Using satellite data to delineate slum and non-slum sample domains for an urban population survey in Uttar Pradesh, India.

Abstract

We describe our approach to develop a sampling frame for slum and non-slum areas using satellite data for an impact evaluation of a reproductive health and family planning program in six cities of Uttar Pradesh, India. The methods used in the paper were developed as part of the Measurement, Learning & Evaluation project (MLE) of the Urban Health Initiative (UHI) program in Uttar Pradesh (UP), India. Increasing access and use of high-quality modern family planning services, particularly among the urban poor, is a primary goal of UHI. Family planning use is associated with lower maternal and child mortality rates, and unintended pregnancies (Cates, Contraception 81(6):460–461, 2010; Cleland et al., Lancet 368(9549):1810–1827, 2006, Lancet 380(9837):149–156, 2012). To evaluate the UHI program in UP, India, a total of six cities were selected for the impact evaluation: Agra, Aligarh, Allahabad, Gorakhpur, Moradabad and Varanasi. This paper describes the methodology developed to design a representative population-based sample of slum and non-slum areas to be surveyed before, during and at the end of UHI program implementation to capture behavior changes among the urban poor, slum and non-slum residents. Geographic data for slums was obtained from the Remote Sensing Applications Center (RSAC). These data were combined with road networks, administrative boundary data and remotely sensed imagery to delineate the study cities into slum and non-slum primary sampling units using Geographic Information Systems.