Suchergebnisse

Latent variable models can only be compared across groups when these groups exhibit measurement equivalence or "invariance," since otherwise substantive differences may be confounded with measurement differences. This article suggests examining directly whether measurement differences present could confound substantive analyses, by examining the expected parameter change (EPC)-interest. The EPC-interest approximates the change in parameters of interest that can be expected when freeing cross-group invariance restrictions. Monte Carlo simulations suggest that the EPC-interest approximates these changes well. Three empirical applications show that the EPC-interest can help avoid two undesirable situations: first, it can prevent unnecessarily concluding that groups are incomparable, and second, it alerts the user when comparisons of interest may still be invalidated even when the invariance model appears to fit the data. R code and data for the examples discussed in this article are provided in the electronic appendix (http://hdl.handle.net/1902.1/21816).

Social and behavioral scientists are increasingly employing technologies such as fMRI, smartphones, and gene sequencing, which yield 'high-dimensional' datasets with more columns than rows. There is increasing interest, but little substantive theory, in the role the variables in these data play in known processes. This necessitates exploratory mediation analysis, for which structural equation modeling is the benchmark method. However, this method cannot perform mediation analysis with more variables than observations. One option is to run a series of univariate mediation models, which incorrectly assumes independence of the mediators. Another option is regularization, but the available implementations may lead to high false-positive rates. In this article, we develop a hybrid approach which uses components of both filter and regularization: the 'Coordinate-wise Mediation Filter'. It performs filtering conditional on the other selected mediators. We show through simulation that it improves performance over existing methods. Finally, we provide an empirical example, showing how our method may be used for epigenetic research.

In the social and behavioral sciences, variables are often categorical and people are often nested in groups. Models for such data, such as multilevel logistic regression or the multilevel latent class model, should account for not only the categorical nature of the variables, but also the nested structure of the persons. To assess whether the model accomplishes this goal adequately, local fit measures for multilevel categorical data were recently introduced by Nagelkerke, Oberski, and Vermunt (2015). The BVR-group evaluates the variable–group fit, and the BVR-pair evaluates the person–person fit within groups. In this article, we evaluate the performance of these 2 measures for the multilevel latent class model (Vermunt, 2003). An extensive simulation study indicates that whenever multilevel latent class modeling itself is viable, Type I error is controlled and power is adequate for both fit statistics. Thus, the BVR-group and BVR-pair are useful measures to locate important sources of misfit in multilevel latent class analysis.

Latent class analysis is used in the political science literature in both substantive applications and as a tool to estimate measurement error. Many studies in the social and political sciences relate estimated class assignments from a latent class model to external variables. Although common, such a "three-step" procedure effectively ignores classification error in the class assignments; Vermunt (2010, "Latent class modeling with covariates: Two improved three-step approaches,"Political Analysis18:450–69) showed that this leads to inconsistent parameter estimates and proposed a correction. Although this correction for bias is now implemented in standard software, inconsistency is not the only consequence of classification error. We demonstrate that the correction method introduces an additional source of variance in the estimates, so that standard errors and confidence intervals are overly optimistic when not taking this into account. We derive the asymptotic variance of the third-step estimates of interest, as well as several candidate-corrected sample estimators of the standard errors. These corrected standard error estimators are evaluated using a Monte Carlo study, and we provide practical advice to researchers as to which should be used so that valid inferences can be obtained when relating estimated class membership to external variables.

Latent Markov models with covariates can be estimated via 1-step maximum likelihood. However, this 1-step approach has various disadvantages, such as that the inclusion of covariates in the model might alter the formation of the latent states and that parameter estimation could become infeasible with large numbers of time points, responses, and covariates. This is why researchers typically prefer performing the analysis in a stepwise manner; that is, they first construct the measurement model, then obtain the latent state classifications, and subsequently study the relationship between covariates and latent state memberships. However, such a stepwise approach yields downward-biased estimates of the covariate effects on initial state and transition probabilities. This article, shows how to overcome this problem using a generalization of the bias-corrected 3-step estimation method proposed for latent class analysis (Asparouhov & Muthén, 2014; Bolck, Croon, & Hagenaars, 2004; Vermunt, 2010). We give a formal derivation of the generalization to latent Markov models and discuss how it can be used with many time points by incorporating it into a Baum–Welch type of expectation-maximization algorithm. We evaluate the method through a simulation study and illustrate it using an application on household financial portfolio change. Our study shows that the proposed correction method yields unbiased parameter estimates and accurate standard errors, except for situations with very poorly separated classes and a small sample.

Many variables crucial to the social sciences are not directly observed but instead are latent and measured indirectly. When an external variable of interest affects this measurement, estimates of its relationship with the latent variable will then be biased. Such violations of "measurement invariance" may, for example, confound true differences across countries in postmaterialism with measurement differences. To deal with this problem, researchers commonly aim at "partial measurement invariance" that is, to account for those differences that may be present and important. To evaluate this importance directly through sensitivity analysis, the "EPC-interest" was recently introduced for continuous data. However, latent variable models in the social sciences often use categorical data. The current paper therefore extends the EPC-interest to latent variable models for categorical data and demonstrates its use in example analyses of U.S. Senate votes as well as respondent rankings of postmaterialism values in the World Values Study.