Last week (FAIR.org, 2/2/22), I suggested that a new ABC News/Ipsos poll (1/30/22) was a poster child for what is wrong with many media-sponsored polls these days. Instead of a serious effort to measure what the public is thinking about any specific issue, the poll glided superficially across a whole range of subjects, never stopping long enough to provide understanding of any one of them—creating an illusion of public opinion that is either misleading, biased or simply inaccurate.
That article focused on the poll’s biased wording on one question about President Joe Biden’s promise to nominate a Black woman for the Supreme Court vacancy. In this article, I examine the nine presidential approval questions the poll included.
Origin Of Presidential Approval
In October 1938, George Gallup asked a presidential approval question for the first time, some three years after he launched his newspaper column “America Speaks.” The intent was to measure a president’s political strength between elections.
For the past seven-plus decades, Gallup has continued to ask the general approval question: “Do you approve or disapprove of the way [president’s first and last name] is handling his job as president?” Other polling organizations these days routinely include this question, or a minor variant of it, in their own polls.
The question appears to capture the president’s and his party’s overall popularity with the public. According to Frank Newport and Lydia Saad, writing in Public Opinion Quarterly (Spring/21), the general approval question correlates moderately with how well a president does when running for re-election, and how well his party fares in midterm elections.
That’s why numerous analysts (see here, here and here) have suggested that if Biden’s low approval rating persists into the fall, the midterms this year could be disastrous for the Democrats.
While the general presidential approval rating provides some relevant insight into public opinion, the same cannot be said for presidential approval questions that focus on individual issues. The January ABC/Ipsos poll included nine such questions:
If we take the questions and answers at face value, the results present an amazing picture of the American public. Here are nine very different, and quite complicated issues. And for each one, the respondents are asked to assess how well Biden is dealing with each issue.
The numbers indicate that 97% or more of American adults know what Biden is doing for each issue. And, presumably, after careful thought, they have come to a meaningful opinion as to whether they approve or disapprove of his actions.
This is the media-polling Myth of the American Electorate—fully informed and attentive to all issues, with well-considered views about the way they should be addressed.
That myth, of course, bears little resemblance to reality. In fact, on most issues, the vast majority of the public is simply uninformed—about the issue, and especially about what any president is doing to address the issue.*
The Illusion Of Public Opinion
Media polls get virtually all respondents to answer a question, regardless of whether they actually have an opinion, by the way in which questions are phrased. They often use a “forced-choice” format, which provides explicit answers like “approve” or “disapprove,” but fails to provide an explicit “unsure” or “don’t know” option. Respondents can volunteer a “no opinion” response, but the vast majority feel pressured to come up with one of the explicit answers, and—as the ABC/Ipsos poll reveals—that pressure works.
Since most people really don’t know much about how the president is addressing each of the issues, their forced-choice responses can be based on a variety of factors. For some respondents, the mention of Biden’s name will cause them automatically to “approve”; for others, it will cause them automatically to “disapprove.”
Some respondents will be influenced by what vague news stories they’ve recently read or heard. In that case, they’re reacting not to what the president is actually doing, but to whether they perceive the topic itself (such as the economic recovery, or the situation with Russia and the Ukraine) as good or bad.
Some issues are inherently “bad”—like crime, gun violence and inflation—and, except for the most devoted partisans, most people will “disapprove” of any president’s “handling” of those issues.
The ABC/Ipsos results thus tell us almost nothing important about public opinion on any of those issues. The news organization could have focused on one or another topic, and asked a variety of questions to probe what Americans are thinking.
Instead, we get a whole lot of soundbites, and only the illusion of public opinion.
Besides the article referenced in this paragraph, numerous studies over the years have demonstrated that large segments of the public are unengaged on any given issue. Among the many studies that explicitly address this issue are Daniel Yankelovich, Coming to Public Judgment: Making Democracy Work in a Complex World (Syracuse University Press, 1991; George Bishop, The Illusion of Public Opinion: Fact and Artifact in American Public Opinion Polls (Rowman & Littlefield, 2004); David W. Moore, The Opinion Makers: An Insider Exposes the Truth Behind the Polls (Beacon Press, 2009); Christopher H. Achen & Larry Bartels, Democracy for Realists: Why Elections Do Not Produce Responsive Government (Princeton University Press, 2017).