Four surprises from the final statistics on Iowa's 2012 election

The Iowa Secretary of State’s Office posted the statewide statistical report on the 2012 general election this week (pdf). For those wondering what took so long: Iowa county auditors had 60 days after the general election to submit their final reports.  

A few things surprised me when I looked over the numbers for the first time and compared them to the 2008 statewide statistical report (pdf).

UPDATE: I asked the Secretary of State’s Office for a comment on the discrepancy between the certified election results, which showed that 1,589,899 Iowans cast ballots in the general election, and the statewide statistical report’s “total voted” number of 1,572,198. The explanation is at the end of this post.

1. Registered Iowa Republicans turned out at a higher rate than Democrats in 2012.

According to the statewide statistical report, 82.2 percent of registered Republicans cast ballots in the general election (546,729 voted out of 664,945 registered). By comparison, 76.9 percent of registered Democrats cast ballots (525,624 voted out of 683,200 registered).

Barack Obama must have beaten Mitt Romney badly among independents to have won Iowa by roughly 92,000 votes. Or rather, Obama’s re-election campaign and the Iowa Democratic Party did a much better job identifying and turning out sympathetic no-party voters compared to the Romney campaign/Iowa GOP.

By the way, the turnout rate among no-party voters in Iowa was just 62.7 percent (497,496 voted out of 792,957 registered). That’s still higher than the 61.3 percent turnout rate for Iowa independents in 2008 (467,762 voted out of 763,520 registered) and much higher than the usual turnout rate for independents in midterm elections.

2. Senior citizens showed the biggest jump in turnout percentage from 2008 to 2012.

Iowa was one of the few states to show higher voter turnout in 2012 (73.3 percent) than in 2008 (72.6 percent). Because of our aging population, one would expect more senior citizens to cast ballots in the latest presidential election than in 2008. But I was surprised to see that the turnout rate among the 65 and over voters increased from 81.9 percent in 2008 to 84.9 percent in the most recent presidential election.

3. Women were more likely than men to vote early.

As usual, women cast more ballots in the general election (840,877 women voted compared to 731,321 men). As usual, the turnout rate for women (74.7 percent) was higher than turnout among men (71.7 percent).

I didn’t realize that women were significantly more receptive to voting early, either by mail or in person. Among all Iowa voters, 43.2 percent cast early ballots. For women, that figure was 45.8 percent, compared to just 40.2 percent for men. Since women are more likely to be primary caregivers for children or older relatives, maybe women were more concerned about a family emergency preventing them from getting to the polls on election day.

4. Adults between the ages of 25 and 50 were least likely to vote early.

According to the statewide statistical report, senior citizens were the most likely to vote early: 58.3 percent of Iowans over 65 who voted cast early ballots. That makes sense, because seniors might have more concerns about getting out in bad weather on election day, or may have illnesses that prevent them from leaving home, or may spend the colder months out of state following retirement.

I had assumed that among Iowans under age 65, more seasoned voters would be more attached to the experience of going to the polls on election day. But in fact, 44.5 percent of voters between the ages of 50 and 64 cast early ballots. That was a higher early voting rate than the 42.1 percent for voters aged 18 to 24.

In contrast, just 33.1 percent of voters aged 25 to 34 cast early ballots, and 32.5 percent of voters aged 35 to 49 did so.

Any comments about the 2012 general election are welcome in this thread.

UPDATE: Chad Olsen, communications director for Iowa Secretary of State Matt Schultz, explained why the statewide statistical report shows that 1,572,198 Iowans voted in the general election. He said the certified number of 1,589,899 Iowans casting ballots is correct, but due to a quirk in the iVoters system, the statistical report can only be pulled two months after the general election,

and lists “deteriorate” gradually each month. What I mean by that is of the 1,589,899 voters in the 2012 election, by the time we pulled the data 60 days later, about 17,000 Iowans who voted in the election have been removed from the list-most likely as a result moving out of state and cancellation notices being received by auditors, or cancellation because of death, etc.

That makes sense–I know five Iowans who have died since voting in the November election. I had not realized that the “total voted” number in the statistical report might differ from the number of ballots cast.

About the Author(s)


  • Your first "surprise" really isn't one, it's always that way

    Look at past elections, even in 2008 the GOP had a higher turnout rate than Dems:  80.07% to 78.2%.

    Maybe you’re thinking the larger margin this time was the surprise, but was it really?  Everyone knew Dems were highly motivated in the ’08 wave and less so this time.

  • #2

    I will speculate that this is just a manifestation of boomers hitting the 65+ category. The issue is with the range of the category. Turnout rates drop off sharply when you get out to the late 70s/80s. An influx of new voters at the “young” end of the category would produce a bump in turnout rate relative to earlier years.

    #4 Is this an issue of affinity for early voting by age category or more an effect based on campaign targeting strategies? Targeting older (say 60+) and under-24 would not be particularly surprising.

  • Are the registration figures

    ALL registered voters, or just Active status? (Fron the total it looks like all.) Democrats tend to be more mobile and more likely to be on Inactive status because mail had been returned undeliverable. (Independents, even more so.) Using all voters artificially inflates D registration totals and thus artificially decreases turnout percentage relative to Republicans.

    • the total includes all registered voters

      as far as I can tell.

      • I think so too

        And that means all the registration numbers are artificially high and the turnout percentages artificially low. “Inactive” status is not what a staffer would call “weak voting.” It means the auditor has gotted returned mail indicating the voter has moved away. It’s basically preliminary cancellation. The process requires multiple repeat efforts to contact the voter by mail. They stay on this status through two general elections, and only then are they completely cancelled.

        Active-only numbers are much more accurate and more comparable to the pre-1994 numbers. Prior to motor voter, it was simple: four years without voting or some registration change, you were out.