Pollster.com

Articles and Analysis

 

Low Information Voters and Television Ads

Topics: 2008 , Associated Press , Barack Obama , Hillary Clinton , Jon Cohen , Pew Research Center

TNR's Michael Crowley blogged this question from a reader last night:

I haven't seen all the exit data, but listening to the talking heads it seems Obama is finally cracking the code of the working class white voter. But maybe it is simpler than that.

Could it be that downscale voters are also "low-information" voters when compared with their Volvo-driving broadband-surfing upscale brethren? If so, it would suggest that all it was going to take was a bit of time for the word to get through that Obama is looking like a winner. Wealthier voters may be the leading edge of a wave. Downscale voters may be the ones who catch trends later--and then really give them mass market power. If so, it's bad news for Senator Clinton going into Ohio and Pennsylvania.

I can answer part of that question. Downscale voters -- those with less education and lower incomes -- are absolutely "low information voters" as compared to their more upscale brethren. That finding has been a consistent theme of 40 or 50 years of political survey research and was vividly reaffirmed by the updated political knowledge study released by the Pew Research Center last year. Here is a slightly condensed version of a table from that report showing that lower income and less well educated Americans are by far the least informed on an index based on 23 questions of political knowledge:

02-13 Pew knowledge table.png

The harder question to answer is whether the Virginia and Maryland results represent some sort of "breakthrough" for Obama among downscale white voters. It is not immediately clear from the exit polls, as Noam Scheiber put it, whether "Obama's strong showing in Virginia a sign of an expanding coalition, or...the predictable result of a contest waged on favorable terrain." Scheiber sees signs of "genuine growth for Obama" in the exit poll results, although he sees signs of improvement among older whites and Catholics. It is hard to be certain about any progress among downscale whites since, as Scheiber points out, the exit poll tabulations do not break out the results by education and race or by income and race.

A different but important piece of this puzzle comes from exit poll question that received suprisingly little attention last night (at least on the coverage that I watched). In both Virginia and Maryland, exit pollsters asked voters to "rate the importance of campaign ads...in your vote in today's presidential primary." As the table below shows, Obama did much better in both states among those who rated political ads as important.

02-13 exit poll ad importance.png

The Associated Press reports that Obama "far outspent his rival on television advertising" in Virginia and Maryland, and these exit poll results are consistent with a perceived Obama advantage in campaign ads. A Clinton supporter who saw nothing but Obama advertisements on television is more likely to say the ads were "not important," while an Obama backer would be more inclined to react favorably to ads that seemed mostly about Obama.

It is thus difficult to tease out from these results the actual importance of the ads in persuading voters in Virginia and Maryland. Still, this is a topic worth digging into more deeply. Did the content of those advertisements -- or simply the fact that Obama had a substantial advantage on that score -- help move some of the lower information voters that have been more supportive of Clinton in other states?

Update: The Washington Post's Jon Cohen reports some additional and highly relevant exit poll numbers which we discuss here.

 

Comments
Dyna:

The "Volvo driving" stereotype is worn out. Volvo doesn't even make cars anymore, they sold that division to Ford. They do make lots of big trucks though... Today's Volvo driver is a blue collar driver who uses his or her laptop with the Wi-Fi at the truckstop to do business and keep in contact with family and friends on the road.

____________________

Jeff:

Gang: SurveyUSA is out with two new versions of its pollster report card, here:

http://www.surveyusa.com/index.php/2008/02/13/2008-pollster-report-card-updated-to-include-potomac-primaries/

One is done by median error, where SurveyUSA itself is tied for first; the other is the mean error. Both updated through yesterday's primaries.

I know it doesn't fit this blog topic, but just noticed it

____________________

Chris S.:

Crowley's colleague at TNR, Noam Scheiber, has written about this phenomenon before, that downscale / low information voters don't follow the election that closely, so they basically end up going with whichever candidate is seen to be "stronger", which often translates into whichever candidate is winning. This possibly explains why downscale voters went with Kerry in the 2004 primaries, even though Edwards would seem like a natural fit for them.

The TNR archives from 2004 are still accessible via archive.org:

http://web.archive.org/web/20040312124126/http://www.tnr.com/etc.mhtml?pid=1342

Quote from Scheiber in Feb. 2004:

--------------

"Wisconsin exit polls are turning up what looks like a paradox: Despite John Kerry's aloof-liberal-Brahmin rap, and despite John Edwards's heavy "son-of-a-mill-worker" shtick, Kerry did better last night among less educated, less affluent, blue-collar, and rural voters than he did among more educated, more affluent, white-collar, suburban voters, while the opposite was true for Edwards.
.
.
.
My own hunch is that what we're seeing is an important divide between less sophisticated voters and more sophisticated voters. Just about the only thing less sophisticated voters--who, I'm guessing, tend to be poorer and less well-educated--know about John Kerry is that he's been winning, and possibly that he's a longtime Senator and a Vietnam veteran....Which is to say, less affluent, less educated voters are looking at John Kerry's string of primary victories and concluding from them that he's electable."
-----

If downscale voters make the same conclusions from Obama's victories, then Clinton's in big trouble.

