Call Us: US - +1 845 478 5244 | UK - +44 20 7193 7850 | AUS - +61 2 8005 4826

Journal of Tropical Pediatry

To estimate changes over time and compare them between intervention and control sublocations more rigorously we need to construct a panel. Our child-level observations are unsuitable for this for three reasons. First, individual child identifiers are not consistent across time in the data set. Second, MUAC data are not available for all children in all months. Third, and most importantly, the sample of children will necessarily change over time. A large proportion of MUAC observations is lost over the three year 3 As a robustness check we also asked district project managers to classify sublocations in their districts into treatment and control groups. For brevity we focus on the investment-based treatment and control classifications. 9 period from 2005/06 to 2008/09 as many children observed in the early years have exited the 6-59 month age group and children born since 2005/06 were added to the sample. For the 128 sublocations we constructed a two period panel for 2005/06 and 2008/09 of sublocation specific MUAC z-scores by summarizing the child-level MUAC z-scores in sublocation summary statistics. To focus primarily on malnourished children the results presented below are based on summary statistics that focus on that subpopulation such as the median Z-score of all children with Z-scores below zero, or the proportion of children with MUAC Z-scores below -1 and -2 standard deviations, focusing on standard cut-off levels to capture the prevalence of mild or severe malnutrition.4