[Image via FVAP]

Reprinted from April 4, 2019 electionlineWeekly – thanks, Mindy!

One of the biggest changes I have experienced in my move to the Elections Research team at Fors Marsh Group is the extent to which I get a close-up look at key research before it’s released. As I’ve told several people, it’s like the #electiongeek equivalent of transitioning from foodie to sous chef – going from a connoisseur of fine data to someone intimately familiar with the process of preparing and sharing it with the community.

That feeling is especially strong with regard to a new research note our team developed, in partnership with the Federal Voting Assistance Program (FVAP) and the Council of State Governments’ Overseas Voting Initiative (CSG OVI), looking at how transactional-level data about the Uniformed and Overseas Citizens Absentee Voting Act (UOCAVA) voting process illuminates key trends in military and overseas voting.

That data, which was produced starting in 2016 as part of FVAP’s efforts to seek new and different ways to assess the UOCAVA voting experience, looks past the summary jurisdiction-level data reported in Section B of the Election Assistance Commission’s Election Administration and Voting Survey (EAVS) to focus on voter-level transactions associated with requesting and returning a ballot. Participating states and jurisdictions (14 in 2016) produce data in accordance with an EAVS Section B (ESB) standard which, when combined with other data about the voting process, allows for rich and fascinating observations about how well (or not) the UOCAVA process works.

Thus, for example, using the 2016 ESB data, the new research note finds that:

  • About two in three ballot requests (65%) were received at least 45 days before the election;
  • About one in 25 (4%) waited until the week before the election;
  • Voters who received their ballots earlier were more slightly more likely to return them with less chance of being rejected for inaccuracy or lateness; and, most interesting
  • Voters who received their ballots by mail were slightly more likely to return them than voters receiving ballots electronically.

You can see the relationship between days until the election and successful return of ballots in the following graphs – first, the likelihood of timely return:

And then, likelihood of rejection:

Obviously, there are many other factors involved in the likelihood of timely return and acceptance of UOCAVA ballots – including who the affected voters are and why they choose when and how to request and return ballots – but these data provide an insight into the military and overseas voting process that deserves attention from election officials and advocates alike. It’s consistent with an overall trend in the field to seek greater standardization of election data transfer; an exciting new approach that I know FVAP (as well as those of us at Fors Marsh and CSG OVI) are interested in continuing to explore.

We were honored to be a part of this process – which could not have happened without the support of FVAP’s David Beirne and focus and dedication of CSG’s Jared Marcotte and Fors Marsh’s Colin MacFarlane – and we look forward to learning more as the ESB data pool expands in 2018 and beyond!

Bottom line: this election data is delicious. We’ll make more. Stay tuned …