Stochastic dominance approach to measuring child development

Tahsin Mehdi

    Research output: Contribution to journalArticlepeer-review


    We measure global performance when it comes to child development outcomes based on Save the Children UK’s Child Development Index, a composite measure comprising health, education, and nutrition. We rely on consistent tests of stochastic dominance efficiency to derive the most optimistic scenario for measured child development where more countries achieve better measured outcomes based on the least variable combination of components of the index. Using the same approach, we also derive the most pessimistic scenario where more countries achieve worse measured outcomes. This approach presents an opportunity to study the sensitivity of the Child Development Index and allows us to better understand the aspects of child development by offering a glimpse into the index components most responsible for driving or hindering improvements in measured child development across countries. To gain a sense of the evolution of the dimensions, we consider four time periods: 1995 - 1999, 2000 - 2004, 2005 - 2010, and 2011 - 2016. We find that in the most optimistic scenario for measured child development, increasing the weight attached to the education dimension over time allows more countries to achieve better measured outcomes. On the other hand, shifting majority of the weight towards health results in the most pessimistic scenario. These results indicate that improvements in children’s education outcomes have outpaced health and nutrition. That is, relative to health and nutrition, more countries find it easier to achieve better education outcomes.
    Original languageEnglish
    Pages (from-to)1567-1588
    JournalChild Indicators Research
    Issue number5
    Early online date24 Aug 2018
    Publication statusPublished - 1 Oct 2019


    Dive into the research topics of 'Stochastic dominance approach to measuring child development'. Together they form a unique fingerprint.

    Cite this