by Tony Woodlief
“I would rather be governed by the first 2,000 people in the telephone directory than by the Harvard University faculty.” —William F. Buckley, Jr.
While American partisans have altered their policy opinions to match the ideologies of the political class, regular Americans have ignored that marching order. Partisans no longer agree with the Other Side on anything, but average Americans don’t let team allegiance dominate their views. Even most Americans who are registered as Democrats or Republicans still favor some policies desired by majorities in the other party. Average citizens demonstrate greater independence of thought than the ideological conformists so revered by political scientists.
Political scientists still contend, however, that Americans are in no condition to vote responsibly, let alone engage in self-governance. By evaluating citizens through a unidimensional, ideological lens, they’ve concluded that there’s neither rhyme, reason, nor consistency behind our voting patterns. In the words of Kinder and Kalmoe, “many Americans simply don’t know what they want from government.”
Is the fact that most Americans don’t flock to either pole on the ideological spectrum proof that their opinions aren’t held together by an underlying value system? Some opinion researchers have pondered the possibility that force-fitting survey answers to the liberal/conservative spectrum incorrectly casts everyday Americans as flighty and unserious. “Perhaps ordinary citizens’ issue preferences lacked ‘constraint,’” speculate Christopher Achen and Larry Bartels, “because they had thoughtfully constructed their own personal political belief systems transcending conventional ideologies and party lines?”
Despite more than a hint of derision in this question, there is indeed evidence that when we unpack the liberal/conservative continuum, Americans aren’t as scatterbrained as scholars make them out to be. Researchers have separated American opinions about foreign policy and defense, economic policy, and moral issues, and found that we hold somewhat consistent beliefs within those issue areas. The problem for theorists is that, when taken in total, those beliefs don’t fit onto their unidimensional, ideological spectrum. Many Americans, for example, are very religious and oppose abortion, which political scientists would consider conservative. Yet these same people also embrace government aid to minorities and laws that ensure equal access to job opportunities, which are typically considered liberal positions. Likewise, a significant portion of America embraces free enterprise and limited government spending, but also abortion rights and gay adoption. We’re “liberal” on some things, and “conservative” on others, and this drives theorists batty.
The damning reality about Americans, however, which makes political scientists so confident in their negative assessments, is that our political opinions fluctuate from year to year. One year we favor more aid to minorities; two years later we oppose it. We say the government should provide health insurance, then we say it shouldn’t. We believe the US should intervene less in foreign affairs, then we’re for war. If average Americans did have an underlying value structure informing their policy preferences, goes the reasoning, their survey answers wouldn’t jump around so much. Maybe there’s an ideology that leads someone to be for both free trade and government-provided health insurance, but there’s no ideology that leads a person to favor these policies one year, and disapprove of them the next.
Before we throw in the towel on the American mind, however, let’s consider a pretend survey question. It asks you to express, on a seven-point scale, your agreement or disagreement with this statement: “People will be better off if they have children.” One means you very strongly disagree; seven means you agree very strongly.
Before you protest that this question lacks all context (“What people?” “How old are they?” “How many children will they have?”), let me remind you that I didn’t make the survey rules. Here’s a real statement, for example, that the American National Election Studies has used for more than sixty years to evaluate American opinions about foreign policy: “This country would be better off if we just stayed home and did not concern ourselves with problems in other parts of the world.”
Let’s tease that one out before returning to my hypothetical childbearing question. Imagine a survey respondent in 2002 who has just recently watched President George W. Bush’s “axis of evil” speech. Inspired by the urgent imperative to stop weapons of mass destruction from proliferating in the hands of evildoers, he might feel strongly that the US will be worse off if it doesn’t get more involved overseas. So, he chooses “disagree” on the survey.
Fast forward to 2012. American soldiers have suffered terrible casualties in Iraq and Afghanistan. Reports of civilian deaths and dismemberments, meanwhile, are staggering. It now appears clear that intelligence failures, even deliberate misstatements, were used to justify our military interventions. No weapons of mass destruction were found, and many of the people we told ourselves we were going to liberate want us out of their countries.
Our same citizen, finding himself once again contemplating this survey question, has significantly cooled toward military interventions. Maybe he still believes some kind of action was warranted against the people who masterminded the September 11th attacks, but he no longer supports wholesale invasions. So, this time he chooses “agree.”
This is entirely reasonable logic. One might even argue that this citizen has better judgment — and certainly more humility — than the politicians and bureaucrats who remain resolutely unapologetic for plunging America into a twenty-year war costing $6.5 trillion and more than 7,000 American lives. Yet still this respondent will be judged as inconsistent by pollsters, because he changed his answer to their survey question.
Now, imagine what might go through someone’s mind when answering my hypothetical survey question about having children. One respondent has just been around parents yelling at their kids in the local park. Another has an unmarried teenage niece with a six-month-old baby. Still another recently watched her daughter win a state championship in wrestling. Do you think this personal context matters? If it does, do you think the very same people, two or four or eight years later, might give significantly different answers that will have been colored by their recent experiences? If so, does this prove they have inconsistent beliefs about children and parenting?
