Audit Plans for CAST and Trinomial Methods

Share:

Description

An audit.plan is returned by CAST.calc.sample, containing details of how to audit for a desired level of confidence. It has a print method for pretty output.

The audit.plan.tri, similarly, is an object that holds information about conduting a PPEB election audit, in particular an audit that will use the trinomial bound to analyze resultant audit data. It is what is returned by the tri.calc.sample method.

Usage

1
2
3
4
5
6
7
8
## S3 method for class 'audit.plan'
print(x, ...)
## S3 method for class 'audit.plan'
is(x)
## S3 method for class 'audit.plan.tri'
is(x)
## S3 method for class 'audit.plan.tri'
print(x, ...)

Arguments

x

An audit plan (or trinomial audit plan).

...

Unused.

Details

Theoretically, auditors will use the plan and go out and generate actual audit data. (You can fake it with simulations–see make.truth.) The audit data should be stored in a new data frame with new vote totals, or overstatements, for the candidates in the audited precincts. To convert from totals to overstatements, use audit.totals.to.OS. You can store that in a elec.data object under “audit”, or keep it seperate.

Author(s)

Luke W. Miratrix

See Also

CAST.calc.sample tri.calc.sample

Want to suggest features or report bugs for rdrr.io? Use the GitHub issue tracker.