Michael R. Elliott
Department of Biostatistics
University of Michigan School of Public Health
M4041, SPH II
1420 Washington Heights
Ann Arbor, MI 48105
Voice: (734) 647-5160 Fax: (734) 763-2215
Survey Methodology Program
Institute for Social Research
Rm 4065, 426 Thompson St.
Ann Arbor, MI 48106
Voice: (734) 647-5563 Fax: (734) 764-8263
Welcome to Michael Elliott's homepage. I am an associate professor in the
Department of Biostatistics at the University of Michigan
School of Public Health, and an associate research professor at the Institute for Social Research.
Below find links to classes that I teach, my current
research interests, and useful or amusing Web links.
Back
to SPH bio page.
Email me: mrelliot@umich.edu
Research Interests
Statistical Methodology
-
Design and Analysis of Sample Surveys
-
Generalized Growth Mixture Models (GGMMs)
Collaborative Research
My collaborative research interests have been organized to some degree around injury
control.
Partners in
Child Passenger Safety uses a known-probability sample of children in
accidents involving claims to State
Farm Insurance Companies to investigate child occupant protection in vehicular
crashes.
(Profiled in the University of Michigan School of Public Health Findings magazine.)
The Social and Behavioral Analysis
Division of the University of Michigan Transportation Research Institute
conducts research that advances understanding of the social and behavioral issues important to transportation, including
developing and testing intervention programs that promote safe driving, expanding the knowledge of
social and behavioral factors related to high-risk driving behavior, and
understanding the relationship between public policy and social and behavioral factors in transportation.
Collaboration with Wharton
Risk Management and Decision Processes Center focuses on the analysis
of the RMP*Info
database, which includes the safety record from 1994-2005 of over 15,000
industrial facilities using above-threshold quantities of 77 toxic or 66
flammable chemicals.
The Philadelphia Gun and Alcohol Study is a major case-control study
assessing the impact of alcohol outlets on risk of firearms injury. The study uses novel recruiting strategies
that attempt to map cases and control into a spatial-temporal pattern to assess risk of injury after
adjusting for confounding due to alcohol outlets being located in business districts with high traffic volumes.
Datasets and other materials for classes are here:
S-Plus Introduction:
Presentation
Code examples
Bootstrapping
exercise
Bootstrapping
code
Missing Data Lectures at the Advanced School and Conference on Statistics and Applied Probability in the Life
Sciences at the Abdus Salam Institute for Theoretical Physics, Trieste, Italy:
Handling Missing Data in Statistical Analyses