1 vote

Paul leads polls among young voters; most polls fail to report.

In response to my post regarding Ron Paul’s success with non-white and young voters in a recent California primary poll, a commenter suggested the Paul campaign should summarize Congressman Paul’s success with younger voters in the national polls. I thought it was a good idea, so I did it.

Here are the results from all the national polls released in August 2011. These are the general results:

Date Pollster Total
9/1/2011 Politico 10%
8/31/2011 NBC 9%
8/31/2011 Fox News 7%
8/27/2011 Quinnipiac 9%
8/25/2011 CNN 6%
8/21/2011 Gallup 11%
8/21/2011 PPP 6%
8/15/2011 Rasmussen Reports 9%
8/9/2011 Fox News 6%
8/7/2011 CNN 12%
8/7/2011 USA Today 14%
8/4/2011 McClatchy/Marist 3%

Here are the results for the 18-29 age range (sorry for the messy table).

7-Aug...USA......Does not provide tables for Poll
4-Aug...MAR....."Under 45 = 33%" NA NA NA NA NA

You’ll notice that Paul is generally at top when it comes to voters in the 18-29 age category. But you’ll also notice that there are a lot of NAs; the information on who young voters support isn’t very complete. That information gap deserves attention, so I looked closer at the polls.

Neither of the Fox News polls provided age breakdowns of their sample. They also didn’t provide any kind of breakdowns for their main survey question (“which candidate would you like to see as the nominee?”). The August 31 poll did provide breakdowns for question #28 though. That question was: “Are there any announced or potential candidates for the Republican nomination who you feel are just too extreme to be seriously considered? If yes, who?” In the age range “Under Age 35” (the closest to the “young voter” category), here was the result: Bachmann 12%; Perry 9%; Palin 7%; Paul 4%; and Romney 1%. In other words, Paul does second only to Romney among young voters, in terms of who they find least extreme.

The earlier Fox News poll provided age breakdowns only for their “Candidate Name – Do you think this person would make a good president or not?” questions. In the “Under Age 35” category, here are the results for answers in the affirmative: Romney 32% (yes); Bachmann 19%; Perry 18%; Palin 21%; Paul 32%. In other words, among young voters Paul was tied with Romney at the top, way above the rest.

Now look at the two CNN polls. The CNN polls also don’t provide any information regarding the age breakdowns of their overall sample. But here’s what’s fascinating. On page 5 of the most recent CNN poll , they provide age breakdowns for the responses to their question 28 (their main question): “please tell me which candidate you would be most likely to support for the Republican nomination for President in the year 2012, or if you would support someone else.” Actually I should say they give the results for the category “50-64”. And they give the results for the category “65+”. And they give the results for the category “Under 50”. And they even give the results for the category “50 and Older”. But their results for the category “18-34”? NA! And it’s the same for the category “35-49”. So how does that make sense? Why are the results NA? It seems that either they didn’t have enough people in those categories to justify reporting the results, or else they excluded them for some other reason. Both are disturbing answers. If their reason was the former, it means that the CNN poll was basically a poll of people over 50. It’s worth asking if that age-bias in their sample would affect Paul’s results, since we know he tends to not do as well among older voters.

Interestingly, Paul’s total was better in the earlier CNN poll from August 8. In that poll he was third at 12% (tied with Palin and Giuliani), beaten by Romney (at 17%) and Perry (at 15%). Again in that poll CNN did not provide the age breakdowns of its sample, and also reported NA for the age categories of 18-34 and 35-49. But in that poll, the results for the Under 50 category have Paul in second place, and he’s only beaten by Giuliani (who isn’t even announced as a candidate)! Paul took 16% of the population under 50, below Giuliani at 17%, above Palin at 11%; above Romney at 14%; and above Perry at 14%. What would his results have been if CNN had had people under 35 or 29 in their sample? Or what were his results for that age category that CNN didn’t report?

USA Today and Rasmussen don’t provide the tables for their polls. The very early McClatchy/Marist poll reported Paul at 3%. They didn’t provide very precise age breakdowns for their polls, but they did report for their Republican primary voters that 33% were “Under 45”. That means that in their poll well over half their sample was over the age of 45. It seems like older people get polled, then those polls are used to decide who can win, then younger people get to choose between two candidates that were chosen by the older people.

Most people intuitively recognize why these polls matter. It’s not because they’re an accurate measure of people’s attitudes or an accurate predictor of their voting behavior. They matter because they actually influence that behavior by becoming a central reference point for what is sometimes called “consensus information”.