Of course it doesn’t. Ordinary people, when asked abstract philosophical questions, will draw on recent, concrete experience to inform their answers. Cognitive psychologists Amos Tversky and Daniel Kahneman labeled this tendency “availability bias.” “Perhaps the most obvious demonstration of availability in real life,” they wrote in the academic journal Cognitive Psychology, “is the impact of the fortuitous availability of incidents or scenarios.” Asked to evaluate abstractions, the natural human response is to draw upon experience, and often our most recent experience is the most accessible. You may hold bedrock beliefs that are proven out in the way you live and how you treat others, yet which don’t shine through in opinion surveys that, lacking sufficient context, invite emotion, varied interpretations, and recency to affect your choice of a number on a scale.
With that said, there is a small group of people whose answers to the parenting question wouldn’t vary: those with such strong convictions about childbearing that context doesn’t matter. Some people have firm religious beliefs, for example, that everyone is called to “be fruitful and multiply,” and that God will work out the circumstances. Others believe the world’s resources are so depleted that it’s imperative for everyone to stop having children before the planet dies.
People who believe a principle should be adhered to no matter what the cost — ideologues, in other words — are likely to be much more consistent in survey after survey than the rest of us. What’s more, they’ll struggle to imagine how people whose responses depend on context can be anything other than shallow. If they happen to be the scholars constructing the surveys and interpreting the results, well, you get today’s near-consensus about American public opinion, which is that most citizens are shortsighted, biased, forgetful, and relatively unprincipled. Not well-suited, in other words, to govern themselves according to the vision of the American Founders.
It’s worth noting that a more charitable view of their fellow man might evoke curiosity among scholars about why their surveys indicate citizenship incompetence among a wide swath of Americans. Given a well-established psychological literature revealing the human tendency to explain one’s own behaviors (and inconsistencies) with more grace than one generally affords others, we might be justified in saying to the professors who hold such a damning view of everyday Americans: “Physician, heal thyself.”
The plain truth is that the machinery of public-opinion surveys, crafted by ideologues, is geared to detect ideology. Ideology is the only mechanism they imagine can drive political opinion in a coherent, predictable direction. There are entire academic treatises on the nature of ideology, its formation and its actualization. The fact that you obey the law, pay your taxes, and participate in the market economy is proof, in some interpretations, that you are embedded in a web of ideology. That discussion is not worth delving into here. The question at hand is whether Americans have shared beliefs that not only lead them to respond to surveys with answers that cut across the academic conceptualization of liberal vs. conservative, but which also explain the variation in their policy preferences over time.
Can “We the People” Be Trusted?
This isn’t just a philosophical question. What’s at stake is the American republic. If most citizens really are indifferent, and the remainder blindly partisan, then the faith of the Founders was mislaid, and we are in no condition to govern ourselves. Far better to leave all those policy decisions to the attentive politicos in DC, provided we can find a way to keep them from plunging us into civil war.
In short, where we go from here depends on an honest answer to this question: is there something other than capriciousness and low information that drives the political opinions of average Americans? Something that makes their desires for our country trustworthy?
I believe there is, for two reasons. First, public-opinion scholars assess citizens’ knowledge of issues and politics like they’re grading a midterm exam. Political scientist James Gibson points out, for example, that President Richard Nixon repeatedly mangled the name of the man he himself nominated to the Supreme Court, William Rehnquist. Coding standards applied by the American National Election Studies would have required pollsters to record Nixon as not knowing his own Supreme Court nominee. So were real survey respondents, when presented with William Rehnquist’s name, recorded as not knowing who he was if they answered with something like: “head honcho of the Supreme Court.” Using a more reasonable standard of knowledge, Gibson found that 72 percent of responses recorded as wrong were in fact correct. The surveys employed to prove most Americans are ignorant, in other words, appear to be off the mark.
More importantly, surveys of American beliefs about government focus almost exclusively on policy levers. They ask whether respondents believe government should provide health insurance, and whether it should spend more or less on welfare, public health, education, crime—even space exploration. They ask whether courts should be harsher or more lenient with criminals, and whether more or fewer immigrants should be allowed into the country. This is like asking an average person to diagnose his car trouble from the driver’s seat. He can no more tell you how much welfare spending is adequate than he can determine how much transmission fluid he’s lacking. If you force him to offer diagnoses, he’s going to grasp at impressions. Wasn’t there a grinding sound last week? Didn’t I notice a funny smell?
Ask him repeatedly over the years, and his answers are going to jump around. All his seeming schizophrenia proves is that he isn’t a mechanic. It doesn’t mean that he has no consistent vision of where he wants his car to go, and how he wants it to get there. Likewise for Americans when it comes to their conceptualizations of government and the common good. Everyday people don’t know how much money government at all levels spends on education, or how much it should spend. This doesn’t mean they lack coherent opinions about what a child’s education ought to look like.
