A generalized and reproducible R analysis pipeline for two-part (hurdle) structural equation modeling designed for zero-inflated psychological and behavioral outcome variables. The workflow integrates probit latent-response models using the WLSMV estimator in lavaan (Part A: entry transition) and continuous latent models using the MLR estimator (Part B: conditional severity). It supports 1 to 3 mediators and 1 to 2 moderators with automated Johnson-Neyman regions of significance, complemented by robust count and continuous Generalized Linear Models (ZINB and Gamma-log GLM). Includes simulated toy data and a concise methodology guide.
Two-Part-SEM-Template-R: Generalized Two-Part (Hurdle) Structural Equation Modeling Pipeline in R
alvani amin
2026
Abstract
A generalized and reproducible R analysis pipeline for two-part (hurdle) structural equation modeling designed for zero-inflated psychological and behavioral outcome variables. The workflow integrates probit latent-response models using the WLSMV estimator in lavaan (Part A: entry transition) and continuous latent models using the MLR estimator (Part B: conditional severity). It supports 1 to 3 mediators and 1 to 2 moderators with automated Johnson-Neyman regions of significance, complemented by robust count and continuous Generalized Linear Models (ZINB and Gamma-log GLM). Includes simulated toy data and a concise methodology guide.File in questo prodotto:
Non ci sono file associati a questo prodotto.
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
