SECTION 1. Chapter 54 of the general laws, as appearing in the 2008 Official Edition, is hereby amended by inserting after section 109 the following section:- 

Section 109A.  Audits of election results

Purpose.  The purpose of this section is to provide a routine cross-check on the accuracy of election night vote totals, both for ballots counted originally by hand and for ballots counted by machine. The audit compares reported preliminary (unofficial) election vote totals for certain election contests to hand-to-eye counts of the same ballots and contests in order to independently confirm election outcomes.  In the case that the audit uncovers discrepancies that reach a specified level, a contest-wide hand count is mandated. 

Definitions

“Contest”,  A race in an election between two or more candidates on a ballot or between the Yes and No position of a ballot question

“Draw”,  A random selection of an audit unit or precinct

“Error Rate”,  see subsection (p)

(a)  Contests to be Audited.  Following each primary, general or special election, random hand-count audits shall be conducted for the following offices: US President, US Senator, US Representative, Governor, Secretary of the Commonwealth, State Attorney General and one of the following three state offices chosen by random drawing:  Lieutenant Governor, Treasurer and Auditor.  All statewide initiative and referenda questions shall be audited.  Contests in which only one candidate’s name appears on the ballot shall not be audited.  The state secretary, the governor, or the state auditor may require the audit of additional offices or precincts when anomalous conditions are observed.  In addition, the chief election officer of any municipality may order an audit of any contest within that municipality.

Primaries that take place less than sixty days before a general election may be exempted from the audit at the discretion of the state secretary except when winning candidates are separated by less than 7% of the vote share from the nearest losing candidates. 

(b)  Discretionary Audits.  Any candidate on the ballot who has received at least 12% of vote share, or any  person who has received at least 12% of the vote share in the current election as a write-in, may request that up to four specified additional precincts be counted by hand.

These discretionary audit counts may be requested up to five days after the ordinary audits and their secondary expansions, if any, are completed for contests that are audited under this act.  For contests not being audited under this act the request must be submitted within ten days of the election.

(c)  The number of precincts to audit shall be based on the reported contest-wide Margin of Victory for each audited contest.  The Margin of Victory percentage shall be calculated for each audited contest by dividing the difference between the total numbers of votes received by each of the two leading candidates or between the Yes and No referenda positions by the total number of ballots cast in all the precincts in which the audited contest appeared on the ballot. 

(d) Report of preliminary results prior to random selection. 

1. The preliminary unofficial election results for auditable contests, excluding unprocessed ballots such as provisional and late absentee ballots, shall be immediately reported to the state secretary and shall be provided in written format to members of the public.  The state secretary shall make public the number and percentages of votes received for auditable contests, including blank votes and over-votes,  before the random selection of precincts for the audit.  The state secretary or the auditor shall calculate the Margins of Victory of the apparent winners.

2. Any representative of a candidate, ballot question or non-profit good-government organization present at close of polls may request and receive a printed copy of the signed tally slip/s on election night in order to verify the publicly posted results.

3. A mechanism shall be implemented for timely, accessible public posting, such as on the Internet, of scanned copies of all precinct tally slips to the public no later than 2 years after the implementation of this law.  

4. A mechanism shall be implemented to convert the numeric information contained on the tally slips into a format amenable to mathematical manipulation for purposes of verification by mathematicians and the public no later than 4 years after implementation of this law.

(a) The ballots for all contests susceptible to audit in all precincts shall be confined in secure facilities that prevent single person and single party access until completion of all audits, recounts and legal challenges except when such audits, recounts or legal challenges are taking place in full public view.

(b) Time and Place of the Audit. Each municipality shall designate in advance the location where it will conduct an audit should it be selected for an audit in the random drawing.  The audit shall commence no later than 48 hours following the selection of precincts to be audited.  The general mechanism for scheduling and sequencing such audits in various areas of the state shall be developed by the state secretary in advance of the election as specified in subsection (v) of this section.

(c) Random selection of precincts to be audited.

Statewide contests shall be selected from a pool including all precincts in the state. US Congressional contests may be taken sequentially from the same audit sample.  However, if additional Congressional precinct selections are needed they shall be drawn from a pool including all precincts in a single congressional district.

