Fig. 2From: Mapping the prevalence of cancer risk factors at the small area level in AustraliaSchematic describing the components of the linear predictor for \(\text {logit}\left( \pi _{ij} \right)\) in the stage 1 model. *The non-outcome risk factor categorical covariate was derived from the interaction of the binary risk factor outcomes not directly associated with the risk factor being modelled. For more details see Section C of the Additional File 1. SA2 Statistical area level 2Back to article page