Pooling methods for likelihood ratio tests in multiply imputed data sets.

Abstract

Likelihood ratio tests (LRTs) are a popular tool for comparing statistical models. However, missing data are also common in empirical research, and multiple imputation (MI) is often used to deal with them. In multiply imputed data, there are multiple options for conducting LRTs, and new methods are still being proposed. In this article, we compare all available methods in multiple simulations covering applications in linear regression, generalized linear models, and structural equation modeling. In addition, we implemented these methods in an R package, and we illustrate its application in an example analysis concerned with the investigation of measurement invariance.

Missing data in the analysis of multilevel and dependent data

“…In principle, the methods considered in this chapter can also be applied to other types of dependent data such as three-level, crossclassified, or multiple-membership data (for an overview, see Goldstein, 2011). For example, Wijesuriya et al (2020) and considered a variety of MI approaches for three-level data, and Grund et al (2023b) proposed JM and FCS approaches for the imputation of cross-classified data (see also Yucel et al, 2008). Especially in cross-classified data, the treatment of missing data comes with additional challenges due to the non-hierarchical structure of the data and the more complex pattern of dependency that they imply (Luo & Kwok, 2009;Meyers & Beretvas, 2006;Shi et al, 2010).…”