Shady business: understanding the spatial ecology of exophilic Anopheles mosquitoes

Malaria Journal

Table 1 The effect of landscape characteristics within a 10 m radius of the sampling points on the number of Anopheles mosquitoes caught in resting clay pots, as shown by zero-inflated negative binomial regression, followed by backward selection of non-significant independent variables

Variables	Estimate	Std. error	z value	Pr(> \|z\|)
Females
Count model coefficients (negbin with log link)
(Intercept)	− 1.2435	0.6274	− 1.982	0.04747*
Distance to nearest dwelling (m)	0.0136	0.0052	2.599	0.0094**
Percent canopy cover	0.0238	0.0068	3.483	0.0005***
Zero-inflation model coefficients (binomial with logit link)
(Intercept)	1.3602	0.6274	1.640	0.1010
Distance to nearest dwelling (m)	− 0.0016	0.0087	− 0.185	0.8530
Percent canopy cover	− 0.0279	0.0094	− 2.963	0.0030**
Males
Count model coefficients (negbin with log link)
(Intercept)	− 1.4289	1.2516	− 1.142	0.2536
Percent canopy cover	0.0215	0.0151	1.420	0.1555
Percent ground vegetation	− 0.0386	0.0231	− 1.672	0.0946
Zero-inflation model coefficients (binomial with logit link)
(Intercept)	2.5915	1.5563	1.665	0.0959
Percent canopy cover	− 0.0432	0.0290	− 1.490	0.1361
Percent ground vegetation	0.0075	0.0503	0.149	0.8819

negbin negative binomial, log link logarithmic link, logit link logistic link
* P < 0.05, **P < 0.01, ***P < 0.001

ISSN: 1475-2875