The goal of this work is to use the 2019 Behavioral Risk Factor Surveillance System (BRFSS) data produced by the Center for Disease Control and Prevention to determine the main non-genetic factors that can be used to predict alcoholism. Additionally, we plan on utilizing multiple reweighting techniques on the BRFSS data in order create a nationally representative sample. Then we will analyze how the differently reweighted data varies from one predictive model to another. After examining predictive models for alcoholism using unweighted, weighted, and reweighted data in the form of Raking and Matching, the results show little variation between the 4 predictive models. Further, we determined that those who smoke, are in the upper bracket of income level, and are healthy and active have the greatest chance of being an alcoholic.


Frazier, Marian




Applied Statistics | Multivariate Analysis | Statistical Models | Statistical Theory

Publication Date


Degree Granted

Bachelor of Arts

Document Type

Senior Independent Study Thesis



© Copyright 2021 Trevor Krisan