Measuring a hidden population: A novel technique to estimate the population size of women with sexual violence related pregnancies in South Kivu Province, Democratic Republic of Congo

Abstract

Successive sampling (SS)–population size estimation (PSE) is a technique used to estimate the sizes of hidden populations using data collected in respondent-driven sampling (RDS) surveys. We assess past estimations and use new data from an RDS survey to calculate a new PSE. In 2012, 852 adult women in South Kivu Province, Democratic Republic of Congo, who self-identified as survivors of sexual violence, resulting in a pregnancy, since the start of the war (in 1996) were sampled using RDS. We used imputed visibility, enrollment order, and prior estimates for PSE using SS-PSE in RDS Analyst. Prior estimates varied between Congolese local experts and researchers. We calculated the PSE of women with a sexual violence-related pregnancy in South Kivu using researchers’ priors to be approximately 17,400. SS–PSE is an effective method for estimating the population sizes of hidden populations, useful for providing evidence for services and resource allocation. SS–PSE is beneficial because population sizes can be calculated after conducting the survey and do not rely on separate studies or additional data (as in network scale-up, multiplier, and capture-recapture methods).

Katherine R. McLaughlin

Assistant Professor of Statistics

My research interests include sampling methods and social network analysis.