Skip to main content

Table 3 Results of the geostatistical variable selection. Models selected with the highest model posterior probability are presented, together with posterior inclusion probability of each explored predictor

From: Bayesian risk profiling of soil-transmitted helminth infections and estimates of preventive chemotherapy for school-aged children in Côte d'Ivoire

  Selected model with predictor posterior inclusion probability
Predictor Hookworm A. lumbricoides T. trichiura
Day land surface temperature (LST)a,b,c Not selected (27.6 %) Not selected (33.0 %) Not selected (40.6 %)
Night land surface temperature (LST)b,c Not selected (12.9 %) Not selected (23.1 %) Not selected (23.9 %)
LST difference Not selected (10.6 %) Not selected (22.2 %) Not selected (24.7 %)
Land covera,b,c Not selected (18.2 %) Not selected (16.3 %) Not selected (31.3 %)
Normalized difference vegetation indexc Not selected (8.1 %) Not selected (10.8 %) Not selected (34.2 %)
Rainfallc Not selected (14.4 %) Not selected (12.3 %) Not selected (19.6 %)
Rainfall coefficient of variation (CV) Not selected (7.1 %) Not selected (17.1 %) Selected (71.8 %)
Altitudeb,c Not selected (23.6 %) Not selected (42.5 %) Not selected (39.4 %)
Soil acidity (pH)b Not selected (7.4 %) Selected (59.0 %) Not selected (37.2 %)
Soil moisturea,c Not selected (17.7 %) Selected (90.8 %) Not selected (22.5 %)
Ecological zonea,b,c Not selected (17.4 %) Not selected (30.3 %) Not selected (24.7 %)
Rural/urban settinga Selected (84.3 %) Not selected (26.2 %) Not selected (44.6 %)
Human influence index (HII)b,c Not selected (25.3 %) Not selected (36.1 %) Not selected (32.7 %)
Improved sanitation Not selected (23.5 %) Not selected (23.1 %) Not selected (19.1 %)
Improved drinking water Not selected (7.1 %) Not selected (21.9 %) Not selected (29.6 %)
Model posterior probability 8.9 % 3.2 % 0.4 %
  1. aCategorised for hookworm; bCategorised for A. lumbricoides; cCategorised for T. trichiura