A recent Economist / YouGov Poll found that Republicans (76%) are more likely to say the U.S. should help its allies when asked than are Democrats (57%) or Independents (51%). It also found that MAGA Republicans are more likely than non-MAGA Republicans to say the U.S. should help allies (81% vs. 66%). This difference was surprising, because in some other contexts, Republicans and especially those who are MAGA supporters have been less likely to support U.S. aid for allies. For example, both groups are more likely to favor reducing military aid to Ukraine.

We considered two possible explanations for this unexpected pattern. First, immediately before the question about the U.S. helping allies on the recent Economist / YouGov Poll, the survey asked whether U.S. allies should agree to help the U.S. when asked. It found that Republicans and especially MAGA supporters are more likely to say allies should agree to help the U.S. Republicans — and particularly MAGA Republicans — may be more likely, in the spirit of reciprocity, to say the U.S. should help its allies because so many of them had just said they would expect the same in return.

Second, the Economist / YouGov Poll included many questions about the war in Iran before asking about the U.S. helping allies. That context may have made respondents more likely to think about allies that Republicans are more inclined than Democrats to support — such as Israel and the Gulf states — and less likely to think about countries such as Ukraine that Republicans are less inclined than Democrats to support. To test these hypotheses, we ran two survey experiments.

Experiment 1: Reciprocity

In the first experiment, we randomized the order in which we asked two questions about allies helping the U.S. and the U.S. helping allies. We randomly selected half of respondents to answer the questions in the same order in which they appeared on the original Economist / YouGov Poll. First these respondents answered the question, "If the U.S. asked for help from its allies, should they agree to help?" and then they answered the question, "If U.S. allies asked for help from the U.S., should it agree to help?" The other half of respondents answered these two questions in the opposite order. If respondents are motivated by reciprocity, we would expect them to be more likely to support the U.S. helping allies if they were first asked a question about allies helping the U.S. However, if they are not motivated by reciprocity, the question order should not make a difference.

Results of this experiment support the hypothesis that respondents are motivated by reciprocity and that this motivation particularly affects MAGA Republicans.

Americans are 7 percentage points more likely to say that the U.S. should help allies after answering a question about the U.S. receiving help compared to when the question about the U.S. receiving help comes later (57% vs. 51%).

This effect varies significantly by party identification. Republicans experience the largest effect: They are 13 points more likely to say the U.S. should help allies when they are first asked about the U.S. receiving help (64% vs. 52%). The effect is 10 points among Independents (50% vs. 41%). There is not evidence of such an effect among Democrats: They are 2 percentage points less likely to say the U.S. should help its allies after being asked about the U.S. receiving help (59% vs. 61%), a difference that is within the margin of error for this survey. Republicans are less likely than Democrats to say the U.S. should agree to help allies when they are asked this question first (52% vs. 61%), but slightly more likely than Democrats to say the U.S. should help allies when asked after a question about the U.S. receiving help from allies (64% vs. 59%).

The effect of reciprocity is particularly pronounced among Republicans who say they are MAGA supporters. MAGA Republicans are 21 points more likely to say the U.S. should help allies when they are first asked about allies helping the U.S. (74% vs. 53%). With the order of questions that emphasizes reciprocity with alliances, MAGA Republicans are much more likely than the average Americans to say the U.S. should agree to help allies, but when asked without a preceding question, MAGA Republicans' responses are not significantly different from those of the average American.

Experiment 2: Context

In the second experiment, we randomized the order in which three sets of questions were asked. A randomly selected half of respondents were first asked questions about the war in Iran, then questions about allies and the U.S. helping each other, and finally questions about the war in Ukraine. The other half of respondents were first asked questions about Ukraine, then about the U.S. and allies, and finally about Iran. The first group's order of questions is more similar to the Economist / YouGov survey, which included questions about the war in Iran before questions about the U.S. and allies. We hypothesized that some respondents may be more motivated to support the allies involved in the war in Iran — such as Israel or the Gulf states — than to support allies involved in the war in Ukraine — such as Ukraine itself or European allies. If this is the case, we would expect more of those respondents to say the U.S. should agree to help allies if they are in the first group, in which respondents were recently thinking about the war in Iran, and less of those in the other group to say the U.S. should agree to help allies, when the war in Ukraine was fresh in their minds.

Results of the second experiment do not provide strong support for the hypothesis that the context of the war in Iran drove more respondents to support the U.S. helping allies — among Americans overall or among Republicans.

Americans are only 3 points more likely to say the U.S. should agree to help allies after answering questions about the war in Iran compared to after answering questions about the war in Ukraine (66% vs. 63%). This difference is within the margin of error for this survey, meaning we do not have strong evidence of a true difference.

There was a slightly larger effect among Independents (4 points) and Republicans (2 points) than Democrats (0 points), but all of these differences are within the margin of error. There also was not a statistically significant difference among MAGA Republicans or non-MAGA Republicans.

The results of these experiments suggest that certain groups of Americans are more motivated by a sense of reciprocity in foreign relations than others. In particular, MAGA Republicans are much more supportive of the U.S. offering support to allies when they have previously answered a question about allies helping the U.S. This suggests that we must take into account context from surrounding survey questions in order to fully understand Americans' beliefs about U.S. alliances. On the other hand, these experiments do not find strong evidence that Americans answer questions about U.S. alliances differently when they are asked in the context of different conflicts. While some Americans seem to view the U.S.'s commitments to its allies as conditional on reciprocity, we did not find evidence that Americans favor allies in the context of some conflicts more than in others.

— Carl Bialik contributed to this article

Related articles:

Methodology: Two YouGov polls were conducted online in March 2026. The first was among 1,098 U.S. adult citizens on March 24 - 27 and the second was among 1,106 U.S. adult citizens on March 25 - 27. Respondents were selected from YouGov’s opt-in panel to be representative of adult U.S. citizens. The sample was weighted according to gender, age, race, education, region, 2024 presidential vote, 2020 election turnout and presidential vote, baseline party identification, and current voter registration status. Demographic weighting targets come from the 2019 American Community Survey. 2024 presidential vote, at time of weighting, was estimated to be 48% Harris and 50% Trump. Baseline party identification is the respondent’s most recent answer given around November 8, 2024, and is weighted to the estimated distribution at that time (31% Democratic, 33% Republican). The margin of error for each overall sample is approximately 4 points.

See the results of the March 24 - 27, 2026 poll and the March 25 - 27, 2026 poll

Image: Getty (Win McNamee / Staff)

What do you really think about President Trump, American politics in general, and everything else? Share your reality, join the YouGov panel, and get paid to share your thoughts. Sign up here.

Subscribe to the YouGov newsletter