Skip Navigation

Design Deficiencies and Lost Votes

In 2010, tens of thou­sands of votes in New York did not count due to over­votes — the invalid selec­tion of more than one candid­ate. This report demon­strates how the lack of adequate over­vote protec­tions dispro­por­tion­ately affected the state’s poorest communit­ies, suggests common­sense reforms, and exam­ines national implic­a­tions.

Exec­ut­ive Summary

In 2010, New York­ers voted on elec­tronic, optical-scan voting machines for the first time. Citizens went to their polling places on Elec­tion Day, filled out paper ballots and fed them into the brand-new optical scan machines.

But tens of thou­sands of their votes did not count. Specific­ally, about 20,000 voters in New York State did not have their votes for governor coun­ted because the machines read their choices as “over­votes” — the invalid selec­tion of more than one candid­ate. Even more votes were lost in other contests — 30,000 to 40,000 more. In a pres­id­en­tial year, with nearly twice the turnout, we expect that the number of votes lost because of over­vot­ing would more than double, possibly result­ing in more than 100,000 lost votes.

In modern history, New York has never seen so many lost votes due to over­vot­ing. Unlike the new optical scan voting system, New York’s old lever machines did not allow over­vot­ing. But even so, the numbers of lost votes due to over­vot­ing in 2010 were far greater than they should have been. Over­votes are almost always unin­ten­tional. A well-func­tion­ing voting system, even one that includes optical scan equip­ment, should have over­vote rates very close to zero.

A great irony of this new prob­lem is that the federal mandate to purchase new machines was specific­ally meant to reduce over­votes nation­wide. Tens of thou­sands of votes were voided as over­votes in 2000, in places like Flor­ida on punch card and other voting systems. The Help Amer­ica Vote Act (“HAVA”), passed by Congress in 2002, requires that new voting systems used in polling places in United States must:

(i) notify the voter [when she] has selec­ted more than one candid­ate for a single office on the ballot;

(ii) notify the voter before the ballot is cast and coun­ted of the effect of cast­ing multiple votes for the office; and

(iii) provide the voter with the oppor­tun­ity to correct the ballot before the ballot is cast and coun­ted.

The Bren­nan Center, NAACP New York State Confer­ence and other civil rights and good govern­ment groups have argued that New York’s over­vote protec­tions did not satisfy these require­ments and predicted in a lawsuit filed in 2010 that these inad­equate protec­tions would lead to such high over­vote rates. Specific­ally, they poin­ted to a message voters would see if the machine could not discern the voter’s intent; the groups argued this message would confuse voters, making it more likely they would cast invalid votes, and less likely that they would correct their ballots to ensure they were accur­ately coun­ted.

As we demon­strate in this paper, the lack of adequate over­vote protec­tions had a dispro­por­tion­ately negat­ive impact on the state’s poorest communit­ies. Lost votes due to over­vot­ing occurred far more frequently in areas with higher popu­la­tions of low-income resid­ents, people of color, and immig­rants.

Black and Hispanic voters were at least twice as likely to lose votes due to over­vot­ing as non-Hispanic whites. Shock­ingly, in two Bronx elec­tion districts, nearly 40 percent of the votes cast for governor were voided as over­votes.

The good news is that the New York State Board of Elec­tions has agreed to adopt a better over­vote warn­ing when a voting machine cannot discern voter intent, hope­fully in time for the Novem­ber 2012 elec­tion: such a warn­ing will inform the voter of the prob­lem in plain English (“you have filled in too many ovals”), and clearly explain the consequences of cast­ing an over­vote (“your vote will not count”).

This should signi­fic­antly reduce the number of over­votes in 2012, but it will not elim­in­ate the prob­lem. There is more that our public offi­cials, and espe­cially our state legis­lat­ors, could do. In this report, we discuss how common­sense solu­tions, like requir­ing boards of elec­tions to publish precinct-level elec­tion results, can improve detec­tion and correc­tion of machine-related prob­lems. Crit­ic­ally, we also explore how better ballot design require­ments can reduce over­votes.

Finally, we exam­ine the national implic­a­tions of our find­ings in New York.