Who benefits from network intervention programs? TERGM analysis across ten Philippine low-income communities

Introduction

People living in poverty who lack access to institutions and services need social networks just to get by in their lives. Many programs try to strengthen the networks of the poor but with little research-based evidence (Perkins et al., 2015). Faith-based organizations are large providers of such interventions (Summerskill and Horton, 2015; Green, 2016) and we would like to know who benefits most from them? Do programs that bring together members of low-income communities help socially isolated individuals access social support and resources? Specifically, we are asking what is the role of pre-existing social networks and what are the relationships between change in social networks and participants’ self-reported health, wealth, and faith?

We focus on a program called “Transform” that has been implemented in a standardized format in thousands of communities in multiple countries by International Care Ministries (ICM), a faith-based non-governmental organization. We collected complete longitudinal network data from before and after this 16-week intervention organized in ten low-income communities in the Philippines. The intervention brings together marginalized women and men in each community for weekly sessions that include training on health promotion, income generation, and Christian values. Conditioning on the participants’ pre-existing networks, we analysed the ten community networks both jointly in a common model and also separately. We applied a version of temporal exponential random graph models (TERGMs) to this data to analyse the social network evolution among the intervention participants during the program and the interplay of networks and the participants’ changing personal attributes.