Table 5.

Interaction between programme impact in Angola and distance to Namibian border

Outcome	Fever episode in 2 weeks prior to survey
	Linear probability model	Logistic regression model
	(1)	(2)
Treated (midline)	− 0.112** (− 0.200 to − 0.0233)	0.447*** (0.244 to 0.817)
Distance	− 0.000580 (− 0.00802 to 0.00686)	0.996 (0.955 to 1.040)
Treated* distance	0.0111*** (0.00384 to 0.0184)	1.078*** (1.032 to 1.127)
Baseline	0.211*** (0.121 to 0.302)	0.268*** (0.160 to 0.449)
Observations	1469	1469
R-squared	0.013

Multivariable regression results from analysis for treatment effect modification in Angola. Results based on linear probability model are shown in Column 1. Results based on logistic regression model are shown in Column 2. 95% confidence intervals shown in parentheses are based on Huber’s cluster robust variance estimator. Sample is restricted to Angolan villages receiving treatment between baseline and midline

*** p < 0.01, ** p < 0.05, * p < 0.1