Analyzing spatial clustering and the spatiotemporal nature and trends of HIV/AIDS prevalence using GIS: the case of Malawi, 1994-2010

Table 2 Summary of regression coefficients for possible drivers of HIV prevalence

	Model 1			Model 2			Model 3			Model 4
Variable description	Beta	Tolerance	VIF	Beta	Tolerance	VIF	Beta	Tolerance	VIF	Beta	Tolerance	VIF
Attended senior primary sch. (%)										-0.287*	0.836	1.196
Percentage that had taken an HIV test ever							0.435**	0.621	1.611	0.360**	0.591	1.691
Mean time to transport, age 30_44 (min)				0.369**	0.998	1.002	0.483***	0.899	1.113	0.507**	0.892	1.121
Mean dist. to main roads (km)	0.605***	1.000	1.000	0.589***	0.998	1.002	-0.341*	0.656	1.523	-0.291*	0.641	1.560
R²	Model 1	0.366		Model 2	0.5		Model 3	0.62		Model 4	0.688
R² adjusted		0.344			0.467			0.467			0.64
F-ratio		16.757			14.662			14.354			14.130
F-ratio significance		.000			0.000			0.000			0.000

N = 31 districts. Significance of the coefficients for individual variables was * for p = 0.05, ** for p = 0.001, and *** for p = 0.001 or lower. Tolerance is proportion of the variance explained by the variable alone and VIF is the Variance Inflation Factor, both collinearity diagnostic statistics. Model 4 was chosen as the “best” model.

ISSN: 1471-2334