Skip to main content

Table 2 Estimated selection probabilities from probit models

From: Heckman-type selection models to obtain unbiased estimates with missing measures outcome: theoretical considerations and an application to missing birth weight data

Probit models
VariableUnadjusted coeffAdjusted coeff
dy/dx95% CIdy/dx95% CI
IFAS: No(+)
 Yes0.171***0.150; 0.1920.243***0.221; 0.265
Distance
 Distance−0.024***-0.027; -0.021− 0.000− 0.006; 0.005
Educational attainment: No schooling(+)
 Primary0.025**0.002; 0.048
 Coranic0.014−0.033; 0.062
 Secondary or higher0.080***0.044; 0.117
Marital status: Unmarried(+)
 Common-law union−0.019−0.057; 0.019
 Married0.064***0.024; 0.103
 Divorced/widowed0.055−0.061; 0.170
Maternal age (years): 20–34(+)
 15–19  0.016−0.014; 0.047
 35–49  0.051***0.021; 0.081
Socioeconomic status: Most poor(+)
 Poor  0.018−0.013; 0.048
 Middle  0.097***0.065; 0.129
 Rich  0.100***0.068; 0.131
 Most rich  0.105***0.066; 0.145
Anaemia: No(+)
 Yes  0.050***0.021; 0.079
Lack of appetite: No(+)
 Yes0.039***0.014; 0.064
Previous births: 0 child(+)
 1–4−0.029-0.086; 0.028
 ≥ 5−0.092***-0.154; -0.029
Twin births: No(+)
 Yes0.092***0.043; 0.140
 observations7325 7325 
  1. *** p < 0.01. ** p < 0.05.* p < 0.1;(+) Reference category; dy/dx = Marginal effect is a change in the probability that Y = 1 with a specific change in X. Adjusted model controls village fixed effects; CI Confidence interval