The precincts to be audited, including any selected for an expansion of said audit, shall be chosen using a random selection process that is fully observable and which can be easily understood to be random by members of the public. 

The time and place of such random selection shall be publicly noticed at least ten days in advance of the election. The time and place of any expanded selection of precincts shall be publicly noticed not less than three days before it occurs.  The drawing shall be open to one observer representing each candidate or referendum position that may be  audited, who shall be positioned where s/he can meaningfully observe the process, and it shall be open to and be observed by the public.  The selection shall take place as soon as possible after the close of the polls but after preliminary results and vote margins have been reported for auditable contests.   In the event that preliminary precinct results are not published by a municipality by 48 hours after the election, such precincts shall be audited in addition to those randomly selected as specified in subsection (k) of this section.  The precincts selected and the municipalities to which they belong  shall be announced aloud as they are selected and shall be recorded in the order they are selected.  The precincts and the order in which they were selected shall be published on the website of the state secretary as soon as possible but not later than fifteen hours after the random selection takes place.  Specific location information shall be added to this website no later than 24 hours after the drawing. 

(d) Sequential selection of precincts to count for different statewide contests.  The precincts chosen to audit statewide contests shall be recorded in the order in which they are drawn.  The same sequence of randomly selected precincts may be used in order they were drawn for all statewide and Congressional contests. The minimum number of precincts that must be audited for each contest shall vary in accordance with the contest-wide Margin of Victory percentages as set forth in Tables One and Two in subsection (k) of this section.  At the discretion of the state secretary, several contests may be grouped together and audited according to the highest number of draws required by the audit of any contest within that group.

(i) Cost. The cost of the audits for all statewide and federal offices shall be the responsibility of the Commonwealth of Massachusetts excluding the salaries of city and town employees for normal hours of work. 

(j) Audit units other than precincts may be used for implementation of this audit if such are determined by the Secretary of the Commonwealth in consultation with experts in the field of election audits to be in general at least as efficient and efficacious as counting entire precincts.

(k) General protocol for random selection of precincts to be audited.

1.The audit shall follow a protocol whereby, for each audited race or ballot question, the chance that a precinct will be selected for inclusion in the random sample shall be proportional to the number of ballots cast therein, or, if the number of ballots cast cannot be provided in time, shall be proportional to the number of registered voters in each precinct as of a single specified date within the month preceding the election that shall be used for all precincts in the state. 

2.A draw with replacement method shall be used, such that, once a precinct has been drawn it shall immediately be made available to be randomly selected again in subsequent draws.  In such a case such precinct shall be counted manually only once but if tabulation is necessary it shall be included as many times as it is drawn. 

3.The number of precincts to be selected randomly for hand counting shall be determined by the size of the margin between the winning candidate and the top losing candidate or between the Yes and No votes on a ballot question as set forth in the following tables. .

