Skip to content
theory and code for RLAs for Colorado in June and November 2018
Branch: master
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Type Name Latest commit message Commit time
Failed to load latest commit information.
code
.gitignore
CO18.pdf
CO18.tex
LICENSE.md
Makefile
README.md
ballotPollError.tex
comparisonError.tex
llncs.cls
pbsBib.bib
requirements.txt
splncs04.bst
stratumRisk.tex
subcollections.tex

README.md

Binder

Risk-Limiting Audits by Stratified Union-Intersection Tests of Elections (SUITE)

by Kellie Ottoboni, Philip B. Stark, Mark Lindeman, and Neal McBurnett

Risk-limiting audits (RLAs) offer a statistical guarantee: if a full manual tally of the paper ballots would show that the reported election outcome is wrong, an RLA has a known minimum chance of leading to a full manual tally. RLAs generally rely on random samples. Stratified sampling---partitioning the population of ballots into disjoint strata and sampling independently from the strata---may simplify logistics or increase efficiency compared to simpler sampling designs, but makes risk calculations harder. We present SUITE, a novel RLA method for stratified samples. SUITE tests all possible partitions of outcome-changing error across strata. For each partition, SUITE combines P-values for each stratum's error into a combined P-value. (SUITE is agnostic about the methods for finding stratum-level P-values.) The combined P-value is maximized over all such partitions. The audit can stop if the maximum combined P-value is less than the risk limit. SUITE is immediately useful in Colorado. Voting systems in some Colorado counties (comprising 98.2% of voters) allow auditors to check how the system interpreted each ballot, which allows efficient ballot-level comparison RLAs. (Other counties use ballot polling, which is less efficient.) Extant approaches to conducting an RLA of a statewide contest would require Colorado to make major procedural changes, or would sacrifice the efficiency of ballot-level comparison. In contrast, SUITE requires little change to Colorado's procedures and is substantially more efficient than a statewide ballot-polling RLA. The two strata comprise ballots cast in counties that can conduct ballot-level comparisons, and the rest. Stratum-level P-values can be found using modifications of ballot-polling and ballot-level comparison, derived here. We provide an open-source reference implementation and exemplar calculations in Jupyter notebooks.

Fisher's combination method illustration

Example Notebook 1

Example Notebook 2

You can’t perform that action at this time.