Disentangling the relation between wealth and contraceptive use in India: a multilevel probit regression approach

Abstract

This paper discusses the poverty-wealth impact on contraceptive use both at individual and community levels. A multilevel probit regression framework was set up to explain the use of contraceptive methods. We use data from the Indian 2005–06 National Family Health Survey (NFHS). It contains data on both the socio-demographic and health characteristics of Indian women, as well as an assembled measure of household wealth: the wealth index. The individual-level model confirms the association between contraceptive use and wealth. When controlling the community level, i.e. where the women live, the multilevel regression results show it has a significant effect on contraceptive use, explaining 19 % of the total variance. Finally, decomposing the wealth index at individual and community levels (average community wealth; individual centered wealth values), the effect of the community poverty-wealth level is significant and positive. To sum up, this framework of nested models reveals the significant effect of the community’s poverty-wealth dimension on each woman’s decision about contraception and suggests that the longstanding urban-rural differential in the contraceptive prevalence is mainly poverty-wealth driven.