Instead of Survey Respondents, Citizens
Imagine that instead of asking Americans to be government mechanics, we instead asked them to think like citizens. Rather than quiz them regarding what policy levers ought to be pulled, they would be questioned about what outcomes they believed were best for our country. People disagree vehemently about government-provided health insurance, for example, but share a desire to see as many Americans as possible receive the medical care they need. People disagree about whether parents should be allowed to divert public funds to private schools, but share a desire to see every American child receive a suitable education. We’re divided over what levers to pull, but not nearly so divided as the political class when it comes to the ends we want to achieve, because we are far more united in our core values than they are.
How do I know? Most directly, I have experienced this reality firsthand — as I suspect you have as well — in many conversations with friends, neighbors, relatives, coworkers, and fellow parishioners whose political opinions vary. Beyond one’s own sense from those conversations, hints of an underlying consensus on American values can be found in the same surveys used by scholars to claim American beliefs are incoherent. Occasionally, a question on those broad national surveys reveals — perhaps without meaning to — values and desires of Americans regarding public policy and the common good.
The American National Election Studies, for example, has asked Americans for decades how they feel about government support for people who need jobs. Between 1956 and 1960, on average 58 percent of Americans said the government should see to it that people who needed jobs should get them. Opposing that goal were 26 percent of respondents, with another 17 percent stating either that they didn’t know or didn’t care. In 1964, however, the percentage of respondents who agreed with this lofty aspiration fell almost by half, to 31 percent. Those who disagreed, meanwhile, rose to 43 percent.
On the surface, this appears to be another example of American ideological schizophrenia. Either that, or a sizable portion of Americans lost their charitable instincts in four short years. A closer look at the question’s wording, however, reveals that between 1956-1960, Americans were asked to either agree or disagree with this statement:
The government in Washington ought to see to it that everybody who wants to work can find a job.
The statement’s wording was altered in 1964, however, replacing a simpler declaration with this version:
In general, some people feel that the government in Washington should see to it that every person has a job and a good standard of living. Others think the government should just let each person get ahead on his own. Have you been interested enough in this to favor one side over the other?
This is a very different question, isn’t it? Before, Americans were asked whether they wanted government to help everyone willing to work find a job. The revised question, in contrast, asked whether Americans believed government should provide everyone a job (no mention of willingness to work), and beyond that a good standard of living. The dramatic change in subsequent survey responses doesn’t simply illustrate the sensitivity of surveys to how questions are worded. It illuminates, as demonstrated with greater proof below, a core conviction that informs how everyday Americans think about everything from welfare to immigration, namely that we should help people who are trying to help themselves.
This is a widespread and stable value that directly affects how Americans feel about welfare, preferential hiring, aid to minorities, immigration, and other policies. The majority of survey questions about these topics, however, pretend this sentiment doesn’t exist. The two most consistently administered and academically rigorous survey batteries in America — the American National Election Studies (University of Michigan) and General Social Survey (University of Chicago) — don’t ask Americans to distinguish between welfare recipients who have one child out of wedlock versus three, or immigrants willing to work versus those who subsist on crime or welfare. Yet these are exactly the considerations, as anyone who’s talked to regular Americans for even a few minutes about these topics can attest, that determine how generous or stingy most Americans will be. Little wonder responses to survey questions about how much we should spend on social services, or how many immigrants we should allow into the country, fluctuate. Lacking context, respondents base their answers on what’s prominent in the news or other media, alongside immediate personal experiences.
Another effect of the aforementioned change to the wording of the jobs question reveals something else about how Americans respond to surveys. When surveyors altered the wording in 1964, the percentage of respondents who subsequently replied that they either didn’t know or didn’t care rose by more than half, from 17 percent to 26 percent. When the surveyors switched, eight years later, from a Yes/No format to a seven-point scale, the “don’t know” responses fell by half. Forty-two percent of respondents, furthermore, placed themselves in the middle of the scale, choosing a three, four, or five.
Between 1964-1972, in other words, ANES administrators forced respondents to consider a false choice: either government guarantees everyone a job and a good standard of living, or it leaves people to fend for themselves. Anyone acquainted with everyday life understands there’s a third alternative in which assistance comes from families, churches, and communities. This isn’t an uncommon phenomenon in surveys. Ideologically-minded researchers manufacture false choices, and Americans respond by either opting out of the questions altogether, or placing themselves in the middle of a scale when it’s available, which gets interpreted as mindless moderation borne of ignorance and shallow beliefs. Americans appear schizophrenic to academics on many policy issues because they’re being asked the wrong questions.
Fortunately, some academics have invested in the more painstaking work of asking Americans what they think about government, public policies, and the common good, and recording what respondents have to say in their own words. The findings of these scholars offer a sharp — and encouraging — contrast to the work of pollsters. While survey researchers paint a picture of Americans as ignorant and indifferent, scholars who take the time to actually talk with the subjects of these surveys describe people who sound like they’re capable of — and willing to be — the kinds of citizens the American Founders envisioned.
– – –
Tony Woodlief is Executive Vice President at State Policy Network, a nationwide community that cultivates and supports state-based organizations working on behalf of citizen freedom and self-determination.
Photo “People Voting” by Phil Roeder. CC BY 2.0.
