Respondent-Driven Sampling (RDS) is an increasingly common sampling and analysis strategy for studies that recruit hidden, often socially excluded populations. That is, populations that cannot be reached using traditional sampling methods. Similar to snow-ball sampling, RDS relies on peer recruitment. Initial study participants, called seeds, are selected by the research team from the desired population. Each seed is then asked to recruit k (typically between three and five) individuals, who are then asked to recruit up to k individuals, and so on, until the desired sample size is reached. Through leveraging information about participants’ social networks, including adjustments for homophily and unequal probability of recruitment, RDS provides statistically valid and robust estimators of population traits of interest, such as the percentage of individuals who tested positive for COVID-19, or the proportion of individuals with diabetes in the population. RDS has been successfully applied in a variety of public health settings, including persons who inject drugs (PWIDs), transgendered people, and home-care workers. This interactive workshop will present an overview of the appropriate use of rds and highlight some of our recent methodological efforts to determine which analysis strategy is optimal depending on the research question and population of interest. Key questions that the workshop will address: At the end of the workshop, participants will be able to: -Identify research questions and populations for which RDS methods are best suited. -Recognize the methodological strengths, limitations and unique challenges of RDS. -Apply the RDS estimator that is most appropriate for their research context. -Develop a basic understanding of how to analyze RDS data using appropriately adjusted descriptive and inferential methods.