Making Multimethod Latent State-Trait Models for Random and Fixed Situations Accessible: A Tutorial.

Area of Science:

• Psychology
• Quantitative Psychology
• Longitudinal Research Methods

Background:

• Longitudinal research designs integrating multiple methods and situations are increasingly common.
• Existing analytical methods, such as multimethod latent state-trait models (MM-LST-RF), are powerful but complex.
• This complexity presents a significant barrier to the widespread application of these models.

Purpose of the Study:

• To facilitate the application of complex multimethod latent state-trait models for random and fixed situations (MM-LST-RF).
• To provide researchers with accessible tools for analyzing data from sophisticated longitudinal study designs.

Main Methods:

• The study presents two simplified methodological approaches by decomposing the full MM-LST-RF model into its multimethod and random/fixed situation components.
• A user-friendly shiny app, powered by a new R function, is introduced to guide users through model specification, estimation, and interpretation.
• Key parameters and model coefficients are illustrated using a motivational example.

Main Results:

• The tutorial breaks down the MM-LST-RF model into manageable components, aiding understanding.
• The developed shiny app and R function provide a practical and guided workflow for applying MM-LST-RF models.
• Detailed explanations and practical recommendations enhance the usability of the models.

Conclusions:

• The developed shiny app significantly lowers the barrier to entry for using MM-LST-RF models in psychological research.
• This facilitates deeper insights into psychological constructs by enabling robust analysis of complex longitudinal data.
• The tools support researchers in effectively utilizing advanced statistical methods for multifaceted research designs.