Both Opposition to and Support for Ukraine Aid May Be Less Than Polls Show

Last month, the Biden administration requested an additional $24 billion to aid Ukraine in its war with Russia. Some Republican leaders are skeptical or outright opposed to new funding, prompting Senate Minority Leader Mitch McConnell to urge his fellow legislators, “It’s certainly not the time to go wobbly.” That sentiment, of course, was reinforced by President Joe Biden during Ukraine President Volodymyr Zelenskyy’s recent visit to the United States.

At first glance, however, support among Republican voters appears to be wobbly already. Late last month, Daily Kos (8/31/23) headlined a story that noted declining support among Republican voters for supporting Ukraine: “McConnell Abandoned by Post-Trump Republican Electorate.” And three recent polls suggest that rank-and-file Republicans are indeed negative toward aid to Ukraine.

But all three polls wildly overstate how engaged Americans, including Republicans, are in this issue. Opposition, as well as support, is probably far lower than what the media tell us.

Polls report GOP opposition

The most recent poll by CBS/YouGov (9/10/23) finds support for aid to Ukraine among Americans overall, but a decline in support among Republicans since last February.

Overall, 64% of Americans are positive about support for Ukraine—saying the Biden administration is either “handling things as they should be” (38%) or should be doing more (26%). Only 36% say it should be doing less. Among Republicans, 56% say the administration should be doing less.

An earlier poll by Fox (8/17/23) reports similar figures. Overall, 61% of registered voters have positive views about US support for Ukraine—40% who believe the US is giving the right amount of aid, and another 21% who want the US to do even more. Just 36% say the US should be doing less. Among Republicans, 56% believe the US should be doing less, the same figure CBS found.

The most negative results about aid to Ukraine are found in last month’s CNN poll (8/4/23), which reported that a majority of Americans overall believe the US has “done enough to assist Ukraine” (51%) and “should not authorize additional funding to support Ukraine in its war with Russia” (55%). Among Republicans, 59% say the US has done enough, and 71% are opposed to additional funding.

Wording makes a difference

When CNN (8/4/23) asks if the US “should do more to stop” Russia, do respondents think that means continuing aid or increasing aid?

So all three polls report a majority of Republicans opposed to additional funding for Ukraine. But two of the polls, by CBS and Fox, find a net positive view of aid to Ukraine among Americans overall, while only CNN finds majority opposition.

The difference between CNN‘s and the other two polls is largely because of CNN’s tendentious wording:

CBS: Do you think the Biden administration should be doing more to help Ukraine in its conflict with Russia, should it be doing less, or is it handling things about as they should be?

Fox: Do you think the United States should be doing more to help Ukraine in its war with Russia, should be doing less, or is the US doing about the right amount to help Ukraine?

CNN: Do you think the United States should do more to stop Russian military actions in Ukraine, or has it already done enough?

(Note: Both CBS and CNN randomly rotated their response options.)

The CNN question gives just two options, compared with three in the other two polls. By itself, that is not a problem. What makes that question tendentious is that it provides a reason not to do more for Ukraine (because the US has “already done enough”), but provides no reason to do more (like, say, “the Russians refuse to stop their aggression”).

Also, the question is somewhat ambiguous: What does it mean for the US to do “more”? Does CNN mean more than the US has been doing, or does it mean to continue to provide aid at the same level? The other two polls make the issue clear—“more” means more than the US is doing now, because the middle option in those two polls (“doing the right amount” and “handling things as they should be,” respectively) essentially says the US should continue providing aid at the level it is currently doing. (The US has given Ukraine $77 billion so far over a year and a half of war, though it’s unclear how many respondents are aware of that.)

Given the problems with the CNN question wording, I’m inclined to discount its results in favor of the other two polls.

An idealized public

Still, even the other two polls have credibility problems. All three describe an idealized citizenry that is utterly at odds with reality. CBS suggests that 100% of Americans/voters have an opinion about the level of aid the US/Biden administration is providing Ukraine. For CNN, the comparable number is 99%. For Fox, 97%.

Such high responsiveness reinforces what two researchers have called the “folklore theory of democracy.” This notion of democracy posits that the vast majority of voters are well-informed and engaged on policy issues, so that when election time comes, they can make a sound judgment as to how well their elected leaders reflect the will of the people.

The reality, of course, is far different. As those authors make clear, the political science literature is replete with studies that describe widespread public ignorance of policy issues, as well as a lack of basic knowledge about the American government.

The illusion of public opinion

So, how did the three polls show virtually all Americans with an opinion on aid to Ukraine? Two major techniques.

First, they ask “forced-choice” questions, which give respondents positive and negative options to choose from, but do not provide an explicit “unsure” or “don’t know” option. Respondents feel obligated to give some answer, regardless of whether they have actually developed any opinion about it.

