Skip to main content

Table 2 Prevention, malaria and poverty in Uganda: OLS and 3SLS GMM regressions results

From: Malaria and protective behaviours: is there a malaria trap?

  (1) (2) (3) (4) (5) (6)
  OLS OLS OLS 3SLS GMM
  Dep Var is % using an ever treated net last night in the village Dep var is malaria prevalence in the village Dep var is poverty incidence in the village Dep Var is % using an ever treated net last night in the village Dep var is malaria prevalence in the village Dep var is poverty incidence in the village
Malaria - −0.044 - - −0.862*** -
Prevention   (0.130)    (0.308)  
Poverty −0.046 0.376*** - −0.323* 0.543*** -
Incidence (0.078) (0.096)   (0.194) (0.132)  
Malaria −0.022 - 0.242*** 0.438** - 0.302**
Prevalence (0.063)   (0.067) (0.175)   (0.124)
Intercept 0.208*** 0.400*** −0.013 0.062 1.364*** −0.290***
  (0.044) (0.058) (0.047) (0.078) (0.174) (0.064)
Observations 170 170 170 170 170 170
R-squared 0.379 0.550 0.668 - - -
  1. The coefficients attached to each variable are presented (standard errors, adjusted for heteroscedasticity in parentheses). All regressions include regional dummies. ***denotes statistical significance at the 1% level, ** at the 5% level, * at the 10% level. The Hansen J Test of overidentifying restrictions shows that the instruments are well identified in 3SLS GMM regressions (Hansen's J chi2= 12.247; p value = 0.140). A rejection of the null hypothesis implies that the instruments are not satisfying the orthogonality conditions required for their employment (i.e. that they are uncorrelated with the error term of the estimated Equation).