BEGIN:VCALENDAR VERSION:2.0 PRODID:-//132.216.98.100//NONSGML kigkonsult.se iCalcreator 2.20.4// BEGIN:VEVENT UID:20261001T051750EDT-7549ukxvLB@132.216.98.100 DTSTAMP:20261001T091750Z DESCRIPTION:Krista J. Gile\, PhD\n\nAssistant Professor\, Department of Mat hematics and Statistics\, University of Massachusetts\n\nInference and Dia gnostics for Respondent-Driven Sampling Data\n\nALL ARE WELCOME\n\nAbstrac t:\n\nRespondent-Driven Sampling is type of link-tracing network sampling used to study hard-to-reach populations.  Beginning with a convenience sam ple\, each person sampled is given 2-3 uniquely identified coupons to dist ribute to other members of the target population\, making them eligible fo r enrollment in the study. This is effective at collecting large diverse s amples from many populations. \n\nUnfortunately\, sampling is affected by many features of the network and sampling process\, which complicate infer ence.  In this talk\, I highlight key methodological challenges arising fr om data collected in this manner.  I then introduce key methods for diagno stics and inference in these settings\, and describe new methods under dev elopment.\n\nBio:\n\nKrista J. Gile's research focuses on developing stati stical methodology for social and behavioral science research\, particular ly related to making inference from partially-observed social network stru ctures. Most of her current work is focused on understanding the strengths and limitations of data sampled with link-tracing designs such as snowbal l sampling\, contact tracing\, and respondent-driven sampling.\n\nhttp://p eople.math.umass.edu/~gile/\n\n \n DTSTART:20160315T193000Z DTEND:20160315T203000Z LOCATION:Room 24\, Purvis Hall\, CA\, QC\, Montreal\, H3A 1A2\, 1020 avenue des Pins Ouest SUMMARY:Biostatistics Seminar: 'Inference and Diagnostics for Respondent-Dr iven Sampling Data' URL:https://www.mcgill.ca/epi-biostat-occh/channels/event/biostatistics-sem inar-inference-and-diagnostics-respondent-driven-sampling-data-259383 END:VEVENT END:VCALENDAR