Second, the respondents are all “performing” for the interviewers. There is an implicit understanding that the respondents are there to answer questions. That is their “job.” If they didn’t want to answer questions, they wouldn’t be taking the poll. If the interviewer (or if the electronic form that respondents fill out online) explicitly offers the option of “no opinion,” then the respondent would feel free to choose that option. But with the forced-choice questions, respondents understand that they are expected to provide an answer.

CNN actually follows up volunteered “no opinion” responses by asking respondents if they “lean” toward one option or the other, thus ensuring they get close to 100% responses.

Unreliable results from unengaged citizens

Seventy-six percent of the respondents whose opinions Pew (6/15/23) cites say they are not paying “very” close attention to the Ukraine War.

How reliable are responses from people who are relatively uninformed? Again, political science research has long answered that question, and the answer is—not very. As one researcher explains:

The consequences of asking uninformed people to state opinions on topics to which they have given little, if any, previous thought are quite predictable: Their opinion statements give every indication of being rough and superficial…. [They] vacillate randomly across repeated interviews of the same people.

How many people are “uninformed”? That’s a bit tricky to measure, because it’s not a simple matter of informed vs. uninformed. People have varying degrees of knowledge. Pollsters avoid the problem by mostly ignoring it. But now and then, pollsters do try to measure how much people know about a given issue.

Last June, for example, a Reuters/Ipsos poll (6/28/23) reported that only 18% of Americans were following stories about the Russian invasion of Ukraine “very closely.” Another 39% said “somewhat closely,” leaving 43% saying not closely (or they didn’t know).

An earlier poll by Pew (6/15/23) also found few people paying particular attention to the war in Ukraine: 9% saying extremely closely and 15% very closely. Another 35% said somewhat closely. Again, 42% said not too, or not at all, closely (or they didn’t know).

Of course, people with little to no knowledge on an issue can still express an opinion about it, and sometimes even feel strongly about it—probably because they see the issue linked to something else they do feel strongly about, like party identification, or perhaps a political leader with whom they closely identify.

Still, if the poll question provides respondents with an explicit “don’t know” option, people who don’t know much about an issue will often choose that. And respondents who express an opinion, but don’t really care one way or the other, are likely to admit it if asked.

Few with strong feelings 

We can see this dynamic in a Pew poll last June (6/15/23), which—unlike the three polls described earlier—explicitly provided respondents with a “not sure” option. The result: Overall, 24% chose “not sure,” and another 1% did not respond.

Even that level of participation—75% expressing an opinion—may overstate the public’s level of engagement. It could reflect the “job” that respondents have taken on, to answer poll questions, regardless of how much they’ve really thought about the issue.

Evidence for this idea is found in the question asked of Pew respondents immediately prior to the one about continued aid: “Do you approve or disapprove of the Biden administration’s response to Russia’s invasion of Ukraine?” Options allowed respondents to express intensity of opinion.

As the table makes clear, overall just 30% of the respondents express a “strong” opinion: 13% who approve, 17% who disapprove.

Another 44% express mild opinions: 26% approve, 18% disapprove. Another 26% have no opinion.

What to make of the respondents who “somewhat” approve or disapprove?

Andrew Smith and I presented a paper at the annual meeting of the American Association for Public Opinion Research in 2010, which included research showing that respondents who expressed mild opinions (characterizing their feelings as “not strongly” or “somewhat”) also said in a follow-up question that they would not be “upset” if their opinion did not prevail.

The conclusions we drew were that large numbers of respondents who express an opinion on a “forced-choice” question, like the ones in the CBS, Fox and CNN polls, are not really invested in their own responses. They are simply not engaged enough to care strongly one way or the other.

Using that criterion, the Pew poll suggests that overall, about 7 in 10 Americans are unengaged in the issue of US aid to Ukraine. Among Republicans, about 65%; among Democrats, 72%.

Among people who are engaged, Republicans are clearly quite negative, by a margin of 31% who strongly disapprove to 4% who strongly approve. Engaged Democrats are more positive: 23% strongly approve, while just 5% strongly disapprove.

Had the other three polls also provided an explicit “unsure” option, and then measured intensity of opinion, the percentage of Republicans who strongly disapprove would no doubt be considerably below a majority. By the same token, the percentage of Democrats who approve would also be considerably below a majority. Most people are simply unengaged in this issue.

Performative vs. realistic polls

As a general rule, news media are not fans of polls that reveal how disengaged the public is on most issues. They prefer what I call “performative polls,” because such polls give the illusion of an attentive and informed public that is consistent with our general conception of how US democracy should work.

More importantly, reporting on polls that regularly show large segments of the public unengaged on the issues would call into question the utility of conducting the polls in the first place. Perhaps the media should spend more effort to keep the public informed on current issues than on performative polls that do little to enlighten.