A Quantitative Approach to Reckon Intersectional Inequality: An Application to Health Data

Chakraborty, Achin; Mukhopadhyay, Simantini (2016). 'A Quantitative Approach to Reckon Intersectional Inequality: An Application to Health Data' Paper presented at the annual conference of the HDCA, Tokyo 2016.

abstract Introduction While the last few decades have witnessed a surge in health equity research, most studies have analysed health inequalities along single dimensions of social power, implicitly assuming that these dimensions are inherently separable and mutually exclusive. Only a few have considered the complex interactions of multiple identities on health inequalities using the framework of intersectionality. Intersectionality literature, particularly in the realm of feminist studies, has predominantly used qualitative methodologies. Few quantitative studies have used the regression approach, treating the dependent health variable as categorical and creating a set of dummy variables for each intersecting category (Sen et al., 2009; Mukhopadhyay, 2015; Mukhopadhyay, 2016). By doing so, the studies have examined the differences between groups across the social spectrum, to see if the disadvantage associated with a particular group identity is offset by the advantages stemming from some other identity, for instance whether poor women belonging to upper castes can leverage some benefits (in terms of some health outcome) from their caste identity, as compared to poor women from backward castes. While the purpose of such analysis is undoubtedly important and interesting, there is still no study that measures the contribution of intersectional inter-group inequality in total interpersonal inequality.  Drawing on the literature on the decomposition of income inequality, we attempt to quantify how much of total inequality can be explained by inter-group intersectional inequality.   Data and Methods We decompose inequality in nutritional status of children, measured by the height for age z-scores of children below five years. We use data on 46,655 children from the latest round of the National Family Health Survey of India, conducted in 2005-06. Inequality is measured by the most commonly decomposed measures of the General Entropy Class. However, since the decomposition of total inequality into ‘between group’ and ‘within group’ components is sensitive to the number and relative sizes of the groups under examination, following Elbers et al. (2008), we evaluate the contribution of inter-group intersectional inequality against the benchmark of maximum inter-group inequality, given the number and relative sizes of groups.   Preliminary results Corroborating the inconclusive evidence on sex inequality in nutritional status of children (Mishra et al., 2004; Mukhopadhyay, 2016), we find that the contribution of ‘between sex’ inequality in total nutritional inequality is miniscule. We question if the contribution of sex inequality in total inequality changes when we explore sex at its intersection with other important dimensions of social power, such as caste and economic class. We now consider how much of total inequality is explained by the inter-group intersectional inequality between 12 groups of children (namely, poor ST girls, poor ST boys, poor SC girls, poor SC boys, poor other girls, poor other boys, non-poor ST girls, non-poor ST boys, non-poor SC girls, non-poor SC boys, non-poor other girls and non-poor other boys). However, since the standard decomposition method is sensitive to the number and relative sizes of the groups of interest, the contribution of the ‘between group’ component automatically rises with an increase in the number of groups. We use the method proposed by Elbers et al. (2008), to cull out the contribution of inter-group intersectional inequality in total nutritional inequality.     Conclusion In this paper, we attempt to measure the contribution of inter-group intersectional inequality in total interpersonal inequality in nutritional status of children. Using the approach proposed in the paper, one can make intertemporal or spatial comparisons of inter-group intersectional inequality. Given the well-documented importance of the framework of intersectionality in analysing inequality, the decomposition exercise has important connotations for policy formulation. 

scroll to top