yolo | R Documentation |
This is for measure W in Yolo County, CA, November 2008. The file includes precinct-level reports.
In the actual audit, 6 precincts were selected (see example) and audited by hand-to-eye count by a group of 4 people cross-checking each other. One of the 6 batches had underreported the "yes" votes by 1, and one had overreported the "yes" votes by 1. There were no other errors.
A data frame with 114 observations on the following 8 variables.
Unique identifier for the batches of ballots
The precinct id of the batch
Vote by mail (VBM) or walk-in (PCT)
Number of votes cast in that unit
Number of undervotes (ballots not voted).
Number of overvotes (where someone marked both yes and no).
Reported number of valid ballots marked yes.
Reported number of valid ballots marked no.
Yolo County, CA. Special thanks to Freddie Oakley and Tom Stanionis.
See Stark et al. for papers using this data to illustrate risk-limiting audits of election data.
# Make an elec.data object out of precicnt-level results data(yolo) yolo = elec.data( yolo, C.names=c("y","n","under","over"), tot.votes.col="b" ) # Look at different sample sizes and cuts for setting aside # small precincts CAST.calc.opt.cut( yolo, beta=0.75, stages=1, t=5, plot=TRUE ) print( yolo ) # Get details of the audit plan -- expected work, etc. ap <- CAST.calc.sample( yolo, beta=0.75, stages=1, t=5, small.cut=5 ) print( ap ) # Draw a sample (seed not used for actual audit) CAST.sample(yolo, ap, seed=12345678)
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.