I noted a disparity in this week’s Marquette Law School Poll in the U.S. Senate race, it made me want to look into things further. To review, among “likely/certain voters”, Democrat Russ Feingold led the GOP’s Ron Johnson 51-42, but only led 45.5-40.8 among all registered voters. So that led me to dig into the crosstabs from the Marquette Poll to see what might be causing the disparity.
Once I did so, the answer in the difference became obvious- young voters. Take a look at the difference in these figures among these two subsets of people ages 18-29 in the Feingold-Johnson polls.
Feingold vs. Johnson, age 18-29
Registered Voters Feingold 41.4-40.4
Likely Voters Feingold 51.5-38.6
That right there explains half of the difference between 4.7 and 9 that we see between registered and likely voters in that senate race. Interestingly, you don’t see that difference in the Clinton vs. Trump matchup - Clinton leads in registered voters 18-29 by 1, but Trump leads by 0.3 among likely voters 18-29.
I have a lot of skepticism on whether the tightness in Registered Voter totals for young voters will hold up, because it doesn’t jibe with recent exit poll data among young voters in Wisconsin. 18-29 year-olds in Wisconsin backed Barack Obama 60-37 and went 58-39 for Tammy Baldwin in 2012, and were the only age group supporting Mary Burke vs Scott Walker in 2014. Also note this accounting of a recent Bloomberg news poll indicates Hillary Clinton may be likely to repeat President Obama’s big margins among younger voters. although it seems to have more to do with Drumpf’s idiocy vs approval of Hillary.
Clinton, in this week's Bloomberg Politics national survey, enjoys a lead of 3 to 1 among likely voters aged 18 to 29, and more than a 3-to-2 advantage among 30- to 39-year-olds. These millennials still aren't wild about the presumptive Democratic nominee. Her favorable ratings are far below Barack Obama's. But Trump's negatives are stunning: Among 18- to 29-year-olds, 83 percent have an unfavorable view of him.But yet Wisconsin millennials in 2016 are so different that Clinton vs. Trump and Feingold vs. Johnson is a 1-point race for the 18-29 group? Forgive me if I think that doesn’t reflects reality, and I’m not necessarily ripping Charles Franklin and company for it. Younger people are notoriously hard to reach compared with older voters who have landlines and answer phone calls. You can only report based on the data you get.
In 2012, voters under 40, who made up more than a third of the electorate, provided the winning margin for Obama. Trump in this poll, conducted by Ann Selzer, does considerably worse than the 2012 Republican candidate, Mitt Romney, with these voters.
But given that a whopping 82% of those same 18-29 year-olds voted for Bernie Sanders in April’s Dem primary, and that Sanders told his supporters yesterday that a top priority for November is that Trump “is defeated and defeated badly,” it seems a safe bet that Hillary will have major support from the younger Wisconsinites come November.
The same goes with Feingold, who had Bernie-like positions on campaign finance, the Iraq War and civil liberties before Bernie was ever a U.S. Senator, and is a candidate that Sanders is already raising money for through emails. Other demographic data from the Marquette Law Poll shows that Russ would be wise to link himself with Bernie, as right now Feingold is being seen as “generic Democrat” in key groups instead of a “Bernie Sanders independent.” I draw that conclusion by comparing Feingold’s crosstabs with the Clinton-Trump and Sanders-Trump matchups in the poll, and I’ll use the registered voter numbers, as that shows a closer race for both Clinton-Trump and Feingold-Johnson, to show where the opportunity exists for Dems in November.
Pay extra attention to the numbers with self-described Independents and voters that ended their education after high school.
2016 election matchups Wisconsin, RVs
By Party ID
Clinton-Trump- Trump 78.5-2.6
Feingold-Johnson- Johnson 87.0-6.3
Sanders-Trump- Trump 74.1-14.7
Clinton-Trump- Clinton 91.3-1.6
Feingold-Johnson- Feingold 83.9-5.0
Sanders-Trump- Sanders 95.1-2.1
Clinton-Trump- Clinton 32.5-31.9
Feingold-Johnson- Feingold 43.7-39.1
Sanders-Trump- Sanders 55.4-26.2
No HS Diploma
Clinton-Trump- Trump 31.6-30.2
Feingold-Johnson- Feingold 48.7-32.1
Sanders-Trump- Sanders 54.1-23.7
HS Diploma only
Clinton-Trump- Trump 41.6-35.4
Feingold-Johnson- Johnson 44.9-36.7
Sanders-Trump- Sanders 54.5-37.4
Clinton-Trump- Clinton 43.7-34.2
Feingold-Johnson- Feingold 48.4-38.2
Sanders-Trump- Sanders 61.6-26.1
Clinton-Trump- Clinton 38.9-35.7
Feingold-Johnson- Feingold 43.8-41.8
Sanders-Trump- Sanders 55.3-28.5
Interestingly, voters with 4-year college degrees don’t change all that much across all 3 polls, as the Democrats have solid double-digit leads in all 3 races, so that wouldn’t seem to be as much of a demo of emphasis for the Feingold campaign.
But the demo data above sure makes it seem like a few Bernie-Russ rallies in some blue-collar towns with plenty of young voters could be the way to go (sounds like cities such as GB, Kenosha, Superior and Oshkosh to me!). And these figures show that if Russ can get the relatively low-info, low-interest voter to get past their default position of favoring incumbent Senator Johnson, that he has room to expand on his already-sizable lead. Not a bad position to be in before casual voters start to pay extra attention for the Fall elections.