Anecdotes are illustrative, evocative and memorable—and a staple of election policy debates. Just think back to February’s State of the Union Address, when President Obama introduced Desiline Victor, the Floridian who waited six hours to vote. The President was illustrating why he created a bipartisan election commission.
The election administration of the several states visited was surveyed in a systematic manner with a view to finding which features were working satisfactorily, which unsatisfactorily, and what was the general experience. Emphasis was placed always upon the practical operation of election laws rather than merely the provisions of the statutes. Each survey involved not only a study of the statutes but also detailed, and usually lengthy, interviews with chief election officers, examination of records and equipment, and interviews with politically informed persons outside of the election office …This study was undertaken because of the present backward and generally unsatisfactory administration of elections.
But anecdotes make a weak foundation for public policy. Instead, “evidence-based management” is underpinning all kinds of government services these days, whether the topic is health care, transportation, criminal justice, education or election administration.
For election administration, finding “evidence” is tricky. Every state, and frequently every jurisdiction, conducts elections differently, making comparisons difficult. Data is not gathered uniformly nationwide as it is in many other government arenas. Election costs are hard to track because they’re borne by several levels of government. You get the idea—it is hard to get facts and figures to support election evaluation.
Not that policy wonks haven’t tried. The first nationwide effort came in the late 1920s when the diligent researcher, Joseph P. Harris, surveyed all the states on election administration. In the second of two tomes based on that research, Election Administration in the United States (Brookings Institution,1934), he describes his work this way:
With a linguistic update, those words could describe this year’s effort to measure elections: The Elections Performance Index, from The Pew Charitable Trusts. “Our work was in part a response to election officials at the state and local level getting frustrated with the widespread lack of empirical data to understand the election process, and to get away from the anecdotes that drive policy and media attention,” says Zach Markovits, manager of Pew’s election initiatives team.
In fact, 2013 may well be the year that elections finally meet their evaluative match. Besides the Elections Performance Index, a how-to book, Evaluating Elections: A Handbook of Methods and Standards, by R. Michael Alvarez, Lonna Rae Atkeson and Thad E. Hall, has hit the market. “Our work is different but complementary,” says Hall. “Pew’s work can identify places to look for improvement. Ours provides a process for how to dig into the data and how to fix a problem.”
NCSL has taken the measure of the Elections Performance Index, evaluated Evaluating Elections, and then done a quick survey of the many ways in which states are already helping themselves make better policy choices when it comes to election administration.
1. The Elections Performance Index
What is this Elections Performance Index? It is not a report in the typical sense of that word. Instead, it’s a “baseline tool for legislators and professionals in the election field to use,” says Markovits. Embedded in this tool is a trove of state-by-state data from the 2008 and 2010 general elections. The data represent 17 indicators, each of which addresses one facet of election administration. Data for 2012 will be added as soon as it becomes available.
The EPI isn’t about words. It is about data offered in a way that allows lawmakers and other policy people to shape and re-shape data so they can focus on what’s helpful at any given moment. Interested in just one state? Fine—it’s easy to see all 17 indicators for Alabama or Wyoming or any state in between. Want to drill down on just one topic, such as ballots from overseas voters? Great—it’s easy to see the rates of return for 50 states plus the District of Columbia. Would seeing how data changes over time meet your needs? With the click of a mouse, you can toggle back and forth between 2008 and 2010. (This feature will become more robust as data from additional years are added in.)
As a brand new project, the EPI isn’t perfect. For instance, it doesn’t address costs or customer satisfaction—because there are no good data on these. And yet, what lawmaker can ignore costs or customers? To such questions, Markovits responds, “This is an iterative process. Because our goal is to improve this baseline tool over time, we're interested in hearing how it's being used and ways we might improve it.”
Data and the Indicators
For now, the problem with election-related data isn’t that it’s apples and oranges. It’s a veritable fruit basket, complete with papayas, mangos and kumquats. Therefore, Pew worked with an advisory group that included local and state election officials plus several luminaries in the field of election research, to review existing data sources. They looked for indicators that were available and consistent across time and across states, and that measure something fundamental to elections administration. The group whittled down the indicators from an initial set of 40 to the final 17.
In the end, the data used in the EPI come from the EAC’s Election Administration and Voting Surveys from 2008 and (cont. next column)
(cont. from previous column)
2010, surveys of state election division records, and some of Pew’s own research, such as Being Online Is Not Enough and Being Online Is Still Not Enough.
