public opinion

Cross-posted at Montclair SocioBlog.

We’ve known for a long time that surveys are often very bad at predicting behavior.  To take the example that  Malcom Gladwell uses, if you ask Americans what kind of coffee they want,  most will say “a dark, rich, hearty roast.”  But what they actually prefer to drink is “milky, weak coffee.”

Something that sounds good in the abstract turnsout to be different from the stuff you actually have to drink.

Election polls usually have better luck since indicating your choice to a voting machine isn’t all that different from speaking that choice to a pollster.  But political preference polls as well can run into that abstract-vs.-actual problem.

Real Clear Politics recently printed some poll results that were anything but real clear.  RCP looked at polls matching Obama against the various Republican candidates.  In every case, if you use the average results of the different polls, Obama comes out on top. But in polls that matched Obama against “a Republican,” the Republican wins.

The graph shows only the average of the polls.  RCP also provides the results of the various polls (CNN, Rasmussen, ABCl, etc.)

Apparently, the best strategy for the GOP is nominate a candidate but not tell anyone who it is.


Cross-posted at Montclair Socioblog.

Peter Berger* takes issue with the phrase “on the wrong side of history.”  Mostly, he takes issue with those who use that phrase. Specifically, he refers to proponents of gay marriage who claim that the Defense of Marriage Act is “on the wrong side of history” (or in Berger’s acronym, OTWSOH) The trouble with this statement, Berger says, is that “we cannot know who or what is on the right side.”

Berger is correct (though he doesn’t offer much explanation) because the history that people are referring to hasn’t happened yet. The history of OTWSOH is the future, and we can’t know the future.  However — and here’s where Berger is wrong — we can make a pretty good guess about some things that will happen, at least in the short-run future. We can look at the trend — Americans becoming more accepting of gay marriage — and predict that the trend will continue, especially when we see that the young are more accepting than the old.

But beyond the short-run, who knows? It’s possible that the values, ideas, and even facts that are right today will, decades or centuries from now, be wrong.  So it may turn out that at some time in the future, people will think that gay marriage is a plague on civilization, that human slavery is a pretty good idea, that Shakespeare was a hack, and that Kevin Federline was a great musician.

The trouble with asking history, “Which side are you on?” is that history doesn’t end. It’s like the possibly true story of Henry Kissinger asking Chou En Lai about the implications of the French Revolution. Said the Chinese premier, “It’s too early to tell.”

At what point can we say, “This is it. Now we know which side history is on”?  We can’t, because when we wake up tomorrow, history will still be rolling on. Duncan Watts, in Everything Is Obvious… Once You Know the Answer, makes a similar point using the historical film “Butch Cassidy and the Sundance Kid.” The two robbers flee the US and go to Bolivia. Good idea? Since we know how the movie ends — that sepia freeze frame — we can safely say, “No, bad idea.”

But if we had stopped the movie twenty minutes earlier, it would have seemed like a good idea. The vindictive lawman and his posse were about to find and kill them. A few minutes later in the film, Bolivia seemed again like a bad idea – it was a miserable place. Then, when their robberies in Bolivia were easy and lucrative, it seemed again like a good idea. And then, they got killed. Butch was 42, Sundance 31.

But history is not a movie. It doesn’t end. So at least for the long run, the OTWSOH argument claims certainty  about what is at best speculation. It says, “We know what will happen, and we know that we are on the right side of history, and those who are not with us are on the wrong side of history.” Some religious folks make similar claims not about history but about God.  “We are on God’s side,” they say, “and those who disagree with us are against God.”  They tend to populate the political right.  The OTWSOH argument, Berger says, “comes more naturally to those on the left,” mostly because that is the side that is pushing for historical change.  The two sides are indulging in a similar fallacy — knowing the unknowable — a fallacy which, to those who don’t share their views, makes them appear similarly arrogant.


* Yes, this is the same Peter Berger whose Social Construction of Reality (co-written with Thomas Luckman), published forty-five years ago, has an important place in sociology’s relatively short history.

HT: Gabriel Rossman

New-ish data from the Pew Research Center suggests that inter-racial and -ethnic marriages are on the rise due to cohort changes.  First, the report shows that people who were newly married in 2008 were more likely to be married to someone of a different racial or ethnic group:

This trend is likely facilitated by greater acceptance of intermarriage.  According to the report, in 1987 less than half of Americans said it was okay for White and Black people to date each other, by 2009 that number had risen to 83%.  Among 18- to 32-year-olds, 93% approve.

