A close look at data from this year’s Republican primary race shows that Mitt Romney has in all cases done better among wealthier voters than he has with voters overall — evidence for the recurring knock against him that he has been unable to connect with the less well off.
Throughout the primaries, many commentators have remarked that in Mitt Romney’s bumpy ride to the nomination, he has been dependent on wealthier voters to get him there, while failing to appeal to middle and working class voters.
So TPM took a close look at all the primaries and caucuses so far, for which there has been exit poll data available, including statistically sizable cross-tabs of voters making $100,000 or more.
What we found is that if $100k voters were taken out of the math, Romney would likely still have won many of the states he has, but by slightly less than he did in real life. He would also still have lost a handful of states, but by less of a margin. But in two key instances, when Romney was on the verge of disaster — Michigan and Ohio — it was those voters who made the difference in his victories.
We found a few distinct categories: A decent number of places where he did so well that he would have won even without $100k-plus voters. Newt Gingrich’s big wins in South Carolina and Georgia — where Romney lost so badly, that he even lost the $100k crowd. And the key pair of states, Michigan and Ohio, where Romney lost voters making under $100k, but his advantage among the $100k’s was so strong as to pull out the win.
But not in one single case, for which there is data available, did Mitt Romney do less well among voters making $100,000, than he did among Republican primary and caucus voters overall. And logically, in every case he has performed less well among voters making less than $100,000.
The scatterplot above illustrates the point, with each state given a dot on the graph. The x-axis is Romney’s finish in the top-line results for the contest in question, while the y-axis represents the Romney’s exit-poll percentage among voters making $100,000 or more. And the dot placements mark Romney’s performance in each state.
The diagonal line represents a hypothetical 1-1 ratio. As you can see, every single dot is above the line — showing Romney’s disproportionate performance among wealthier voters. If any of the dots were below the line, that would mean there was a state where Romney did less well among the $100k voters than he did overall — but no such state currently exists, in the contests for which exit-poll data exists.
Iowa
Romney overall: 25%, second place.
Romney $100k: 36%, first place.New Hampshire* (Even without 100k’s, Romney would still have won.)
Romney overall: 39%, first place
Romney $100k: 48%, first placeSouth Carolina
Romney overall: 28%, second place
Romney $100k: 34%, second placeFlorida*
Romney overall: 46%, first place
Romney $100k: 52%, first placeNevada*
Romney overall: 50%, first place
Romney $100k: 58%, first place(Gap — Missouri’s non-binding primary, and caucuses in Minnesota, Colorado and Maine, which do not have exit poll data.)
Michigan** (Romney would have lost, without 100k’s.)
Romney overall: 41%, first place
Romney $100k: 48%, first placeArizona*
Romney overall: 47%, first place
Romney $100k: 52%, first place(Gap — no data for the caucuses in Washington state.)
Super Tuesday
Ohio**
Romney overall: 38%, first place
Romney $100k: 46%, first placeGeorgia
Romney overall: 26%, second place
Romney $100k: 32%, second placeMassachusetts
Romney overall: 72%, first place
Romney $100k: 76%, first placeOklahoma
Romney overall: 28%, second place
Romney $100k: 39%, first placeTennessee
Romney overall: 28%, second place
Romney $100k: 35%, first placeVermont*
Romney overall: 40%, first place
Romney $100k: 44%, first placeVirginia*
Romney overall: 60%, first place
Romney $100k: 64%, first place(Gap — Also on Super Tuesday, there are no entrance polls for the caucuses in Alaska, Idaho and North Dakota.)