Estimating poverty for refugees in data-scarce contexts: an application of cross-survey imputation

Abstract

The increasing growth of forced displacement worldwide has brought more attention to measuring poverty among refugee populations. However, refugee data remain scarce, particularly regarding income or consumption. We offer a first attempt to measure poverty among refugees using cross-survey imputation and administrative and survey data collected by the United Nations High Commissioner for Refugees (UNHCR). Employing a small number of predictors currently available in the UNHCR registration system, the proposed methodology offers out-of-sample predicted poverty rates that are not statistically different from actual poverty rates. These estimates are robust to different poverty lines, perform well according to targeting indicators, and are more accurate than those based on asset indexes or proxy means tests. They can also be obtained with relatively small samples. We additionally show that it is feasible to provide poverty estimates for one geographical region based on existing data from another similar region.

Using Survey-to-Survey Imputation to Fill Poverty Data Gaps at a Low Cost: Evidence from a Randomized Survey Experiment

“…Given these samples, both the estimates and their 95 percent CI almost all fall inside the 95 percent CI around the true rate. The finding that the minimum sample size for the target survey is around 1,000 observations is consistent with previous evidence (Dang and Verme, 2023) and corroborates the findings in Section 4.…”

“…The analytical framework features the poverty imputation method developed in Dang et al (2017), which builds on the survey-to-census poverty mapping method in Elbers et al (2023). 10 The method has been applied to fill poverty data gaps for general household populations in various low-and middle-income countries from different regions, including India, Jordan, Tunisia, Vietnam, and Sub-Saharan African countries (Beegle et al, 2016;Dang and Lanjouw, 2023), as well as to estimate poverty within refugee sub-populations (Dang and Verme, 2023). Most recently, the method is employed in to analyze data from 14 multi-topic household surveys from Ethiopia, Malawi, Nigeria, Tanzania, and Vietnam.…”

“…The target survey can be either an existing, non-consumption survey, such as a Demographic and Health Survey (DHS) or a labor force survey (Stifel and Christiaensen, 2007;Douidich et al, 2016), or a purposefully commissioned survey that only collects the requisite predictors. Recent applications also include sourcing the data for the requisite predictors from administrative records to impute poverty for hard-to-reach refugee populations (Altindag et al, 2021;Dang and Verme, 2023), or phone call detail records to target the ultra-poor (Aiken et al, 2023).…”