Skip Navigation
Report

Pilot Implementation Study of Risk-Limiting Audit Methods in the State of Rhode Island

Summary: The risk-limiting audit is one of the most important tools states can use to bolster voter confidence and secure elections for 2020 and beyond.

  • Rhode Island RLA Working Group
Published: September 3, 2019

Exec­ut­ive Summary

In Octo­ber 2017, Rhode Island Governor Gina Raimondo signed into law a ground­break­ing elec­tion secur­ity meas­ure. Now, state law requires Rhode Island elec­tion offi­cials to conduct risk-limit­ing audits, the “gold stand­ard” of post-elec­tion audits, begin­ning with the 2020 primary. A risk-limit­ing audit (“RLA”) is an innov­at­ive, effi­cient tool to test the accur­acy of elec­tion outcomes. Instead of audit­ing a prede­ter­mined number of ballots, offi­cials conduct­ing an RLA audit enough ballots to find strong stat­ist­ical evid­ence that outcomes are correct. The law, enacted in the after­math of two crit­ical events relat­ing to the 2016 elec­tions, stems from decades of advocacy aimed at increas­ing the effi­ciency, trans­par­ency, and veri­fi­ab­il­ity of polit­ical contests in the state. Rhode Island is now the second state, join­ing trail­blaz­ing Color­ado, to mandate use of this modern tool statewide.

Follow­ing the law’s enact­ment, a group of profes­sion­als with expert­ise in elec­tion secur­ity and elec­tion admin­is­tra­tion formed the Rhode Island Risk-Limit­ing Audit (“RIRLA”) Work­ing Group. As its name suggests, the RIRLA Work­ing Group was estab­lished to assess the condi­tions in Rhode Island to help the state as it prepares to imple­ment the law. The RIRLA Work­ing Group recom­men­ded – and Rhode Island offi­cials agreed – that the state should conduct pilot RLAs in advance of the 2020 dead­line. The Rhode Island Board of Elec­tions chose Janu­ary 2019 as the date for the pilots and, based on several factors, selec­ted Bris­tol, Cran­ston, and Ports­mouth, Rhode Island as parti­cip­at­ing muni­cip­al­it­ies.

Lead­ing up to the pilots, the RIRLA Work­ing Group had regu­lar confer­ence calls, meet­ings, and other corres­pond­ence to gain greater famili­ar­ity with Rhode Island’s elec­tion laws, prac­tices, and voting equip­ment. In part­ner­ship with the state, the RIRLA Work­ing Group set a goal to plan and develop a trio of pilot audits that would both meet the state’s needs and adhere to the Prin­ciples and Best Prac­tices for Post-Elec­tion Tabu­la­tion Audits. Ulti­mately, the RIRLA Work­ing Group draf­ted three separ­ate audit proto­cols, step-by-step instruc­tions to guide those who would conduct the RLAs over the course of two days.

On Janu­ary 16 and 17, the Rhode Island Board of Elec­tions and members of the RIRLA Work­ing Group met in Provid­ence, RI to conduct the pilot risk-limit­ing audits. In conduct­ing three unique RLAs – a ballot-level compar­ison, a ballot polling, and a batch compar­ison audits – the part­ner­ship sought to:

  • Famil­i­ar­ize elec­tion offi­cials with RLAs and provide them some comfort with conduct­ing them through a hands-on learn­ing exper­i­ence;
  • Eval­u­ate Rhode Island’s elec­tion facil­it­ies, equip­ment, and other resources to determ­ine their adequacy for admin­is­ter­ing RLAs;
  • Emulate the actual envir­on­ment and real-world circum­stances in which the RLAs would be conduc­ted, includ­ing by having offi­cials manage unanti­cip­ated scen­arios that could arise;
  • Time the vari­ous audit steps and compile the data to compare the relat­ive effi­ciency of the separ­ate audit approaches;
  • Ascer­tain any gaps or defi­cien­cies that might hinder the initial rollout and imple­ment­a­tion of RLAs or the state’s abil­ity to conduct them beyond 2020;
  • Make a set of recom­mend­a­tions to help offi­cials conduct RLAs pursu­ant to state law and best audit prac­tices.

This report tells that story. It details how, through a collab­or­at­ive effort, Rhode Island success­fully conduc­ted three pilot RLAs. It provides a general over­view of RLAs, includ­ing the advant­ages and disad­vant­ages of each method. The report describes the history of elec­tion admin­is­tra­tion in Rhode Island, which has led to the envir­on­ment in which elec­tions are conduc­ted in the state today. It also lays out the essen­tial compon­ents of the audits– their design, soft­ware tools, and present­a­tion – and provides the results of the audits. Finally, the report describes some of the key lessons learned through­out the process, and it makes recom­mend­a­tions, specific­ally that Rhode Island pursue ballot-level compar­ison audits, so that state offi­cials and the public move towards RLAs seam­lessly and to improve the state’s exper­i­ence with RLAs in the future.

This was a truly collab­or­at­ive effort. It would not have been possible without count­less hours of work from Miguel Nunez and Steve Taylor (Rhode Island Board of Elec­tions); John Marion (Common Cause Rhode Island); Mark Linde­man and John McCarthy (Veri­fied Voting); Wilfred Codring­ton III and Andrea Cordova (Bren­nan Center for Justice); Luther Weeks (Connecti­cut Voters Count); Ron Rivest, Mayuri Srid­har, and Zara Peru­mal (Massachu­setts Insti­tute of Tech­no­logy); Suzanne Mello-Stark (Rhode Island College), inde­pend­ent volun­teers Lynn Garland, Neal McBur­nett, Tom Murphy, and many others who made the pilot audits and this report a success.

A host of recent events, includ­ing equip­ment malfunc­tion, cyber­threats, malad­min­is­tra­tion, and human error, have under­mined public confid­ence in Amer­ican elec­tions. Unfor­tu­nately, these types of occur­rences are not likely to disap­pear. But the good news is that our offi­cials can help, even in the face of constraints on their time and public resources. They can take steps both to decrease the number of incid­ents and to show that they are will­ing and able to address the prob­lems as they arise. They can start by conduct­ing RLAs, to assure the public that the repor­ted winners of elec­tions are the actual winners. Risk-limit­ing audits are an effi­cient, effect­ive, and straight­for­ward way to enhance public confid­ence in our elec­tions that takes into account the real­it­ies of elec­tion admin­is­tra­tion. The success of the pilots in Rhode Island can and should serve as a model for what state and local offi­cials across the coun­try can accom­plish, and how other indi­vidu­als and organ­iz­a­tions can provide valu­able assist­ance. We hope to make that clear with this report.