Twitter | GitHub | Google Scholar | CV
Makes a mean cornbread
Currently, I am a Statistical Demographer and the Training Director at the Center for Studies in Demography & Ecology (CSDE). My roles include support and training for CSDE students pursuing demographic and population research and providing consulting in statistical and demographic methods to CSDE faculty affiliates and students. I am passionate about teaching, mentorship, and research applications that improve the lives of people and communities
I recently received my Ph.D. from the Department of Statistics at the University of Washington. I was a fellow of and received training in demographic methods from CSDE. My dissertation was advised by Jon Wakefield. My broad research interests include demographic methods, Bayesian spatiotemporal methods, survey statistics, and the places where all of those things overlap.
Born and raised in Alabama, I received my M.S. in Statistics at Auburn University. My thesis, titled “Group lasso for functional logistic regression”, was advised by Nedret Billor. I also received a B.S. in Actuarial Science and a minor in tailgating from the loveliest village on the plains. War eagle!!!
I was the Project Lead for the 2021 and 2022 cohorts of the Applied Research Fellowship Program supported by the University of Washington’s Population Health Initiative in partnership with CSDE. In 2021, along with a team of three graduate student fellows and two undergraduate student fellows, we used American Community Survey data to estimate changes in the distribution of households within King County by household size, renter or owner status, and family status to better understand the varied types of households and their housing needs. Through expert interviews and use of King County assessor data, we also identify housing vulnerabilities and how they vary by owner or renter status in the face of disaster events such as flooding, earthquakes, and extreme heat. In O2022, we used longitudinal King County Assessment data to look the availability of ohousing bynumber of bedrooms and type. Our dashboard is a work in progress, to be updated soon!
This summer I was the Data Scientist for an amazing project and group of Data Science fellows in eScience’s Data Science for Social Good Program. Led by Dr. Jennie Romich and I, our fellows developed different methods for aggregating individual records from the Washington Merged Longitudinal Administrative Data (WMLAD) dataset into households or economic units. Families, households, and economic units are social constructs and incredibly hard to measure, especially for individuals whose support systems do not fit within a traditional heteronormative framework. But, in order to understand poverty and economic insecurity and intergenerational changes in poverty, it is important to be able to quantify earnings at the household level, and not the individual level. Our student’s work can be seen on their project website, and I hope to continue this work in the future.
My thesis, “Spatiotemporal estimation of period child mortality in a low and middle income countries context”, focuses on statistical and demographic methods to improve subnational estimation of child mortality in countries that rely on complex household surveys to provide comprehensive demographic and health information about the population. Specifically, it addresses the following three questions:
Can we combine both design-based direct full birth histories from complex surveys with summary birth histories from censuses and surveys in a discrete space-time smoothing model to make estimates of child mortality at the Admin-2 level?
Can we use model-based methods to estimate child mortality at the Admin-2 level from full birth histories when design-based estimation fails in sparser data settings?