Skip to main content

Table 3 Results of variable selection for the climatic predictors and ITN coverage measures based on Bayesian geostatistical logistic regression models

From: Assessing the effects of malaria interventions on the geographical distribution of parasitaemia risk in Burkina Faso

Posterior inclusion probability (%)

Variable

Model 1

Model 2

Altitude

28

 

Distance to water body

2

 

NDVI

3.30

 

LSTD

6.20

 

LSTN

74.10

 

Rainfall

1.60

 

Distance to rice growing area

0.30

 

Altitudea

5.70

 

Distance to water bodya

4

 

Area type (urban/rural)a

87.20

 

NDVIa

0.20

 

LSTDa

4.50

 

LSTNa

3.50

 

Rainfalla

1

 

Distance to rice growing areaa

74.50

 

ITN coverage

 ITN per person (ITNpPR)

0.20

 U5 sleep under net (ITNsU5)

32.02

 ITN per under 5 years (ITNpU5)

2.46

  1. Posterior inclusion probabilities larger than 50 % indicate an important predictor
  2. aCategorical form