|
Activity Number:
|
532
- Can Statistics Inform Decisions in Social, Economic, and Political Event?
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Wednesday, August 1, 2018 : 10:30 AM to 12:20 PM
|
|
Sponsor:
|
Social Statistics Section
|
|
Abstract #327121
|
|
|
Title:
|
STatistical Election to Partition Sequentially (STEPS) and Its Application in Differentially Private Release and Analysis of Youth Voter Registration Data
|
|
Author(s):
|
Claire Bowen* and Fang Liu
|
|
Companies:
|
University of Notre Dame and University of Notre Dame
|
|
Keywords:
|
DIfferentially Private Data Synthesis;
statistical disclosure limitation;
propensity score;
Kolmogorov-Smirnov statistic;
Universal Histogram;
privacy budget
|
|
Abstract:
|
Voter data is important in political science research and applications such as improving youth voter turnout and predicting the presidential election outcome. Privacy protection is imperative in voter data as this type of data often contains sensitive individual information. DIfferentially Private Data Synthesis (DIPS) techniques produce synthetic data and pseudo individual records in the differential privacy setting, a robust concept for privacy protection. We propose a new DIPS approach called STatistical Election to Partition Sequentially (STEPS) that sequentially partitions data based on the data attributes' statistical differentiability of the variability in the data. Additionally, we propose a Universal Histogram approach to synthesize count data for a general hierarchical histogram and develop a general metric called, SPECKS, to assess the similarity of synthetic data to the actual data. The application of the STEPS pro- cedure on the 2000-2012 Current Population Survey youth voter data suggests STEPS is easy to implement and better preserves the original information than some DIPS approaches.
|
Authors who are presenting talks have a * after their name.
