Versión de HTML Básico
Assessing the Effect of Conditional Cash Transfers in Children Chronic Stunting: The Human Development Bonus in Ecuador
93
13
Analiti a, Revista de análisis estadístico, Vol. 13 (1), 2017
tioned gap. In this sense, this research proposes an observational causal inference design
using a recent important national survey attempting to analyse the effect of BDH on stunt-
ing of children under 5 years old in Ecuador, focusing on programme’s third stage.
4 Empirical strategy
4.1 Data
The study sources out mainly from a secondary quantitative database, the Ecuadorian
Living Standards Measurement Survey (LSMS), which is executed by the National Statis-
tics Bureau (INEC). The LSMS, as part of the Ecuadorian Integrated Households Surveys
System, collects information on the quality of life and well-being and represents an im-
portant tool for research and policymaking. Historically, six LSMS have been collected
since 1994; however, the focus is in the last one collected from November 2013 to Octo-
ber 2014 (INEC, 2015). These databases can be found in the Institute´s web page:
http:
//www.ecuadorencifras.gob.ec/banco-de-informacion/.
The main dimensions under study in the LSMS are income and expenditure, household
production, housing, health, assets, public services access, education, among others. The
survey is statistically representative for the 4 natural regions, 24 provinces, 9 planning zones
and 4 main cities in urban and rural areas (INEC, 2015). The sample includes 28.970
households, 109.694 individuals and 11.473 children under 5 (INEC, 2014).
The indicator of children stunting, is estimated from the height and age collected on
the questionnaire section 3 (Health), part E under “Anthropometrics”. The height-for-age
z-scores computed are compared to the distribution of the measurements for the standard
sample established in 2006 by WHO in their Multicentre Growth Reference Study
2
. Point
estimates under two standard deviations from the median of the reference are considered
stunted. For this study, I computed results using the igrowup stata package in a Stata 14
version (World Health Organization, 2006).
Regarding BDH reception, information is collected in section 7 (Economic Activities),
part G of the survey under “Transfers and Money Aid”. The type of bonus, the frequency,
the amount, the co-responsibilities, among other topics are asked about the BDH in twelve
questions. In the present study, only the type of BDH given to mothers will be considered
in the analysis. The exclusion of other types is justified given the mentioned mediators that
the BDH has to influence nutritional outcomes.
A critical part of the design is the initial reconstruction of the eligibility criteria index in
the survey, which is the instrument for the later application of the regression discontinuity
2
For children under 2 years the accurate measure is length-for-age z-scores, though to simplify I refer only
to height
11