Comparing two approaches for estimating the causal effect of behaviour-change communication messages promoting insecticide-treated bed nets: an analysis of the 2010 Zambia malaria indicator survey

Table 2 Biprobit regression model predicting whether respondent slept under a net the previous night (n = 3,263)

	Coefficient	SE	p-value
EQUATION B: SLEPT UNDER AN ITN LAST NIGHT
Number of nets in household (Ref = 1)
2 nets	0.02	0.08	0.765
3 or more nets	1.16	0.08	0.001
Received malaria information from a health worker	1.18	0.06	0.001
Exposed to malaria information from BCC	0.48	0.12	0.001
Constant	-0.48	0.08	0.001
EQUATION A: EXPOSED TO MALARIA INFORMATION FROM BCC
Age (Ref = 15–24)
25-34 years old	-0.05	0.10	0.642
35-49 years old	1.34	0.11	0.001
Education (Ref = None)
Primary	2.19	0.17	0.001
Secondary or Higher	3.94	0.19	0.001
Upper 2 wealth quintiles	-0.66	0.09	0.001
Province (Ref: Central, Copperbelt)
Eastern, Northern, Luapula	-0.77	0.10	0.001
Lusaka	-0.28	0.16	0.089
Western, Southern, NorthWestern	-0.61	0.10	0.001
Has a child under the age of 6	-0.55	0.09	0.001
Lives in a district that received IRS	-0.39	0.10	0.001
Lives in an urban area	0.04	0.12	0.736
Constant	-2.19	0.21	0.001
Correlation between residuals	0.021	0.08	0.7791