Beyond Monetary Poverty Analysis: The Dynamics of Multidimensional Child Poverty in Developing Countries

Chronically poor and least poor households are determined based on monthly household expenditure per capita.

Child-level questionnaire contains information on school, time use, health, social support, feelings and attitudes, parents and household issues, perception of future, environment, and household wealth, and the Peabody Picture Vocabulary Test (PPVT).

The participating child refers the child that the Young Lives interviews in a household.

Alkire and Foster (2011) use income, health, health insurance, and years of schooling to estimate multidimensional poverty.

Narayan et al. (2000) include material well-being, bodily well-being, social well-being, security, freedom of choice and action, and psychological well-being to measure poverty.

Nussbaum (2000) provides a list of basic functional capabilities a human being requires for existence. It includes life, health, integrity, senses imagination, and thought, emotions, practical reason, affiliation, other species, play, control over one’s environment, and material.

Stiglitz, Sen, and Fitoussi (2009) select eight dimensions to measure multidimensional subjective and objective well-being. These are material living standards, health, education, personal activities including work, political voice and governance, social connectedness and relationship, environment, and economic and physical insecurity.

Biggeri et al. (2006) choose 14 capabilities to measure child poverty. These dimensions are life and physical health, love and care, mental well-being, bodily integrity and safety, social relations, participation, education, freedom from economic and non-economic exploitation, shelter and environment, leisure activities, respect, religion and identity, time-autonomy, and mobility.

UNICEF uses seven dimensions including nutrition, health, water, sanitation, shelter, education and information to measure child poverty in developing countries (Gordon et al. 2003).

UNICEF-ECLAC measures child poverty based on six dimensions including nutrition, water, sanitation, shelter, education, and information (UNICEF-ECLAC 2010).

For the European Union, six dimensions including nutrition, clothing, education, child development, information, and housing are selected for children below minimum compulsory school age. For children under 16, leisure is included. Child development is replaced with social. For children aged 17 and 18, clothing, activity, leisure and social, healthcare access, information, and housing are selected (Chzhen and Neubourg 2014).

Note that WHO recommends height-for-age, weight-for-age, and weight-for-height z-scores to measure stunting, undernourishment, and wasting respectively. Weight-for-age z-score is a good indicator of malnourishment for children under the age of 24 months but not for children over the age of 10 because it can misidentify children experiencing the pubertal growth spurt as children with excess weight. Weight-for-height measures short-term effects of negative environments such as diseases or changes in calorie intake (Setboonsarng 2005). Young Lives, however, did not report weight-for-age z-scores for children in the older cohort and weight-for-height z-scores for both cohorts. Height-for-age z-scores was selected as a health indicator available for both the young and older cohorts. It reflects cumulative effects of nutrition and health condition.

Tetrachoric correlations assumes bivariate normality for latent continuous variables underlying dichotomous variables (Agresti 2010).

Coefficients between indicators across the dimensions are less than 0.3505 in the young cohort and 0.3524 in the older cohort. Within the same dimension, the highest coefficients are 0.7861 in the young cohort and 0.7718 in the older cohort.

When using panel data, education indicators may capture a cohort effect, not deprivation in education. If school enrollment measures educational deprivation, it may identify older children who start to leave school by reaching the age of 15 in 2009 as the ones who are deprived in the education dimension. Years of schooling show a limited variation within the cohort over time. Vocabulary test scores were recorded for both cohorts in 2006 and 2009, however, there is no official age-specific cut-off point for the vocabulary test to measure children’s educational deprivation. The same test sets have been used regardless of children’s age. It implies that a child who repeated the same grade multiple times or who started school one or more years later may get a lower score than other children in the same cohort and be continuously identified as the one deprived in the education dimension.

A child is considered deprived in the education dimension if the child is not enrolled in a preschool or in a formal school when his or her age is five years old or above.

According to United Nations Educational, Scientific and Cultural Organization (UNESCO), the compulsory and primary school ages are 6 years with a 9-year duration in India and Vietnam, 5 years with a 12-year duration in Peru, and 7 years with an 8-year duration in Ethiopia.

Based on the analysis of multidimensional poverty, conditional cash transfer programs, regional development plans, and public services have been implemented, not only to mitigate economic hardships, but also to improve income-generating abilities, living standards, and community infrastructure (OPHI, 2013b).