Probing Latent Interactions Estimated with a Residual Centering Approach

2021 ◽  
Author(s):  
Sunthud Pornprasertmanit ◽  
Alexander M. Schoemann ◽  
G. John Geldhof ◽  
Todd D. Little

Understanding latent interactions is an important need for the structural equation modeler. Plotting and probing latent interactions, however, has not been well defined. We describe methods for plotting and probing two- and three-way latent interactions fit with a variety of approaches (LMS/QML, residual centering, double mean centering). The methods are demonstrated through a small simulation and examples based on existing data.

2007 ◽  
Vol 14 (4) ◽  
pp. 570-580 ◽  
Author(s):  
Herbert W. Marsh ◽  
Zhonglin Wen ◽  
Kit-Tai Hau ◽  
Todd D. Little ◽  
James A. Bovaird ◽  
...  

2020 ◽  
Vol 12 (1) ◽  
pp. 79-95
Author(s):  
Purnamie Titisari ◽  
Arnis Budi Susanto

This study aims to build a model of consumer preference in choosing Islamic banking. This study uses a comparative study between people's understanding of Islamic banking and conventional banking which is influenced by several factors. The sample in this study amounted to 200 people who had different backgrounds. Existing data were analyzed using Structural equation models (SEM). The research results show that there is a level of understanding and preference of the community towards different Islamic banking which is influenced by several factors developed in this research model. In the next research, it is expected to conduct an assessment of the decision to choose Islamic and conventional banking products that concentrate more on specific products in Islamic banking. In addition, further research is expected to take into account environmental factors and government policies as well as other factors that have not been studied in this study so that it is expected to enrich the knowledge in the field of management, especially consumer preferences and decision making using Islamic banking products


2018 ◽  
Vol 2 (2) ◽  
pp. 51-61
Author(s):  
Lutfi Adin Affandi ◽  
Ma'ruf Akbar ◽  
Dedi Purwana

Losing members due to death in battle or desertion from their units is detrimental to the military unit. Both will interfere with the implementation of the overall task. Killed in battle were recorded as a hero and desertion was classified as an unlawful act. The desires of desertion of soldiers must be sought to be prevented, and become part of the task of the unit commander. Likewise, the work environment must be improved so that soldiers become comfortable in working or esprit de corps as a spirit of unity must be maintained and enhanced as the pride of the corps that is able to bind members in a single unit. This quantitative study aims to empirically determine what factors directly influence the level of desertion intention of the soldier. By using structural equation modeling (SEM) estimation techniques on 302 Navy soldiers. The results of processing existing data found that esprit de corps is a very central aspect in an effort to reduce desertion intentions. Leadership also has a significant negative direct effect, although it is not as strong as esprit de corps, while the quality of work life has a direct but not significant negative effect on desertion intentions. It is interesting that the influence of leadership on esprit de corps is very significant. Leadership that pays attention to members through direct communication is capable of increasing unit esprit de corps.


2020 ◽  
Author(s):  
Lesa Hoffman

A primary motivation for conducting longitudinal research is to examine variable relationships at both the between-person (BP) and within-person (WP) levels of analysis. In models with time-varying predictors, the problems of conflating WP residual effects with BP intercept effects are relatively well-known, whereas the potential for additional conflation by BP effects of time has received much less attention. The present study used simulation methods to demonstrate the deleterious impact that ignoring BP relations for time slopes across variables can have on the recovery of contemporaneous or lagged WP effects, as well as BP intercept effects, within common longitudinal models (i.e., in mixed-effects models using person-mean-centering, single-level and multilevel structural equation models, and auto-regressive cross-lag panel models). Recommendations are provided for how to use different options for univariate or multivariate longitudinal models to avoid conflating effects across BP and WP levels of analysis in practice.


Sign in / Sign up

Export Citation Format

Share Document