As used in this article, the following terms have the following meanings:

(a) “Ballot” means original, voter-verifiable paper ballots, including voter-marked paper ballots whether marked manually or via a ballot marking device or system, and, where direct recording electronic (DRE) voting systems are used, the voter-verifiable paper audit trail (VVPAT). It does not mean electronic versions of ballots, digital images of ballots, or paper printouts of ballot images or digital cast vote records.

(b) “Ballot-level comparison audit” means a type of risk-limiting audit that involves both of the following steps:

(1) The elections official uses an independent system to verify that the cast vote records created by the voting system or ballots created independent from the tally or ballot marking system yield the same election results as those reported by the voting system.

(2) The elections official compares some or all of those cast vote records to a hand-to-eye, human interpretation of voter markings from the corresponding ballot marked by the voter or the voter verified paper audit trail, as defined by Section 19271.

(c) “Ballot polling audit” means a type of risk-limiting audit in which elections officials examine voter markings on randomly selected ballots seeking strong evidence that the reported tabulation outcome is correct.

(d) “Cast vote record” means an auditable document or electronic record that purports to reflect the selections a voter made on a ballot. It lists the contests on the ballot and the voter’s selections in each of those contests.

(e) “Cross-jurisdictional contest” means an election contest in which ballots were cast in more than one county.

(f) “Electoral outcome” means the winner or winners of an election contest or whether a measure passed. It does not mean the numerical vote totals.

(g) (1) “Partial risk-limiting audit” of a cross-jurisdictional contest in a given county means any procedure that has at least a 95 percent chance of leading to a full manual tally of the votes in that contest on the ballots cast in that county if the electoral outcome is incorrect in part in that county. If a partial risk-limiting audit leads to such a full manual tally, the vote totals according to that manual tally shall replace the originally reported vote totals for that contest in that county.

(2) An electoral outcome of a cross-jurisdictional contest is incorrect in part in a given county if the tabulation error for that contest in the county, extrapolated in proportion to the number of ballots in the contest cast in that county compared to the total number of ballots cast in the entire contest, would alter the overall electoral outcome of the contest.

(3) The tabulation error of a contest in a county is the difference between the reported vote tally for the contest in that county and what a tally based on manual tally of the votes cast in that county would show, if the manual tally ascertains voter intent by eye, directly from the voter-verifiable paper records.

(h) “Risk-limiting audit” means a post-election process that involves hand-to-eye, human inspection of ballots in such a manner that if a full manual tally of all the ballots cast in the contest would show different outcomes than the results reported by the voting system, there is at most a five percent chance that the post-election process will not lead to such a full manual tally. If this post-election process does lead to a full manual tally, the winner or winners according to that full manual tally replace the winner or winners as reported by the voting system if they differ.

(Added by Stats. 2018, Ch. 913, Sec. 2. (AB 2125) Effective January 1, 2019. Repealed as of January 1, 2021, pursuant to Section 15365.)