For more on election data—what’s available and how reliable it is—see Pew’s Election Administration by the Numbers, a 2012 report that describes the state of election data. The report can save time for policymakers and researchers as they look for quality information.
What do administrators think of the EPI? “We’ve seen raters come and go and we mostly have been able to ignore them in the past,” says Chris Thomas, Michigan’s election director. As for the EPI, he says “This gives us a tool to convince state legislators that they need to take a look at
changing some things. We need data as the foundation to achieve balance between competing political goals.”
While Wisconsin tops the charts and Mississippi brings up the rear, ranking states was not Pew’s goal. “Every state does some things well, but we now know that each can also improve their election administration in a number
of specific ways,” says Markovits. The EPI’s goal is to simply give “states
a new way to think about election policy.” All states can improve—and if all states do an excellent job, then all states can get great grades.
The EPI shows that Mississippi simply had no data for many of the indicators, rather than that it didn’t do well. Even so, Mississippi’s Representative Bill Denny, chair of the House Committee on Apportionment and Elections, isn’t happy. “Being one of 16 states that is under section 5
of the Voting Rights Act of 1965 (and one of eight that is 100 percent
under it) we’ve got a sterling record for elections,” says Denny. “This doesn’t make sense. I can assure you that we run our elections in an excellent manner. We don’t have any complaints at all. We should easily be in the top 10 of the United States.”
As for the nation’s election geeks, see these comments from Doug Chapin, Heather Gerken and Charles Stewart.
2. Evaluating Elections (the book)
Dick Smolka, the grandpère of elections work, says that “it is important to distinguish between election procedures prescribed by law and those introduced by election administrators.” In other words, it’s one thing to set out the law; it’s another to interpret it and follow it. That’s where the book, Evaluating Elections: A Handbook of Methods and Standards, comes in.
“We wrote this for a couple of audiences. We had election officials in mind,” says co-author Thad Hall. “We want them to understand that in some ways they’re already sitting on a wealth of information, and they can use this data to think in the future how they can make things work better.”
“We also wanted this to be thought of by legislators,” he continued. “The laws may be fine; lawmakers need to think about building up capacity at
the state level to promote better training, or to provide localities with resources so they can evaluate existing data. Imagine if a legislature said, ‘We want our universities to help our jurisdictions evaluate their data.’”
The book provides a step-by-step framework to evaluate elections, including details on surveying voters, surveying poll workers, taking observations at the polls, and doing a procedural audit. Hall asks: “Did the ballot boxes come back sealed? How many voters had problems that
weren’t resolved well? How many errors came from poll workers? A procedural audit will look at all of these questions. No state or jurisdiction wants a process problem to blow up on Election Day.”
For more details, contact the authors or buy the book. The publisher is offering a 20 percent discount to NCSL-affiliated folks.
3. State-Based Reports
By no means are the Elections Performance Index and Evaluating Elections the only resources on improving election administration. Indeed, many states do their own evaluative work, either through legislative audits, partnerships between universities and election administrators, or directly through the offices of state or local election officials. NCSL recently gathered many of these reports from the last two years. They are an excellent source of state-specific data and practical innovation. (If we missed one, please let us know so we can practice “continual improvement.”)
For instance, here are reports from Maine, South Carolina, Texas and Virginia, all called for by their legislatures.
This year, legislation is being considered to create general election studies
in Arizona (HB 2160), Massachusetts (HB 635) and New Hampshire (HB 521). Other states have legislation to create studies and commissions with specific mandates:
- California SB 637 would require the secretary of state to study early voting.
- Connecticut SB 433 would require a “democracy index” to be implemented.
- Kentucky HCR 31 would request that the Legislative Research Division conduct a study about the use of vote centers.
- Montana SJR 14 would study combining primary elections with school elections.
- Oregon SB 157 (and companion bill HB 3285) would establish a task force focused on Oregon’s compliance with the Help America Vote Act.
- Pennsylvania SR 28 would study early voting and same-day voter registration.
- New York AB 3188 and Texas HB 579 would call for studies of Internet voting.
From the Chair
Nevada’s Senator Pat Spearman (D) chairs the Senate Legislative Operations and Elections Committee, a wonderful assignment for a freshman legislator. She is a retired military officer and an ordained minister—and is completing her doctorate in business administration. On March 20, she spoke with NCSL about elections policy in the Silver State.
--Our secretary of state, Ross Miller, rolled out a plan that people say was a form of voter ID. That’s not what he’s talking about.