____________________

Alan:

I remain very offended by the language of this post. Every voter has one vote to cast (or not cast) however they want. They know what is important to them: their jobs, their housing, their transportation, their daily life and routine. Who are you to decide that the ability to questions that are only of interest to people in specialized fields like politics or economics makes someone a "low information" or "high information" voter? Where have you scientifically established the relevance of the questions? Where have you scientifically established the basis for the implied value judgments of "low" and "high"?

Furthermore, there appears to be a serious flaw in your scholarship. You have (a) correlated education level and income to an arbitrary "knowledge" scale (that you misidentified as an "information" scale). Second, in a completely separate study, you correlate education level and income with votes cast in the Virginia and Maryland primaries. Then, you make a huge and unsupported leap to claim that there must be a correlation between the "information" scale and the votes cast in the Virginia and Maryland primaries. In fact, this conclusion has not been established in the least. It could well be that the lower income voters at the top of your "information" scale cast their votes differently than the lower income voters at the bottom of your "information" scale, and that while there is a correlation between income and votes, there is no correlation, or even a different correlation, between "information" and votes. You have no way of knowing because you have no data that shows "information" vs. votes for any individuals.

This is what I suggest. If you have identified a group based on income or education, then please stick to reporting the results based on income and education. Do not pull out some arbitrary "information" scale that was not measured by the exit poll and use that to assign implied value labels to the voters. That is unnecessary, in my opinion not helpful, assigns an implied value judgment where none is needed, and last but not least, a scientific error because it draws a conclusion that cannot be made from the data you have.

____________________

Alan:

I should add that the original survey in question in fact measured what should be at best identified as political knowledge. What the survey uncovered is a correlation between income and the study's measurement of political knowledge. If the study were redone to measure automotive repair knowledge, or electrical wiring knowledge, or child care knowledge, it might well find just the opposite correlation, the as income increases, "knowledge" decreases. It is sloppy scholarship to measure a specialized form of knowledge and then present is simply as "knowledge". It is also sloppy scholarship to not analyze the relevance and measurement limitations of particular form of knowledge that was measured.

____________________

VolvoJoe:

Volvo doesn't make cars anymore? That's the weirdest statement I've read in a while. Hopefully it's a weird stab at humor, since there are lots of liberal, well-educated people out there driving cars they believe are Volvos, that certainly say "Volvo" on them, regardless of who the parent company might be. Are Cadillacs not Cadillacs since the original owner sold out to GM decades ago?

____________________

Alan:

Joe, there is in fact a company called Volvo which no longer makes cars. See http://www.volvo.com/ . I guess you can add yourself to the "low information" voter group, since you did not know this important and highly relevant piece of information.

____________________

jamzo:

the right to vote is given to each citizen without regard to his or her "political knowledge"

each person's opinion is equally important regardless of color of skin, gender, level of education, way of earning money, height, weight, ethnicity, hair color, number of fingers and toes,

the simplistic people-groupings of the survey technicians and their politician and media paymasters might be helpful in providing context to campaign questions but they certainly make for bad tv

chris matthews struggled last night

every time he said the word race

he stammered, then aplogized, then rationalized the need to make the racial comment

tv seems to think that breaking exit polling into simplistic groups by color, gender, wages, and education makes for interesting tv

if i had the opportunity to script tv shows and try to make money reporting election results i wonder what i would do?

i might use issues...

i might....

____________________

jsh1120:

Alan,

Your comments are well-taken. The Pew Research study clearly measured textbook political knowledge as opposed to current events or campaign knowledge.

Nevertheless, your complaint that if another type of knowledge had been measured the results would not have shown the same education/income correlation is, I think, misplaced. Political knowledge (as opposed to automotive repair knowledge) is strongly correlated with attention to political campaigns (the unmeasured attribute being discussed.) You're undoubtedly correct that automotive knowledge might well be associated with purchase of a particular brand or attention to automobile commercials, but that's not the focus here.

In general, I'm inclined to give more weight to the "momentum" argument stemming from information saturation than the "breakthrough" hypothesis where Obama's gains are attributed to conversion of specific demographic groups. But as noted in the discussion above, it's difficult to resolve that question with the available data.

____________________

Alan:

My point was: don't call it "knowledge" if in fact it is a specialized form of knowledge (and what isn't?), and further, identify exactly what form of "knowledge" you are measuring and how it is relevant.

> Political knowledge (as opposed to automotive repair knowledge) is strongly correlated with attention to political campaigns (the unmeasured attribute being discussed.)