Consensus information is what people believe about what the majority of people around them believe or how they behave. People change their behavior in response to their perception of how the people around them are behaving. This has been shown in terms of people’s racial beliefs, and for people’s attitudes towards obesity, to name just a few examples. Even more interestingly, when people learn about others’ preferences they sometimes exert less effort in deciding their own, and make poorer decisions too. It’s common knowledge that much of the electorate doesn’t put much thought or effort into their decision about who to vote for. Polls facilitate that by making it easy for people defer to the consensus of people that are identified as similar to them (e.g. “Republicans” or “tea-party”).

The polls, in other words, don’t really help Ron Paul’s chances much, especially since older people get polled more than younger people. But I think it’s possible to improve the situation a bit if the way Paul’s results were framed was changed a bit. Why allow polls to prioritize older voters? Paul leads the candidates among young voters. The polls report that most young voters prefer Paul as the Republican candidate. In a party that has consistently underperformed among the nation’s young voters (and here), Paul represents a chance to win the youth vote, etc.

This is even more important if you realize that Paul doesn’t actually have to change the beliefs of older voters. In fact, people’s beliefs don’t tend to change much during an election period. (That doesn’t mean they can’t, just that for most people they won’t that much.) But that doesn’t really matter because voting doesn’t rely on a very close assimilation of the electorate’s beliefs to the candidates’ principles/positions.

The point is, we don’t have to change people’s beliefs to change their behavior (in this case their vote). A better way to change their behavior is to change their perception of the social norm (their information about the consensus). That is, people are more likely to make a concrete change in their behavior (like a vote) in response to a change in their perception of the social norm than to make a change in their abstract belief. Put in terms of this election, people are less likely to vote for Paul because they’ve reconsidered their belief about the gold standard. They are more likely to vote for Paul because (if) they perceive that most young Republicans think he’s best able to serve as President of the country.

It’s actually quite simple to recognize this when you think about how likely you’d be to wear a Ron Paul t-shirt or pin or otherwise identify yourself with his campaign in certain areas of your life. I think many people would agree that in some places it isn’t easy to put it out there that you like Ron Paul. Part of that’s because we’re not sure what kind of reception we’d get, and we imagine it might be a tiresome one. If that’s true for Ron Paul’s supporters, it’s even more true for those people who don’t really know much about the differences between the different candidates. We don’t tend to vote for someone who we think might make us different from the people we like and think are the same as us (being on a team feels good) – even if we aren’t likely to put it that way or think we’re way too independent to care what others think.

Can the Paul campaign and the grassroots make it public that Paul leads the field among young voters in the polls? Is it possible to flash-mob and grassroots-advertise and in other ways force our way into the mainstream? Of course it’s easier to interact on the internet. I know that I’m not sure I can depend on myself to do much, because it’s easier to let whatever happens happen. So perhaps the best strategy would be to put ourselves into situations where it’s harder and more uncomfortable to not advertise Ron Paul than it is to advertise him. Currently, no one’s going to bother me for not publicly representing Ron Paul’s candidacy, but there’s a good chance they might if I do. So I’m going to try and figure out ways I can put myself in a situation where I’ll definitely be bothered if I don’t, and then hopefully to avoid that irritation I’ll risk the possibility that it’ll be a little inconvenient to be associated with Ron Paul for President.

Anyway, the good news is, Ron Paul leads the polls among young voters.

Trending on the Web

Comment viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.

Polls are alot like silver

Polls are alot like silver and gold priced in dollars.
Both are manipulated and we all know the real numbers are much much higher.

Incompetence is more likely than deliberate manipulation

There are a lot of problems with polls in general, which is why they're less used in the more rigorous areas of social science. Typically the political scientists who get involved in that area are the ones who are more interested in punditry than in real research. That said, when pollsters make mistakes they tend to make them because they just don't know better, not because they're deliberately manipulative. To claim that polls are deliberately manipulate gives the pollsters and the media a lot more credit than they deserve. They operate within fairly limited networks that don't expose them to a lot of critique, and the result is they make a lot of mistakes in the way they conduct and report polls.

I think we'll get better mileage from pointing out the problems with their analyses than we will from criticizing them for being so smart that they can get away with manipulation and the malign use of power.

Could it be...

Could the fact that older voters keep turning up overly represented in these polls because polsters are calling people with land lines vs. cell phones? People with cell phones often have limited minutes so are going to be on "do not solicit" lists or might not like some pollster costing them money while they ask 100 questions. Younger people are more likely to only have cell phones. That's why we "kill" on internet polls but can seem soggy on these polls where they phone grandma and grandpa --- a demographic Dr Paul does not survey high with "He's going to take our social security!"