Caste and religious diversity on formation of social capital : A field survey in India and measurement issues

Purpose – The purpose of this paper is to address the impact of caste and religious diversity on individual‐level social capital. Design/methodology/approach – Based on the World Bank's field tested assessment tools, SOCAT, the field survey of the authors' study considers data related to most of the indicators of social capital. This study considers stratified random sampling method. The selection of four municipalities was done by purposive method and the data include information on 440 households from four municipal wards chosen from four municipalities of West Bengal. Out of four municipalities, two are most homogenous (Serampore) and most heterogeneous (Ashoke Nagar Kalyan Garh) along caste dimension and the other two are most homogenous (Purulia) and most heterogeneous (Santipur) religion based municipalities. Caste and religious heterogeneity are measured by an index of ethnic fractionalization, ETHNIC; the index calculates the probability that two randomly‐chosen individuals will not belong to the same group. Findings – Using different measures of social capital, results of the analysis strongly support adverse effect of ethnicity on individual non‐structural social capital. The data set and results imply that for every dimension, the formation of social capital for each individual is higher among same (homogeneous) individuals along caste and religious dimensions in their social sphere. Moreover the level of structural social capital in the form of associations and cognitive form are lower than individual form of non‐structural social capital for each individual, irrespective of interactions between same or different (heterogeneous) types. Originality/value – The study is important in that it tries to explore the impact of diversity along caste and religious dimensions on the non‐structural individual social capital, in an Indian context; whereas all the previous literature focuses on diversity along ethno‐linguistic and racial lines and most of these studies consider social capital at aggregative level.