Elizabeth Petraglia has spent the past five years working in the fields of survey design and analysis and statistical programming. Before beginning her Ph.D. program, she worked as a SAS programmer and research assistant for a major public policy research firm, primarily performing data cleaning and analysis on large-scale survey data. She is currently a research assistant for Ohio State's Statistical Consulting Service (SCS).

In her own research, she helped develop a Bayesian hierarchical time-series model to predict missing values in the Uniform Crime Reports (UCR) property crime series. Her current dissertation research is on augmenting UCR county-level crime counts with NCVS large-area crime estimates in order to improve the accuracy of county-level crime rates.