TABLE ONE: For statewide contests 

draw 13 precincts draw when the margin is equal to or greater than 15%

draw 17 precincts when the margin is at least 12% but under 15%

draw 21 precincts when the margin is at least 10% but under 12%

draw 24 precincts  when the margin is at least 9% but under 10%

draw 27 precincts when the margin is at least 8% but under 9%

draw 31 precincts when the margin is at least 7% but under 8%

draw 37 precincts when the margin is at least 6% but under 7%

draw 44 precincts when the margin is at least 5% but under 6%

draw 49 precincts when the margin is at least 4.50% but under 5%

draw 52 precincts when the margin is at least  4.25% but under 4.5%

draw 56 precincts when the margin is at least 4.00% but under 4.25%

draw 60 precincts when the margin is at least 3.75% but under 4.00%

draw 64 precincts when the margin is at least 3.50% but under 3.75%

draw  69 precincts when the margin is at least 3.25% but under 3.50%

draw  75 precincts when the margin is at least 3.00% but under 3.25%

draw  78 precincts when the margin is at least 2.90% but under 3.00%

draw  80 precincts when the margin is at least 2.80% but under 2.90%

draw  83 precincts when the margin is at least 2.70% but under 2.80%

draw  87 precincts when the margin is at least 2.60% but under 2.70%

draw  90 precincts when the margin is at least 2.50% but under 2.60%

draw  94 precincts when the margin is at least 2.40% but under 2.50%

draw  98 precincts when the margin is at least 2.30% but under 2.40%

draw 103 precincts when the margin is at least 2.20% but under 2.30%

draw 108 precincts when the margin is at least 2.10% but under 2.20%

draw 113 precincts when the margin is at least 2.00% but under 2.10%

draw 119 precincts when the margin is at least 1.90% but under 2.00%

draw 126 precincts when the margin is at least 1.80% but under 1.90%

draw 134 precincts when the margin is at least 1.70% but under 1.80%

draw 142 precincts when the margin is at least 1.60% but under 1.70%

draw 152 precincts when the margin is at least 1.50% but under 1.60%

draw 157 precincts when the margin is at least 1.45% but under 1.50%

draw 163 precincts when the margin is at least 1.40% but under 1.45%

draw 169 precincts when the margin is at least 1.35% but under 1.40%

draw 175 precincts when the margin is at least 1.30% but under 1.35%

draw 182 precincts when the margin is at least 1.25% but under 1.30%

draw 190 precincts when the margin is at least 1.20% but under 1.25%

draw 198 precincts when the margin is at least 1.15% but under 1.20%

draw 208 precincts when the margin is at least 1.10% but under 1.15%

draw 217 precincts when the margin is at least 1.05% but under 1.10%

draw 228 precincts when the margin is at least 1.00% but under 1.05%

draw 241 precincts when the margin is at least .95% but under 1.00%

draw 254 precincts when the margin is at least .90% but under .95%

draw 269 precincts when the margin is at least .85% but under .90%

draw 286 precincts when the margin is at least .80% but under .85%

draw 305 precincts when the margin is at least .75% but under .80%

draw 327 precincts when the margin is at least .70% but under .75%

draw 352 precincts when the margin is at least .65% but under .70%

draw 382 precincts when the margin is at least .60% but under .65%

draw 417 precincts when the margin is at least .55% but under .60%

draw 459 precincts when the margin is at least .50% but under .55%

draw 510 precincts when the margin is at least .45% but under .50%

draw 574 precincts when the margin is at least .40% but under .45%

draw 656 precincts when the margin is at least .35% but under .40%

draw 766 precincts when the margin is at least .30% but under .35%

draw 919 precincts when the margin is at least .25% but under .30%

when the margin is 0.2 % or less a full recount shall be conducted

TABLE TWO: For U.S. Congressional contests

draw 5 precincts when the margin is equal to or greater than 20%

draw 7 precincts when the margin is at least 15% but under 20%

draw 9 precincts when the margin is at least 12% but under 15%

draw 11 precincts when the margin is at least 10% but under 12%

draw12 precincts when the margin is at least 9% but under 10%

draw 14 precincts when the margin is at least 8% but under 9%

draw 16 precincts when the margin is at least 7% but under 8%

draw 22 precincts when the margin is at least 6% but under 7%

draw 25 precincts when the margin is at least 5% but under 6%

draw 28 precincts when the margin is at least 4.50% but under 5%

draw 32 precincts when the margin is at least 4.00% but under 4.50%

draw 35 precincts when the margin is at least 3.50% but under 4.00%

draw 38 precincts when the margin is at least 3.00% but under 3.50%

draw 42 precincts when the margin is at least 2.70% but under 3.00%

draw 49 precincts when the margin is at least 2.30% but under 2.70%

draw 57 precincts when the margin is at least 2.00% but under 2.30%

draw 63 precincts when the margin is at least 1.70% but under 2.00%

draw 67 precincts when the margin is at least 1.60% but under 1.70%

draw 71 precincts when the margin is at least 1.50% but under 1.60%

draw 76 precincts when the margin is at least 1.45% but under 1.50%

draw 79 precincts when the margin is at least 1.35% but under 1.45%

draw 82 precincts when the margin is at least 1.30% but under 1.35%

draw 85 precincts when the margin is at least 1.25% but under 1.30%

draw 88 precincts when the margin is at least 1.15% but under 1.25%

