Thomas Lumley - Analysing Multistage Survey Data Using R

Thank you to New York Presbyterian for hosting.

About the Talk:
Multistage surveys run by all levels of government and academic research groups are a rich source of data. Traditionally, their analysis involved specialised software (and even specialised hardware); one aim of survey research in recent decades has been to make survey analysis relatively boring, so that it can be done safely by people with domain expertise. I will talk about the R survey package, which just had its 16th birthday. The package allows for exploratory analysis and modelling, and supports research into more efficient estimation in two-phase samples.

About Thomas:
Thomas Lumley is originally from Australia, and was at the University of Washington in Seattle for 15 years. He is now Professor of Biostatistics at the University of Auckland. His research interests include study design in epidemiology, semiparametric theory, computing and graphics, and genomics. Thomas blogs about statistics in the media at statschat.org.nz and about other topics at notstatschat.rbind.io. He is a Fellow of the Royal Society of New Zealand and the American Statistical Association, and a member of R Core, and has taught short courses on R in thirteen time zones.