Comparison of Different Child Poverty Measures
: Empirical Evidence from Indonesia

  • Erlangga Agustino Landiyanto

Student thesis: Doctoral ThesisDoctor of Philosophy (PhD)

Abstract

The aim of this thesis was to investigate the extent and nature of child poverty in Indonesia through comparing different child poverty measures. Further, given different conceptualisations of poverty, a secondary aim was to consider which approach to assessment of child poverty is most useful in the Indonesian context.
The aim was met using secondary analysis of existing data sets, the Indonesian family life survey (IFLS) 5 and IFLS East, conducted in 2014 and 2012 respectively. This thesis developed, used, and compared monetary and multidimensional measures (absolute and relative deprivation) of child poverty. The analysis of monetary poverty estimated poverty based on expenditure relative to absolute and relative poverty thresholds. The absolute deprivation measure used a human rights-based approach to identify domains and indicators. The relative deprivation measure identified the preliminary list of items based on domains and indicators from previous studies in Indonesia and then selected reliable, valid and additive items for the analysis.
The final sample consisted 13,192 households and 21,396 children. Using these analyses, 8.7 million (10.4%) of Indonesian children live in absolute monetary poverty, 14.5 million (17.3%) in absolute deprivation and 21.5 million (25.7%) in relative deprivation. Although there are 28.4 million children (34%) who are poor according to one of those measures, only 3.2 million (3.9%) of children suffer all three types of child poverty. Subgroup comparison shows children in rural area and outside Java, in non-Muslim households, in households with more assets, and whose household head had lower levels of education experienced significantly higher rates of poverty (estimated by all measures) than their peers.
There is no best single child poverty measure. However, sensitivity and specificity analysis supported by the receiver operating characteristics (ROC) curve suggested both absolute deprivation and relative deprivation were robust measures of child poverty while monetary measure was less appropriate.
Date of Award23 Jan 2019
Original languageEnglish
Awarding Institution
  • The University of Bristol
SupervisorPatricia J Lucas (Supervisor) & Sebnem Eroglu-Hawksworth (Supervisor)

Cite this

Comparison of Different Child Poverty Measures: Empirical Evidence from Indonesia
Landiyanto, E. A. (Author). 23 Jan 2019

Student thesis: Doctoral ThesisDoctor of Philosophy (PhD)