Fig. 2: Spatial clustering of BMI. | Nutrition & Diabetes

Fig. 2: Spatial clustering of BMI.

From: Overlapping spatial clusters of sugar-sweetened beverage intake and body mass index in Geneva state, Switzerland

Fig. 2

Getis-Ord Gi clusters calculated for 15,423 Bus santé participants (1995–2014) for the raw body mass index (BMI) variable (a) and adjusted for covariates (b). White dots correspond to individuals with a non-significant Z-score. Red dots correspond to individuals with a statistically significant positive Z-score (α = 0.05), meaning that higher values cluster within a spatial buffer of 1200 m and are found closer together than expected if the underlying spatial process was random. Blue dots correspond to individuals with a statistically significant negative Z-score (α = −0.05), meaning that lower values cluster within a spatial buffer of 1200 m and are found closer together than expected if the underlying spatial process was random. Indicative landmarks numbered 1–10 are displayed on the maps and used to support the description of the results

Back to article page