Better Elections Performance Through Measurement: Pew's New Index
[Screenshot image courtesy of pewstates.org]
“If you can not measure it, you can not improve it.” – Lord Kelvin
The future of election administration got a huge boost yesterday with the release of the Pew Center on the States’ new Elections Performance Index (EPI).
From the release:
“Election officials can use this data to benchmark their own performance over time, and help assess which policies have been working most effectively for their own citizens,” said David Becker, director of Pew’s Election Initiatives project. “Pew’s goal in developing this new online interactive tool is to promote the highest standards of accuracy, cost-effectiveness, convenience, and security in America’s election administration system.”
The study builds a new baseline for measuring election administration by looking at such indicators as polling location wait times, availability of voting information tools online, the number of rejected voter registrations, the percentage of voters with registration or absentee ballot problems, how many military and overseas ballots were rejected, voter turnout, and the accuracy of voting technology.
The part of the EPI that will get the most short-term attention is the “score” each state is given on the 17 indicators Pew collected from a variety of sources for election years 2008 and 2010. [The full methodology is here.]
But to me, the primary benefit of the EPI isn’t to pick winners and losers but rather to identify areas in which each of the states – including the high-performing ones – can improve their performance. Some of this represents policy choices by Pew: for example, states get credit for online voter registration, which was available in only a handful of states by 2010. I would expect a number of states to improve their “scores” in the 2012 EPI simply by adopting online registration.
The EPI also allows election officials to begin a conversation about the factors that affect performance. For example, in 2010 California rejected absentee ballots at a higher rate than any other state; in response, Los Angeles’ Dean Logan observed that the high rejection rate included ballots that were returned as undeliverable as well as signature match problems associated with society’s transition away from the reliance on consistent “wet ink” signatures. Both of these problems – which are a by-product of registration roll quality and the impact of technology
on elections – can and should be the starting point for discussions in California and elsewhere about how to reduce the rate of misdirected or rejected ballots. The data also provides a nice jumping-off point for discussions about how these indicators vary within states – a crucial idea in our decentralized system of election administration.
Whether looking back or looking ahead, however, it’s undeniable that having a common pool of quality data on which to base elections policy discussions is an invaluable asset to the field. Kudos to the full Pew elections team – especially Sean Greene, Zachary Markovits, Aleena Oberthur, Andreas Westgaard, and Maria Ho – for some quality work.
I’m going to give the last word to MIT’s Charles Stewart, who was vital to the success of the EPI. His quote to the New York Times encapsulates not just the need for the Index but also the reason why it’s so important to the future of elections:
Among all important areas of public policy, election administration is probably the most episodic and prone to the problem of short attention spans … What would the world be like if we only gave intense attention to education, corrections, transportation and public health problems for a one-week period every four years?