approaches (specification of the nonlinear functional relationship) and the flexibility of semiparametric structural equation mixture approaches for approximating the nonnormality of latent predictor variables. In a comparative simulation study, the advantages of the proposed mixture procedure over contemporary approaches [Latent Moderated Structural Equations approach (LMS) and the extended unconstrained approach] are shown for varying degrees of skewness of the latent predictor variables. Whereas the conventional approaches show either biased parameter estimates or standard errors of the nonlinear effects, the proposed mixture approach provides unbiased estimates and standard errors. We present an empirical example from educational research. Guidelines for applications of the approaches and limitations are discussed.

Qualitätsanforderungen an einen psychologischen Test (Testgütekriterien)

Wenn man mit der Frage konfrontiert wird, worin der eigentliche Unterschied zwischen einem unwissenschaftlichen »Test« (etwa einer Fragensammlung) und einem wissenschaftlich fundierten, psychologischen Test besteht, so ist die Antwort darin zu sehen, dass sich ein psychologischer Test dadurch unterscheidet, dass er hinsichtlich der Erfullung der sog. Testgutekriterien empirisch uberpruft wurde.

A Bayesian Model for the Estimation of Latent Interaction and Quadratic Effects When Latent Variables Are Non-Normally Distributed.

Structural equation models with interaction and quadratic effects have become a standard tool for testing nonlinear hypotheses in the social sciences. Most of the current approaches assume normally distributed latent predictor variables. In this article, we present a Bayesian model for the estimation of latent nonlinear effects when the latent predictor variables are nonnormally distributed. The nonnormal predictor distribution is approximated by a finite mixture distribution. We conduct a simulation study that demonstrates the advantages of the proposed Bayesian model over contemporary approaches (Latent Moderated Structural Equations [LMS], Quasi-Maximum-Likelihood [QML], and the extended unconstrained approach) when the latent predictor variables follow a nonnormal distribution. The conventional approaches show biased estimates of the nonlinear effects; the proposed Bayesian model provides unbiased estimates. We present an empirical example from work and stress research and provide syntax for substantive researchers. Advantages and limitations of the new model are discussed.

A Simulation Study Comparing Recent Approaches for the Estimation of Nonlinear Effects in SEM Under the Condition of Nonnormality

In the past decade new approaches for the estimation of latent nonlinear interaction and quadratic effects in structural equation modeling have been proposed (Kelava & Brandt, 2009; Klein & Moosbrugger, 2000; Klein & Muthén, 2007; Marsh, Wen, & Hau, 2004; Mooijaart & Bentler, 2010; Wall & Amemiya, 2003). Most approaches have been developed for the analysis of normally distributed latent predictor variables. In this article, we investigate the performance of five recent approaches under the condition of nonnormally distributed data: the extended unconstrained approach (Kelava & Brandt, 2009), LMS (Klein & Moosbrugger, 2000), QML (Klein & Muthén, 2007), the 2SMM approach (Wall & Amemiya, 2003), and the method of moments approach by Mooijaart and Bentler (2010). Advantages and limitations of the approaches are discussed.

A general non-linear multilevel structural equation mixture model

In the past 2 decades latent variable modeling has become a standard tool in the social sciences. In the same time period, traditional linear structural equation models have been extended to include non-linear interaction and quadratic effects (e.g., Klein and Moosbrugger, 2000), and multilevel modeling (Rabe-Hesketh et al., 2004). We present a general non-linear multilevel structural equation mixture model (GNM-SEMM) that combines recent semiparametric non-linear structural equation models (Kelava and Nagengast, 2012; Kelava et al., 2014) with multilevel structural equation mixture models (Muthén and Asparouhov, 2009) for clustered and non-normally distributed data. The proposed approach allows for semiparametric relationships at the within and at the between levels. We present examples from the educational science to illustrate different submodels from the general framework.