draw 91 precincts when the margin is at least 1.25% but under 1.15%

draw 99 precincts when the margin is at least 1.00% but under 1.05%

draw 109 precincts when the margin is at least 1.05% but under 1.00%

draw 114 precincts when the margin is at least 1.00% but under .1.05%

draw 121 precincts when the margin is at least .95% but under .1.00%

draw 127 precincts when the margin is at least .90 but under 1.00%

draw 143 precincts when the margin is at least .80% but under .90%

draw 153 precincts when the margin is at least .75% but under ..80%

draw 164 precincts when the margin is at least .65% but under .70%

draw 176 precincts when the margin is at least .70% but under .75%

draw 191 precincts when the margin is at least 60% but under .65%

draw 209 precincts when the margin is at least 55% but under .60%

draw 230 precincts when the margin is at least 50% but under .55%

draw 255 precincts when the margin is at least .45% but under .50%

draw 287 precincts when the margin is at least .40% but under .45%

draw 328 precincts when the margin is at least .35% but under .40%

when the margin is 0.3 % or less a full recount shall be conducted

1. A full statewide hand-count shall be conducted for any audited statewide contest with a Margin of Victory of 0.2% percent or less or U.S. Congressional race with a Margin of Victory of 0.3% or less.

2. The state secretary, in consultation with persons with peer-reviewed expertise in statistical protocols and election auditing, may substitute another method of random selection and auditing if it is at least as transparent as the one described herein and at least as effective in ensuring that a full hand recount will occur whenever such a recount would reverse the preliminary outcome reported by the voting system.  

(l) Audit report:  Comparison of precinct audit results to original reported results.   Upon completion of the audit in each municipality the registrars  shall record each contest audited in each precinct on a separate audit report form provided by the state secretary for easy comparison of the following:

1. the audit tallies for each candidate or each yes and no vote and the blank votes and over votes as recorded on the official audit forms pursuant to paragraph (w) 4) of this section 

2. the election night tallies reported for each candidate or each yes and no vote and the blank votes and over-votes for the same ballots that were counted in the audit sample

3. the differences between the audit tallies and the election night tallies for each candidate, yes and no question and blank votes ; the differences should be marked with "+" when the audit result is higher or with "-" when the reported result is higher.

4. The number of votes for each contest for which voter intent is discernible but that are not marked by the voter according to the instructions provided to the voters under Section 48 of Chapter 54, that were identified under (w) (6) of this section. and any explanatory notes related thereto.

5. Any additional notes on perceived causes for discrepancies as provided for in section 48 chapter 54

(m) Reporting audit results. The registrars shall submit the official audit form or forms described in paragraph (l) of this section to the state auditor  who shall make public the information in these reports as soon as practicable thereafter in both human- and machine-readable formats, such as a spreadsheet or comma-separated-value (CSV) file.

(n) Comparison of contest-wide audit sample to preliminary contest-wide results.  Upon receipt of the results of the manual audit counts from the audited precincts, the state auditor shall calculate the total number of votes for candidates, yes and no referenda positions, blanks votes and over-votes in each audited contest as recorded in the manual audit count pursuant to subsection (w) 4 and shall compare them to the corresponding totals reported directly after the election for the same candidates and ballot questions on the same ballots and shall publish any resulting discrepancies.  The auditor shall calculate the error rate and determine whether further hand-counts are required pursuant to subsection (p) of this section.

(o) If after one or more expansions of the audit sample discrepancies in the expanded audit indicate a substantial possibility, as specified in subsection (p), of a different prevailing candidate or outcome than the outcome originally reported in the preliminary election night count, a complete hand count of all ballots cast on which the above contest or contests appeared shall be conducted.  The results shall be reported within sufficient time to meet state and federal election deadlines for final certification but no winner shall be certified until 24 hours after all audits or recounts for that contest have been completed.

If there is a successful petition for a recount pursuant to section 135 of chapter 54 the audit counting for that contest shall and cease.  The state secretary shall determine whether any ballots already counted in the audit need to be recounted.

(p)  The error rate for discrepancies shall be calculated as follows:

The Error Rate equals the vote share as reported on election night for the same ballots that were counted in the audit sample for the first place candidate minus his/her vote share as counted by hand in the audit sample subtracted from the vote share as reported on election night for the same ballots that were counted in the audit sample for the second place candidate minus his/her vote share as counted by hand in the audit sample.