Among Pew’s respondents, 63% said that they approved of inter-racial and -ethnic marriages without reservation and another 17% said that they approved of at least one type of intermarriage, but not others.  Still, overall acceptance of intermarriage still aligns with the familiar racial hierarchy in that Americans are more comfortable with outmarriages to Whites, than to Asians, Hispanics, and especially Blacks.

Acceptance of inter-racial and -ethnic marriage is on the rise, then, in part because younger people are more accepting of it than older people.  Acceptance, however, still reflects a color-based racial hierarchy.

Lisa Wade, PhD is an Associate Professor at Tulane University. She is the author of American Hookup, a book about college sexual culture; a textbook about gender; and a forthcoming introductory text: Terrible Magnificent Sociology. You can follow her on Twitter and Instagram.

Data from the American Religious Identification Survey (collected in 2008) reveals some interesting things about the population of Americans that do not identify with organized religions: atheists, agnostics, and the “spiritual but not religious.”  Most of the non-religious grew up with religious parents.  Only 17% report that neither of their parents identified with a religion:

Being non-religious does not correlate with income or education:

Instead, it’s strongest correlation is with gender.  Women are more likely than men to believe in God, more likely to convert to a faith if raised as a non-believer, and less likely to leave a faith they are raised in.

Younger people are also more likely to be non-religious:

Americans with Irish ancestry make up a significant percentage of the non-religious. They account for about 12% of Americans, but about 1/3rd of all non-religious:

More details at American Nones.

Lisa Wade, PhD is an Associate Professor at Tulane University. She is the author of American Hookup, a book about college sexual culture; a textbook about gender; and a forthcoming introductory text: Terrible Magnificent Sociology. You can follow her on Twitter and Instagram.

The figure below, featured in a paper by political scientist Larry Bartels, maps partisan identification — whether one identifies as a Democrat, an Independent, or a Republican, and how strongly — with opinions as to whether unemployment and inflation had gotten better or worse under Reagan’s presidency (1981-1988).  It shows that partisan beliefs strongly predict people’s opinions about discernable facts.

Via Gin and Tacos.

Lisa Wade, PhD is an Associate Professor at Tulane University. She is the author of American Hookup, a book about college sexual culture; a textbook about gender; and a forthcoming introductory text: Terrible Magnificent Sociology. You can follow her on Twitter and Instagram.

Most Americans, when asked if they are affected by advertising, will say “not really.” They say they skip the print ads in magazine, ignore the ones on the street, mute TV commercials, and are generally too savvy to be swayed by their messages.

Here’s some data illustrating the not-me phenomenon. The Kaiser Family Foundation asked 15- to 17-year-olds whether they and their friends were influenced by sexual content on TV.

Seventy-two percent of teens say that sexual content on TV affects their friends “a lot” or “somewhat”:

But only 22 percent say that sexual content on TV affects them “a lot” or “somewhat”:

Advertisers know that most Americans are wrong about whether advertising affects them.  That’s why they spent $117 billion in 2009 trying to convince you to buy their product. It works. So it must be affecting somebody, right?

Images borrowed from Strasburger’s Children, Adolescents, and the Media.

Lisa Wade, PhD is an Associate Professor at Tulane University. She is the author of American Hookup, a book about college sexual culture; a textbook about gender; and a forthcoming introductory text: Terrible Magnificent Sociology. You can follow her on Twitter and Instagram.

Talking Points Memo posted an article about a study recently released by the Program on International Policy Attitudes at the University of Maryland. The study looks at misinformation about issues related to the 2010 election among the U.S. electorate. The survey sampled 616 individuals who reported that they voted in the November elections, and according to their methodology, was chosen to be representative of the overall U.S. population. After individuals were chosen to take part, they were asked to complete an online survey; those who didn’t have access to a computer were provided with a laptop and internet access. You can read more about this method of collecting data, which uses an online program called Knowledge Networks, here.

