Modeling Relations Among Discrete Developmental Processes: A General Approach to Associative Latent Transition...

Area of Science:

• Developmental psychology
• Quantitative psychology
• Behavioral science

Background:

• Understanding developmental processes requires examining their interrelations.
• Existing statistical models (e.g., latent growth curve models) are limited to continuous, unidimensional trajectories.
• There is a need for methods that can model discrete, multidimensional developmental relationships.

Purpose of the Study:

• To present associative latent transition analysis (ALTA), a flexible approach for modeling developmental relations among discrete, multidimensional latent variables.
• To detail the substantive interpretation of different ALTA models and their expressed hypotheses.
• To empirically demonstrate ALTA using data on adolescent alcohol use and sexual risk behavior.

Main Methods:

• Utilizes the general framework of loglinear modeling with latent variables.
• Introduces associative latent transition analysis (ALTA) for modeling discrete, multidimensional latent variables.
• Applies ALTA to examine the longitudinal association between alcohol use and sexual risk behavior in adolescence.

Main Results:

• ALTA provides a flexible framework for analyzing complex developmental associations.
• The empirical demonstration illustrates the application of ALTA in understanding the interplay between substance use and risk behaviors.
• Specific hypotheses regarding the association between alcohol use and sexual risk development can be tested.

Conclusions:

• Associative latent transition analysis (ALTA) offers a powerful tool for studying the development of multiple, discrete latent variables.
• ALTA advances the statistical modeling of developmental processes, particularly for behaviors like adolescent alcohol use and sexual risk.
• This approach facilitates a deeper understanding of the complex, interconnected pathways in human development.