A comparison of the multilevel MIMIC model to the multilevel regression and mixed ANOVA model for the estimation and testing of a cross-level interaction effect: A simulation study

When observing data on a patient-reported outcome measure in, for example, clinical trials, the variables observed are often correlated and intended to measure a latent variable. In addition, such data are also often characterized by a hierarchical structure, meaning that the outcome is repeatedly measured within patients. To analyze such data, it is important to use an appropriate statistical model, such as structural equation modeling (SEM). However, researchers may rely on simpler statistical models that are applied to an aggregated data structure. For example, correlated variables are combined into one sum score that approximates a latent variable. This may have implications when, for example, the sum score consists of indicators that relate differently to the latent variable being measured. This study compares three models that can be applied to analyze such data: the multilevel multiple indicators multiple causes (ML-MIMIC) model, a univariate multilevel model, and a mixed analysis of variance (ANOVA) model. The focus is on the estimation of a cross-level interaction effect that presents the difference over time on the patient-reported outcome between two treatment groups. The ML-MIMIC model is an SEM-type model that considers the relationship between the indicators and the latent variable in a multilevel setting, whereas the univariate multilevel and mixed ANOVA model rely on sum scores to approximate the latent variable. In addition, the mixed ANOVA model uses aggregated second-level means as outcome. This study showed that the ML-MIMIC model produced unbiased cross-level interaction effect estimates when the relationships between the indicators and the latent variable being measured varied across indicators. In contrast, under similar conditions, the univariate multilevel and mixed ANOVA model underestimated the cross-level interaction effect.

Does Misery Love Company? Multilevel Relationships between Perceived Age Discrimination and Happiness among Older Europeans

The present study examines how contextual age discrimination moderates the individual-level association between perceived age discrimination and happiness among older Europeans. In this endeavor, we test two opposing views: 1) the "social norm" hypothesis that predicts the association between perceived age discrimination and happiness to become weaker in areas with a higher average level of age discrimination; and 2) conversely the "contagion effect" hypothesis that predicts the association to grow stronger in such areas. Using data from the European Social Survey (2008), we estimate two- and three-level mixed effects models to test these opposing hypotheses. Our findings from multilevel analysis lend support to the social norm hypothesis. Specifically, the negative link between perceived age discrimination and happiness is weaker in subnational regions where the proportion of victims of age discrimination is higher.

Give a Man a Fish or Teach a Man to Fish: A Cross-Level Moderated Mediation Model of Cognitive and Performance Responses of Team Members to Help of Team Leaders

Drawing upon theoretical lenses of social cognitive theory, this study explores whether, when, why, and how the helping behaviors of team leaders influence individual work role performance of team members (in terms of individual task proficiency, task adaptivity, and task proactivity) through self-efficacy of team members. The consequences of different types of help of leaders are uncovered in this study. By proposing a cross-level moderated mediation model and using multisource and multistage data from 303 team members in 39 work teams, autonomy-oriented help of leaders was found to have a differential effect on individual work role performance of members via the self-efficacy of the latter when controlling for dependency-oriented help of leaders. Moreover, the multilevel analysis of moderation uncovered that leader–member exchange relationship at the team level engendered a boundary condition for the mediating role of member self-efficacy in the relationship between autonomy-oriented help of leaders and individual work role performance of members in this model.

The Effect of Cross-Level Interaction between Community Factors and Social Capital among Individuals on Physical Activity: Considering Gender Difference

This study examines the effect of cross-level interaction between community physical environment and social capital among individuals on physical activity by considering gender difference. In this regard, we ask two research questions: (1) What is the effect of cross-level interaction between community factors and social capital among individuals on physical activity? (2) Is there gender difference in the effect of the cross-level interaction? To examine the research questions, this study used the 2015 Korea Community Health Survey and used multi-level analyses. The empirical analyses show that while there are both positive and negative cross-level interaction effects between physical activity-supportive community environment and social capital among individuals on physical activity, the positive cross-level interaction effect is more pronounced for women than for men. These findings suggest that local efforts to improve public health should take into account the cross-level interaction effect between community physical environment and social capital among individuals as well as gender difference.