You are still trying to use this to make a two-step correlation: "A is correlated to B and B is correlated to C; therefore A must have the same correlation to C". This is not mathematically valid. In order to correlate A with C, you need data that measures both A and C in a sample of individuals. You cannot take one study that measures A and B in one sample of individuals and combine it with another study that measures B and C in a different sample of individuals, and use that to conclude there is a correlation between A and C.

____________________

Alan:

My point was: don't call it "knowledge" if in fact it is a specialized form of knowledge (and what isn't?), and further, identify exactly what form of "knowledge" you are measuring and how it is relevant.

> Political knowledge (as opposed to automotive repair knowledge) is strongly correlated with attention to political campaigns (the unmeasured attribute being discussed.)

You are still trying to use this to make a two-step correlation: "A is correlated to B and B is correlated to C; therefore A must have the same correlation to C". This is not mathematically valid. In order to correlate A with C, you need data that measures both A and C in a sample of individuals. You cannot take one study that measures A and B in one sample of individuals and combine it with another study that measures B and C in a different sample of individuals, and use that to conclude there is a correlation between A and C.

____________________

Dr. Brad Burke:

Lot's of poorly informed comments here. I think they have misunderstood the intent of what you were saying.

Forever and ever, pundits have divided the electorate into demographics, and then tried to figure out who did or will vote for who. In the old politics, candidates have long attempted to redraw their message to demographic groups.

But this article points to a larger concept which I have believed for a long time... It's not just who voters ARE, it's more how they THINK. The art of voting (at least in this Democratic primary) is not completely unlike decisions regarding typical consumer purchases. You have people who lead new trends (early adaptors), those who typically wait years before trying new things (laggards)- in fact a whole spectrum based upon things like risk, fear, comfort, and other psychosocial factors.

I believe what you said, Alan, is dead on. I have been saying for weeks that the voters who were more actively searching out information were Obama voters. Those who casually observed or didn't really gain much info were Clinton voters. This relates both to name recognition and the idea that those who are the most interested in deep study of the campaigns are also more likely to desire change over security. This has been clearly borne out in exit polls and demographic studies during this primary.

This campaign has truly been a contest between change and the status quo. It's not that change is better for all voters. Many, such as has been generalized in older Latino voters, are simply more comfortable with the already known quantity. So the innovators were the first to jump on the Obama bandwagon. The late adopters (read conservative decision-makers) are still strongly with Clinton. But what we have seen is movement in the middle- as more people in more groups become more aware of "the new guy" they have become more comfortable with him and his policies (even if they come from limited info sources like TV ads). He's not worn out, but he's not as stiff and scratchy like a brand new shirt. He's not as "scary" (read risky or uncertain) to a large group of voters as he was just a few weeks ago.

I have of course generalized. And every voter will make their own decision, regardless of their demographics. But I think this article brings up an important point of discussion. And I think the wise political minds will begin to look more at the thinking patterns of voter groups rather than old divisive categories. It is certainly obvious that Obama, as a candidate and a movement, is rapidly spreading to more and more people- even those who take fewer risks.

____________________

Adam:

Alan -- Volvo does make cars still. But the automotive division is now owned by Ford. But so is Aston Martin, and 51% of Mazda, and Land Rover, to name just a few. The factories, processes, etc., are all the same.

Aston Martin doesn't make cars? Mazda doesn't? Land Rover doesn't make the Range Rover et. al.? I think you've got that one wrong.

____________________

DTM:

Alan,

First, I think most of the people using phrases like "low-information voters" would gladly agree that the phrase "low-information" is intended to refer only to general information about politics and specific information about the particular electoral contest in question, and not to other kinds of information. Of course if that isn't clear to a substantial number of people from the context in which the phrase is being used, then perhaps it is worth making that intention more explicit, as you suggested.

Second, I also think at least many of those people would agree it is very hard to draw any definitive conclusions from the current exit polls when it comes to this issue, for the reason you identified: the exit polls for the most part do not ask the questions we would need them to ask in order to directly test these hypotheses. Still, I don't think it is inappropriate to use the available data to explore these questions to the extent possible. As I would put it, the available data may not be able to decide the issue, but we may be able to use it to at least test how plausible the relevant hypotheses might be.

And personally, that is about how far I think we can get. Specifically, I do think it is plausible that Obama is gradually making progress among voters who have relatively little information about the contest, perhaps as a result of things like the ongoing coverage of his recent successes, his own advertising campaigns, and so on. But I also think we cannot confirm those hypotheses with the limited data set available to us.

____________________

Jane:

The young tend to be early adopters in any event but also they lack the experience of following Hillary Clinton's behavior over decades so it has been much easier for Obama to sell them a false picture of who she is.

