Table 1

Descriptive Statistics and Balance of Covariates at Baseline, Full Baseline Sample

Control (N = 1,612)School Feeding (N = 1,821)School Feeding – Control Difference (SE)
Child age in months102.73103.921.185
Child is male0.540.51−0.027
Child has child fallen ill in last 7 days0.100.09−0.010
Highest grade completed1.671.760.088
Child has repeated a grade0.110.120.011
Absent from school in past 7 days0.120.170.050
Private school0.100.110.010
Height-for-age z-scores−1.11−1.050.062
Number of children of target age3.383.24−0.142
Number of children under 5 years1.060.94−0.117
Household size6.776.60−0.178
Head of the household is male0.810.80−0.004
Head of the household’s age44.0645.521.477*
Mother’s age37.4538.581.128
Mother’s education in years5.226.010.789
Wealth index13.2113.380.174
Sold agriculture produce in the past year0.510.43−0.074
Per capita expenditure2,085.172,092.627.446
Household owns livestock0.680.66−0.012
Northern regions0.430.500.066
  • Notes:

  • * p < 0.1.

  • N = 3,433. This table presents descriptive statistics for the full baseline sample of eligible children at baseline, stratified by assignment to treatment. The sample refers to all children aged 5–15 interviewed at baseline, prior to attrition. Mean and standard deviation in parentheses. The school feeding – control difference column reports the school feeding coefficient of a basic OLS regression with each covariate as an outcome and standard errors clustered at the community level. For each variable, the estimated school feeding coefficient provides the difference between the school feeding and control groups and its standard errors.