--If the DMV is already verifying information for voters who register online, and since they have pictures for their driver’s licenses, why not equip electronic pollbooks with those pictures? Then, when I show up to vote, they type my name in, and my picture comes up. That takes the onus off the voter to provide the identification and places it on the government, if you will.
--I always like to hear what local officials have to say. It’s always a good idea to get their ideas and understanding so we know how they will implement a new plan.
Read the full interview here for more on Senator Spearman's prior work, methods for approaching legislation, and proposed campaign finance legislation.
The Election Administrator's Perspective
Jacquelyn Callanen serves as the elections administrator for Bexar County, Texas, which includes the nation’s seventh largest city, San Antonio. On February 28, NCSL asked her about the election issues in her district.
--When I get to talk to an 80-year-old Hispanic woman who’s going to vote for the first time, it doesn’t get any better than that. Nothing warms your heart more than when you make that difference.
--Bexar County is proud that we are seventh in the country for mailing out overseas ballots. Some of our young military have never voted in person. We send a pdf package to them on just how to vote. The package includes directions, a thank-you for voting, a “security envelope” and a second envelope it goes into, and of course the ballot for them to print out and sign.
--We use 100 percent touch screen voting machines that are 11 years old. Our voters love it, our election officials love it, and we’re making it work. They’re still pumping along, but we’ve had three or four desktop computers here in the office in the meantime.
Read the full interview here for more on poll workers, equipment and a suggestion to legislators to create an Election Day school holiday.
One Big Number: 195
That’s the number of election-related bills enacted in 2012, as measured by NCSL’s elections team in Wrap-Up of 2012 Election Legislation Enactments. Getting more attention in 2012 than in 2011 were online voter registration, polling place requirements and voter registration list maintenance. Continuing in popularity were laws that addressed election crimes, the needs of military and overseas voters, and absentee voting requirements. Less common than in 2011 were enactments relating to all-mail elections, the National Popular Vote compact, primaries, pre-registration for 16- and 17-year-olds, Internet voting pilot programs and vote centers.
This weekly round-up of election-related news, opinion, legislative action and job openings always leads with a breezy look at one timely topic. Most recently it has tackled Law and Order: Elections and First Person Singular: Steve Weir. EW aims at an audience of election administrators, but the whole universe of election watchers reads it. (Editor’s note: every week there’s some little bit of humor to be found—and since “The Book of Election Jokes” is one of the thinnest in the world, we’ll take what we can get.)
While the U.S Supreme Court ponders the fate of Section 5 of the Voting Rights Act of 1965, Washington is considering one of its own: the Washington Voting Rights Act. If enacted, it would likely shift many local races from at-large elections to district-based elections.
The U.S. Supreme Court heard Arizona v. The Inter Tribal Council of Arizona on March 18. This case will decide whether states can add procedures above and beyond what the National Voter Registration Act of 1993 requires—specifically, whether a state can ask for proof of citizenship at the time a voter registers. To understand the case, read this explanation from Lyle Denniston, at SCOTUSblog. Other states with similar laws include Alabama, Georgia, Kansas, South Carolina and Tennessee.
Wisconsin now knows what it would cost to eliminate same-day registration in the Badger State: the Final Report on the Impacts and Costs of Eliminating Election Day Registration in Wisconsin says it could cost $14.5 million. With EDR, Wisconsin isn’t required to offer voter registration at the motor vehicle bureaus, as outlined in the National Voter Registration Act of 1993. Without it, Wisconsin would need to comply, accounting for much of the expected expenses.
Finding good data on how voter ID laws affect voters is hard to do. This recent report, Black and Latino Youth Disproportionately Affected by Voter Identification Laws in 2012 Election, is one—and NCSL would like to find others, as well. The report says that minority youth are less likely to have state-issued IDs than white youth.
The Accessible Voting Technology Research Workshop is coming right up, April 1 and 2. This event, sponsored by the U.S. Election Assistance Commission and the National Institute of Standards and Technology, will be webcast here.
We missed Voter Information in the Digital Age: Grading State Election Websites from the Center for Governmental Studies when it came out in 2012; perhaps you did too.
J. Alex Halderman is noted for his cautionary voice when it comes to Internet voting and, last fall, to online voter registration. He spoke with NCSL about both his belief in online registration and his views on how to make it safe. Read the Q and A.
From NCSL's Elections Team
Is there something you’d like to know about elections? Do you wish you could ask a reference librarian? Well, you probably can ask a professional in your legislative library. Or, ask us. We aren’t “real” librarians, but we are in the information business—we gather it, package it and distribute it. And, as always, please stay in touch.
Jennie Bowser and Wendy Underhill