Developing a WhatsApp-Based Extension Model for Behavioral Change in Child Feeding Practices: An Analysis Using Smart PLS-SEM

Abstract

Indonesia faces persistent stunting, with a 21.5% prevalence among children aged 0–59 months (Indonesian Health Survey, 2023), which exceeds the 2024 target of 14%. Inappropriate feeding practices driven by limited caregiver knowledge remain the key determinants of poor nutrition (Aguayo & Menon, 2016; Diana et al., 2022). This study developed and validated a WhatsApp-based extension model to promote behavioral change toward balanced, local food-based child feeding.A quasi-experimental pre-post design was conducted with 128 parents of children aged 0–6 years in three East Lombok sub-districts (July–October 2025). The control group (n=64) received brochures, and the intervention group (n=64) received brochures plus weekly WhatsApp videos over 10 weeks. Complete pre-post data from all 128 participants were analyzed using change scores (Δ).Smart PLS-SEM integrated with TPB-TAM constructs. Changes in perceived ease of use exerted the strongest effect on changes in feeding behavior (β = 0.415, p < 0.001), followed by changes in perceived WhatsApp diffusion (β = 0.352, p < 0.001). The model explained 40.7% of the variance in feeding behavior change (R² = 0.407) with an acceptable model fit (SRMR = 0.082) and moderate predictive relevance (Q² = 0.226).This study provides the first empirically validated TPB-TAM framework for WhatsApp-delivered nutrition extension in Indonesia, offering a scalable low-cost digital intervention model for rural stunting reduction.