Yes, Obama will become more popular as people become more familiar with him but his popularity will dip as people subsequently become aware of his flaws. The Nuclear Leaks story is an almost perfect example of his tendency to try and placate the middle and to gave to supporters like Exelon and pretend that he hasn't done so. Supporting Lieberman in the Connecticut primary while posing as anti-war is another example. The Democratic activists who have been following at this level of detail are one of the hardest groups for him to crack.

I would be interested in seeing a poll which rates both groups of supporters on basic political facts about each candidate: fact based questions such as who voted which way etc. -- not position questions which are debatable.

____________________

Tim in AnnArbor:

A point of information regarding,
"The "Volvo driving" stereotype is worn out. Volvo doesn't even make cars anymore, they sold that division to Ford. They do make lots of big trucks though... Today's Volvo driver is a blue collar driver who uses his or her laptop with the Wi-Fi at the truckstop to do business and keep in contact with family and friends on the road. Posted by: Dyna | February 13, 2008 12:36 PM."
-- --
It's true that Ford purchased Volvo Cars a number of years ago, but it's _not_ true that "Volvo doesn't even make cars anymore." Except for some exchange of intellectual property (with a majority of the IP flowing from Volvo to Ford), Volvo operates almost as independently as it did before its purchase by Ford -- even slightly more independently than either Jag or Land Rover do, and far more independently than, for example, Ford of Europe or Ford of Australia do.

____________________

Tim in AnnArbor:

Point of information regarding,
"Joe, there is in fact a company called Volvo which no longer makes cars. See http://www.volvo.com/ . I guess you can add yourself to the "low information" voter group, since you did not know this important and highly relevant piece of information. Posted by: Alan | February 13, 2008 1:33 PM."
-- --
Also TRY: http://www.volvocars.com

____________________

DTM:

This debate over the meaningfulness of car brands is not new, of course, nor limited to cars for that matter. Personally, on the one hand I think it is a bit extreme to insist, as some purists do, that any common ownership or shared supply chain implies that the two product lines are not meaningfully distinct. On the other hand, I think it is worth noting that it is not an uncommon practice for companies to acquire the rights to a brand name and to then apply it to products which have little in common with the previous products made under that name, with the possible exception of some superficial styling cues.

So, to me this is a matter of degree, varying from products which have merely been "rebadged" to products which are entirely independent and unique. And the truth is that in the modern automotive industry, there are few if any mass-produced cars which are anywhere close to being 100% independent and unique.

____________________

Daniel T:

I found much of this discussion misguided. Fundamentally, we don't understand why voters vote the way they do on an individual level; the reasons are many and varied. But what we can say is that *at a group level* certain groups with shared characteristics *tend* to vote in similar ways. We don't know *why* they do, but they do. And the "why" is the real issue.

I also find it very curious as to why low income people tended to oppose Edwards, who it was in their interest to support. It is no good to say they were in the "low information" group because obviously they had enough information to make some type of decision and, from a practical perspective, that's all that matters. But likewise, it makes no difference if they were "late adoptors" or "going with the momentum". We are still left with the original question (only at second hand) which is, to repeat, "why are low income voters behaving in a way that seems to be politically disadvantageous."

I would like to suggest that the answer to this question can be found in educational theory. A great deal of work has been done around theories of cultural capital. Low income individuals are not only more likely to be less informed politically, it has been demonstrated for a 100 years that they are less involved educationally as well. For example, low income individuals are far more likely than high income individuals to say that the government is responsible for educating their children as opposed to taking responsibility themselves. Indeed, the most common feature in other fields of low income groups is their overall passivity.

My point is that the behavior of low income individuals is not an issue that is limited to political science; it is an issue that has been studied extensively in the fields of sociology, psychology, and education. While Alan is technically correct about correlation, part of the reason that a=c is appealing is because we know from other fields of study that a=c. It would be odd if it didn't in political science (although that is no excuse for bad research).

____________________

john l:

While I don't entirely agree with everything Alan is saying, as someone with a bachelors in Psychology and Pershing a masters in Industrial/organizational psychology I have had one topic persistently pounded into my brain and that is correlation does not equal causation. With that being said how many people commenting on these forums have an education level lower than college? Do poorer people have as much access to information? Do less educated people have the knowledge of how to access the information they are looking for? Merely questions posed, not assumptions. It would be interesting to find out.

____________________



Post a comment




Please be patient while your comment posts - sometimes it takes a minute or two. To check your comment, please wait 60 seconds and click your browser's refresh button. Note that comments with three or more hyperlinks will be held for approval.

MAP - US, AL, AK, AZ, AR, CA, CO, CT, DE, FL, GA, HI, ID, IL, IN, IA, KS, KY, LA, ME, MD, MA, MI, MN, MS, MO, MT, NE, NV, NH, NJ, NM, NY, NC, ND, OH, OK, OR, PA, RI, SC, SD, TN, TX, UT, VT, VA, WA, WV, WI, WY, PR