RDS Analyst - HPMRG
In 2013, I co founded the Hard-to-Reach Population Methods Research Group (HPMRG) to improve statistical methodologies to estimate prevalence and population sizes of hard to reach populations using respondent driven sampling (RDS).
One goal of this group was to create a more user friendly, open source software for analyzing data collected with RDS. This software, known as RDS Analyst, is now used widely and has numerous features including:latest estimators
ability to upload data from different file types and to download data into different file types
ability to reuse code
weighted descriptive statistics
regression analysis (linear and logistic) and other advanced analysis techniques
trend analysis
recruitment and population homophily
cross tabs
graphics
and more....
RDS Analyst can be downloaded for free at the HPMRG website. A manual to get started with RDS Analyst can be found on this website. I have taught RDS Analyst to RDS practitioners all over the world. See RDS Analyst slides.
In addition, the group has developed a population size estimation technique known as the successive sampling-population size estimation (SS-PSE). The SS-PSE method uses data normally collected in an RDS survey: social network size, date of enrollment, maximum number of coupons. The estimates use a Bayesian framework (i.e., quantifies uncertainty about unknown quantities by relating them to known quantities) incorporating information about a “guess” or prior knowledge of the size of the sampled population. For more information, please see the following publications
Handcock MS, Gile KJ, Mar CM. Estimating Hidden Population Size using Respondent-Driven Sampling Data. Electron J Stat. 2014; 8(1): 1491–1521. (Download PDF)
McLaughlin KR, Johnston LG, Gamble L, Papoyan A, Grigoryan T. Use and interpretation of population size estimations among hidden populations using successive sampling in respondent driven sampling surveys. JMIR Public Health Surveill. 2019. 5(1):e12034. (Download PDF)
Johnston LG, McLaughlin KR, El Rhilani H, Latifi A, Toufik A, Bennani A, Alami K, Elomari B, Handcock MS. Estimating the Size of Hidden Populations Using Respondent-driven Sampling Data: Case Examples from Morocco. Epidemiology. 2015. 26(6):846–52. (Download PDF)
McLaughlin KR, Handcock MS, Johnston LG, Japuki X, Gexha-Bunjaku D, Deva E, et al. Inference for the Visibility Distribution for Respondent-Driven Sampling. In: American Statistical Association. 2015. (Download PDF)
HPMRG is a collaborative interdisciplinary group of researchers:
Dr. Lisa G. Johnston is an epidemiologist, applied researcher and RDS consultant. Dr. Johnston has twelve years of experience providing supervision and training on using RDS methods and HIV/STI biological-behavioral surveillance survey planning and implementation. Since its creation she has written a manual and has taught RDS Analyst to RDS practitioners all over the world. For copies of slides used in past workshops, see RDS Analyst Slides.
Dr. Krista J. Gile is Assistant Professor of Statistics in the Department of Mathematics and Statistics at the University of Massachusetts, Amherst. Her research focuses on developing statistical methodology for social and behavioral science research, particularly related to making inference from partially-observed social network structures. Most of her current work is focused on understanding the strengths and limitations of data sampled with link-tracing designs such as snowball sampling, contact tracing, and respondent-driven sampling. In particular, her dissertation and recent work focus on understanding the implications of assumptions of current RDS methodology, and on introducing improved estimation strategies for RDS data. For details see her web page.
Dr. Mark S. Handcock is Professor of Statistics in the Department of Statistics at the University of California – Los Angeles. His research involves methodological development, and is based largely on motivation from questions in the social and epidemiological sciences. He has published extensively on survey sampling, network inference, and network sampling methods. He teaches Statistical Analysis of Networkand Sample Survey Techniques. For details see his web page.
Dr. Cori M. Mar is the Director of the Statistics Core at the Center for Studies in Demography and Ecology at the University of Washington. Her duties include providing training in statistical methods, data analysis techniques, and statistical programming. Dr. Mar has taught R in a variety of formats from a 2-3 hour one class introduction to one hour a week through a 10 week course. Dr. Mar has extensive experience as a translator between statisticians and the applied researchers. For details see her web page.
Dr. Ian E. Fellows is a professional statistician based out of the University of California, Los Angeles. His research interests range over many sub-disciplines of statistics, with his dissertation work focusing on new methods in the analysis of social network sampling designs (such as RDS). He has designed statistical user interfaces for both academic and corporate clients, and in 2011 one of his designs won the prestigious John Chambers Award. He is the primary author of RDS Analyst’s graphical user interface.
Katherine R. McLaughlin is Assistant Professor of Statistics in the Department of Statistics at the Oregon State University. She works in many areas of statistics, often with application to social demography and global health. She is an expert in social network analysis, network sampling, survey sampling and social statistics. Her dissertation work focuses on modeling preferential recruitment for RDS and Peer-Driven Interventions. For details see her web page.