Of course, any study of misinformation brings up the tricky question of how to identify what is “true,” and how to do so in a way that isn’t itself political. The authors explain at length:

…we used as reference points the conclusions of key government agencies that are run by professional experts and have a strong reputation for being immune to partisan influences. These include the Congressional Budget Office, the Department of Commerce, and the National Academy of Sciences. We also noted efforts to survey elite opinion, such as the regular survey of economists conducted by the Wall Street Journal; however, we only used this as supporting evidence for what constitutes expert opinion. In most cases we inquired about respondents’ views of expert opinion, as well as the respondents’ own views…in designing this study we took the position that some respondents may have had correct information about prevailing expert opinion but nonetheless came to a contrary conclusion, and thus should not be regarded as ‘misinformed.’

On some issues, such as climate change, there is a vocal dissenting minority among experts. Thus questions were framed in terms of whether, among experts, more had one or another view, or views were evenly divided.

The researchers first asked if respondents believed they had seen or heard misleading or incorrect information during the fall campaign. Overall, a majority of voters said they had encountered misinformation during the election, and over half said there was more misinformation this time than usual.

The results also indicated relatively high levels of misinformation on a number of questions. For instance, 40% thought the Trouble Assets Relief Program (TARP, or the bank bailout) was passed under President Obama, when it was actually passed under President Bush; 43% didn’t know that President Obama has increased the number of U.S. troops in Afghanistan. And overall, respondents seemed quite confused about financial policies, with overwhelming majorities of both Republicans and Democrats getting questions about taxes, the stimulus, and the auto maker bailout wrong:

However, for most items, Democrats and Republicans tended to differ on which issues they were misinformed about, in fairly predictable ways:

Finally, the study found that source of information seemed to play a role. Those who said they watched Fox News were more misinformed than any other group, and the more they watched it, the more misinformed they were — whereas with most other news sources, the more news individuals consumed, the less misinformed they were. Viewers of Fox News were more likely to believe the following (incorrect) statements (from p. 20 of the report):

  • most economists estimate the stimulus caused job losses (12 points more likely)
  • most economists have estimated the health care law will worsen the deficit (31 points)
  • the economy is getting worse (26 points)
  • most scientists do not agree that climate change is occurring (30 points)
  • the stimulus legislation did not include any tax cuts (14 points)
  • their own income taxes have gone up (14 points)
  • the auto bailout only occurred under Obama (13 points)
  • when TARP came up for a vote most Republicans opposed it (12 points)
  • it is not clear that Obama was born in the United States (31 points)

This pattern persisted regardless of political affiliation — Democrats who reported watching more Fox News were more misinformed than other Democrats, though less so than Republicans who watch the same amount of Fox News.

This table shows the number of respondents who said most experts believe climate change either isn’t occurring or that the scientific community is evenly split, by source of news and how often they view that source (p. 21):

They have the same breakdown of data for each question. In general, the lowest levels of misinformation were found among those who reported high levels of consumption of news from MSNBC and/or PBS/NPR. However, as the study authors point out, for a number of questions (such as those about the effects of the stimulus program), all groups had quite high levels of misinformation.

Of course, this leaves a number of questions unanswered: are people more misinformed because they watch Fox News? Or are misinformed people more likely to watch Fox News at least in part because it is more likely to reinforce ideas they already have?

And how does the choice of these particular questions, out of all the potential questions we could ask to judge how well- or poorly-informed people are, affect the results? I suspect critics might say that many of these questions are ones liberals are more likely to get right simply by answering based on political ideology, regardless of actual knowledge — for instance, someone who is Democratic might be more likely to say the health care bill wouldn’t add to the deficit, and thus be “right,” but answer that way because health care reform was a Democratic-backed policy and thus something they supported, not because they have any concrete knowledge about it. As we see with the question about the Chamber of Commerce, when a question doesn’t fit so well with liberal-leaning views (the Chamber of Commerce tends to be more popular among conservatives), Democrats showed high levels of misinformation as well. If we asked more of those sorts of questions, would we find that Democrats (or, say, those who report PBS or NPR as their main source of information) were more misinformed than Fox viewers?


Abi, a professor of Materials Engineering at the Indian Institute of Science, pointed out an interesting graph posted by Ezra Klein at the Washington Post. The graph, using data from a survey of public opinion about the U.S. budget, shows how much of the U.S. budget respondents believe goes to foreign aid, how much they think should go to foreign aid…and how much actually does:

The actual figure? 0.6% in 2009.