Heterogeneity in Voter List Maintenance Practices: A Study of Florida CountiesWorking Paper No.: 137
Date Published: 2020-04-22
Jian Cao, California Institute of Technology
Seo-young Silvia Kim, California Institute of Technology
R. Michael Alvarez, California Institute of Technology
How do we ensure the accuracy and integrity of a statewide voter registration database, which often depends on aggregating decentralized, sub-state data with diﬀerent list maintenance practices? We present Bayesian multivariate multilevel model to account for common patterns in local data while detecting anomalous patterns, using Florida as our example. We use monthly snapshots of state’s voter database to estimate countywide change rates for multiple response variables (e.g., changes in voter’s partisan aﬃliation), and then jointly model their changes. We show that there is much heterogeneity in how counties manage voter lists, resulting in very diﬀerent patterns in additions, deletions, or changes of records. Our method identiﬁes several Florida counties with anomalous rates of changes in the 2016 election.