This may also be expressed as (E2/EO - A2/AO) - (E1/EO - A1/AO) where

E1 is Election night vote count for the first place candidate for all ballots selected in the audit sample

E2 is Election night vote count for the second place candidate for all ballots selected in the audit sample

A1 is Audit sample hand-count for the first place candidate

A2 is Audit sample hand-count for the second place candidate

EO is All ballots cast for the office as reported by the preliminary election results

AO is All ballots cast for the office as reported by the audit hand counts

“Candidate" here means a person or a Yes or No for a referendum vote.

A negative error rate indicates that the winning margin decreases.

A positive result confirms the election outcome.

Either or both of the following discrepancies  in the initial audit sample shall trigger further hand counts:

    1. If the contest-wide error rate between the reported election night results and the hand counted audit sample reaches 20 % of the election night Margin of Victory and/or

    2. If the difference in any one precinct between the reported election night results and the hand counted audit results reaches 50 votes an expanded random sample shall be selected and counted that requires three times the number of draws as did the initial sample.

Otherwise the audit counting may be deemed to be complete.

Either or both of the following discrepancies in the expanded audit sample just described shall trigger further hand counts:

    1. If the contest-wide error rate between the reported election night results and the second hand counted-audit sample reaches 30% of the election night victory Margin of Victory, and/or

    2. If the difference in any four precincts between the reported election night results and the hand counted audit results reaches 50 votes a full 100% recount shall be conducted.  Otherwise, at the discretion of the state auditor, the audit counting may be deemed complete.  Notwithstanding this, at the discretion of the state auditor, a new audit sample may be selected and counted that requires at least three times as many draws as the initial sample, in lieu of a full recount.  In such a case, if the contest-wide error rate between the newly expanded combined audit sample and the preliminary election night results reported for the same ballots reaches 50% of the election night victory Margin of Victory a full hand recount shall be conducted.  Otherwise, at the discretion of the state auditor, the audit counting may be deemed complete.

(q) Vote of record.  If there is a discrepancy between a manual count and originally reported tallies, where the original results were obtained using electronic equipment, the manual count of the official paper ballots shall be the official vote of record. 

(r) Archiving audit results. The results of audits and hand-counts, as well as the corresponding data for the originally reported tallies, shall be made publicly available on a precinct-by-precinct basis both in hardcopy and in electronic file format, and shall be archived indefinitely, but for  no less than sixteen years, in order to facilitate research and to provide for continuous improvement of election procedures.

(s) Analysis of discrepancies.  Discrepancies between the hand-counted audit and the original vote tally shall be analyzed in consultation with other experts in the field of election audits to ascertain their cause.  The state secretary shall oversee the analysis and shall publish and make available online the findings within 180 days.

(t) Analyzing discrepancies by machine model.  Whenever the difference between the hand-counted audit and the originally reported tally is more than one percent for any particular machine model, the discrepancy between the hand-counted audit and the original vote tally shall be analyzed to ascertain the cause of the discrepancy and to discern patterns, if any, between counts by each machine model and hand counts.  The state secretary shall oversee the analysis and shall publish and make available online the findings within 180 days.

(u) Training and deployment of audit personnel.  Training for audit personnel shall be based on uniform written and visual training materials issued by the state secretary.  In consultation with municipal clerks or their representatives  the state secretary shall develop a plan for flexibly and timely deploying trained audit personnel to perform audits in precincts selected for audits in various areas in the state.  Personnel may be drawn from pools of workers within certain areas or districts and may be deployed sequentially to perform audits in various locations.

(v) The state secretary shall promulgate procedures to carry out this act and shall solicit public comment for at least 45 days.  The secretary shall revise such procedures as necessary to respond to public comment and shall post final procedures on the state website at least 30 days in advance of each election to which they apply.  These procedures shall be revised, with opportunity to comment, periodically as needed, but at least every four years. 

