A Thematic Mapping Approach to Stunting Segmentation in Indonesia by Province Using Finite Mixture Partial Least Squares Method

Abstract

Stunting remains one of Indonesia’s most persistent public-health problems, and the national
prevalence conceals subtantial variation across the country’s 38 provinces. This study models
seven latent constructs behind stunting, antenatal care, early initiation of breastfeeding,
immunization, sanitation, exclusive breastfeeding, low birth weight, and stunting using Partial
Least Squares Structural Equation Modeling on 2023–2024 provincial health-profile data
published by Indonesia’s Ministry of Health, and then applies Finite Mixture Partial Least
Squares (FIMIX-PLS) to detect unobserved heterogeneity across provinces and to translate the
resulting segments into a thematic map. The measurement model satisfied every recommended
validity and reliability threshold, with an Average Variance Extracted between 0.835 and
0.932 and both Cronbach’s alpha and composite reliability above 0.80. In the pooled national
model, low birth weight was the only construct with a statistically significant direct path to
stunting (β = 0.540, p < .001); antenatal care showed no significant direct path to stunting
(β = −0.143, p = .678) but exerted a significant total effect (β = −0.365) entirely through a
fully mediated pathway via low birth weight (indirect effect β = −0.263, p = .022), together
explaining 63.0% of the variance in stunting. Model-selection criteria favored a three-segment
solution comprising 21, 10, and 7 provinces, dominated respectively by sanitation-related,
birth-weight-related, and breastfeeding-related structural pathways. Mapping these segments
geographically shows that provinces sharing a dominant pathway are not scattered at random
but tend to cluster by region, indicating that stunting-reduction programs should be tailored
to each segment’s dominant pathway rather than applied uniformly across the whole country.