One. The exit poll shows some familiar patterns, with Mitt Romney running strongest among high-income and high-education voters. But there are some interesting departures from pattern.
Romney carries all age groups (well, not quite; he trails a bit among the 40-49 cohort but carries the overlapping 30-44 and 45-64 cohorts). In earlier contests young voters broke for Ron Paul and the middle age cohorts went for Rick Santorum. Not in Illinois.
Romney carries college graduates and non-college graduates. Romney carries all income groups except the under $30,000, which are only 10% of the sample.
Romney, as usual, carries Catholics, with 53% of the vote. Partly this reflects the fact that Catholic Republican primary voters tend to live in the upscale suburbs, where Romney’s appeal has consistently been strongest. But I think there may be another factor. Rick Santorum lives a lifestyle that was typical of many Irish Catholics in the 1950s. And sometimes he seems to be trying to convert us to that lifestyle, as when he volunteered in that caffeinatedthoughts.com interview last October that contraception “is not okay.” But most American Catholics don’t life that lifestyle today. You have to be pretty old to remember, as I do, when there were multiple Catholic families on the block with seven or eight children.
Romney loses white evangelicals, but by only 45%-39%. He carries tea party supporters, including strong supporters. There’s obviously considerable overlap between evangelicals and tea partiers, but in Illinois at least those in these overlapping groups who are more concerned about economic issues are going with Romney, while only those who are (unlike most voters this year) more concerned about cultural issues are sticking with Santorum.
Two. Newt Gingrich finished fourth, with 8% of the vote, 1% behind Ron Paul. Gingrich can continue going through the motions of campaigning, visiting zoos and lunching in New Orleans at Galatoires. But he isn’t getting much notice in the media and for most voters he has become a non-factor. And his effective disappearance hasn’t helped Rick Santorum. The notion that there is a bloc of conservative voters equally attracted to Gingrich and Santorum and equally repelled by Romney is refuted by the Gallup national poll question about the second choice of Gingrich voters: 40% Romney, 39% Santorum. The Illinois result suggests that voters who at some point backed Gingrich broke about the same way. In PPP Illinois polls last year Gingrich had 17% and 23% of the vote—two to three times what he polled in the primary. There is no sign that these voters went entirely to Santorum.
Three. There is an uncanny resemblance between the Illinois results this year and in 2008—except that Mitt Romney was the loser that year and the winner this year. Look at the percentages for the winner (McCain in 2008, Romney in 2012) and the second-place finisher (Romney in 2008, Santorum in 2012) statewide and in the three major regions—Cook County (Chicago and inner suburbs), the Collar counties (DuPage, Kane, Kendall, Lake, McHenry, Will) and Downstate (the other 95 counties).
2012 2008
ILLINOIS 47%-35% 47%-29%
Cook Co 57%-26% 52%-28%
Collar Cos 52%-30% 51%-29%
Downstate 38%-42% 43%-29%
The one significant difference is that Santorum was stronger Downstate than Romney was in 2008, when Mike Huckabee was winning 29% of the vote there (he clocked in at only 11% in Cook County and 13% in the Collar Counties). I’ll leave further explanation to others.
Four. Many commentators last night were saying the turnout was low. Actually, it wasn’t, at least by historic standards. Here are the Illinois Republican presidential primary numbers going back to 1976, with snips of commentary:
2012 921,880 up 10% from 2008
in Cook County
2008 899,422
2004 no contest
2000 736,921 nom already set
1996 818,364
1992 831,140 Buchanan challenging
Bush 41
1988 858,637
1984 no contest
1980 1,130,081 IL Rep. John Anderson
vs. IL native Reagan
1976 775,893
Keep in mind that Illinois has had relatively little population growth, so you would not expect turnout to rise much over time. In that context, it’s fascinating that the high turnout mark was reached 32 years ago. But it also has to be said that this year’s turnout was the second highest, and that some of those other contests (1996, 1988, 1976) were fiercely contested.
Turnout has been low relative to previous contests in many Republican primaries this year, but it has been relatively high in others; there doesn’t seem to be any clear pattern, or at least one that I can discern.