Written procedures shall

    1. describe the random selection process for precincts to be audited

    2. set up a mechanism for municipalities to inform the state secretary, in advance, of the exact location they will provide if any of their precincts are selected for audits and for informing municipalities and candidates when they have been selected for an audit

    3. set up a mechanism to coordinate the times of audits after precincts have been selected and to report such to the public

    4. coordinate audits and recounts for optimum efficiency and to avoid duplicate counting when that is desired and possible

    5. ensure secure multi-partisan chain of custody for all ballots and election equipment, with signed access, to prevent single person or single party access to any ballots pertaining to an election until after all audits, recounts and legal challenges have been completed

    6. set standards for tamper evident seals for elections, audits and recounts

    7. set standards for tamperproof containers and for their secure storage

    8. describe ballot counting and sorting methods that require that each counter, sorter or recorder to be checked by a person of an opposing party

(w)  Audit procedures.   Audits shall be conducted publicly in a similar manner to the procedures for counting ballots in Section 105 of Chapter 54.  In addition, audits shall follow these procedures: 

    1. each precinct audit will be presided over by two registrars of voters from the two leading political parties who shall examine such election records as they deem necessary including but not limited to spoiled and unused ballots, absentee ballots and envelopes, provisional and overseas ballots as available,  and the poll register;

    2. observers representing candidates and ballot questions shall be positioned and may move about if needed so that they can clearly and comfortably observe every aspect of the proceedings, including marks on ballots, seals, signatures, documents and forms pertaining to the audit and to the election.

    3. upon commencing an audit of its contents the registrars shall examine the signatures and the seals securing ballot containers before they are opened and compare them to the signatures and seal numbers recorded after close of polls in the Precinct Log described in paragraph (y) 3 of this section and determine whether either have been disturbed and if so shall record this in the Precinct Log; 

    4. audit personnel shall count and record the total number of ballots cast at the precinct on Election Day, including the unused and spoiled ballots, and count the total votes, blank votes, over-votes and total write-ins for each candidate and the Yes and No votes and blanks for each ballot question on those ballots and record such on official forms provided by the state secretary;

    5. audit personnel shall compare the information on the signed machine tally slips produced at the precinct at close of polls to the vote totals report submitted to the state secretary as required in subsection (d) 1) of this section and shall record discrepancies, if any, on the appropriate audit form ; 

    6. voter intent shall be the standard for counting votes; 

    7. in order to assist with analysis of the error rate, in jurisdictions where votes were counted initially by an electronic voting system, an additional record shall be made on the audit report form of the number of  votes for which voter intent is discernible but ballots were not marked according to the instructions provided to the voters under Section 48 of Chapter 54;

    8. when they have been counted all the ballots shall be enclosed in envelopes or containers, marked with the name of the city or town, ward, if any, and precinct and each envelope or container shall be sealed with a secure tamper-evident seal provided therefor, and the number thereof shall be recorded in the Precinct Election and Audit log.

    9. the registrars or three members of the audit team shall sign and date an affixed seal designed so that it cannot be removed without damaging the signatures.

(x)  Audit personnel shall not work with the preliminary tally numbers at hand but shall record and sign their own hand-counts on forms provided therefor by the state secretary.  If any subsequent counts are taken to clarify perceived errors the final audit report shall contain the results of all counts, the order in which they were performed, the perceived error/s and an explanation thereof. 

(y) Precinct Election and Audit Log

A log book shall be maintained by hand at and for each precinct in which shall be recorded various items pertaining to elections and to their audits, if any.  Items to be recorded therein include:

1) the model and serial numbers of all voting and counting devices used, if any,  signed by two election officials

2) the numbers printed on the seals used to close containers on election night,

3) the numbers printed on seals used to close containers after any audit or recount, as set forth in paragraph (e) 3 of this section,

4) the printed names and signatures of those signing seals and tally slips on election night,

5) the printed names and signatures of those signing seals and tally slips after an audit or recount,

6) any problems or anomalies encountered during the election, or any audit or recount

7) any protests made by those conducting the audit or any citizen or candidate observer present pertaining to the conduct of the audit along with the names and contact information of such persons. 

The Log shall be kept in a secure location before during and after the election but shall be available for public inspection and copying under the supervision of the municipal clerk.

(z)  Candidates. Candidates and referenda sponsors in contests that may potentially be audited must be informed no less than three days before the election of their right to have their own observers present at the audit of their contest if they choose.

This act shall take effect for the earliest applicable election which occurs one year after its enactment.  Pilot audits may be conducted in the interim.