Amelia M. Haviland
Professor of Statistics and Health Policy
Amelia Haviland joined the Heinz College faculty as an Associate Professor in 2011.
Prior to this position she was a Senior Statistician at the RAND Corporation where she worked since receiving her joint PhD in statistics and public policy from Carnegie Mellon University in 2003. She is the recipient of the Anna Loomis McCandless Chair, a Thomas Lord Distinguished Scholar Award (Institute for Civil Justice, RAND), a MacArthur Fellowship for Younger Scholars (MacArthur Research Network on Social Interactions and Economic Inequality), and a Wray Jackson Smith Scholarship (Section on Government Statistics, American Statistical Association).
Dr. Haviland’s research focuses on causal analysis with observational data and analysis of longitudinal and complex survey data applied to policy issues in health and criminology. For example, she recently led a team of researchers assessing the effects of high deductible account-based health insurance plans on health care costs, use, and disparities in the most comprehensive study on the topic to date. Other health policy work involves assessing mechanisms for health disparities for Medicare recipients and exploring connections between patient safety and recent reductions in medical malpractice claims. An example of her work in criminology is methodological work extending group-based trajectory modeling (semi-parametric longitudinal mixture models) to address causal questions with application to assessing the effect of gang membership on violent delinquency. She currently serves on the National Research Council Panel tasked with assessing the research evidence on whether there is a deterrent effect of the death penalty. This and other work of Dr. Haviland’s has been published in journals such as Psychometrika, Psychological Methods, Review of Economics and Statistics, Journal of Human Resources, Survey Methodology, Criminology, Health Affairs, Health Services Research, Medical Care, and the Forum for Health Economics and Policy.