Recent YouGov polls on survey interpretation show that most Americans consider 55% support for a policy in a poll “a majority” but that a policy needs at least 90% support before a majority sees the result as an “overwhelming majority.” But issue polls are used for more than measuring overall support for a policy. Pollsters and journalists often use surveys to describe how different groups — often, Democrats and Republicans — feel about a policy and determine whether the groups agree or disagree on the policy. Lots of polling coverage — especially during election years — also revolves around surveys that ask which political candidate in a given race Americans would vote for if an election were held today.
New YouGov polling asks Americans how they interpret polling results such as these. When a poll shows that 55% of people in one group support a policy and 45% of people in another group support that policy, most Americans (57%) think the groups “are divided on the policy.” If a poll shows that a policy has 90% support in one group and 80% support in another, Americans without college degrees are more likely than those with college degrees to think the groups disagree on the policy. When it comes to election polls where one political candidate has 46% support and their opponent has 42% support, most Americans think the first candidate has a slight lead but not a large lead. Republicans are likelier than Democrats and Independents to think a political candidate with a lead in a poll has the better chance to win the election.
Policy support among two groups
Are two groups united or divided about a policy? Americans, appropriately, often are divided on that question when presented hypothetical poll results. YouGov presented respondents with polls showing that a policy has majority support among one group but not another — the poll didn't specify whether these two groups were Democrats and Republicans, or any other characteristics of the groups. These hypothetical surveys showed that 55%, 70%, or 90% of people in one group supported a policy — as did 10%, 30%, and 45% of people in a second group Respondents then were asked whether the groups “are divided on the policy,” “disagree on the policy,” “agree on the policy,” or “are united on the policy.”
When a poll shows that 55% of people in one group support a policy and 45% of people in another group support the policy, most Americans (57%) think the groups “are divided on the policy” and about half (46%) think they disagree on the policy. The survey made it possible to characterize the same survey results in more than one way. In this scenario, 27% of Americans think the groups agree on the policy. When a survey instead shows 70% support for a policy in one group vs. 30% support in another group, Americans are much less likely to think the two groups agree (13%).
While in general, Americans are more likely to say two groups disagree or are divided over a policy as the difference in policy support between the two groups increases, even at big differences in support not all see disagreement. 67% of Americans identify 90% vs. 10% support in a poll as disagreement, compared to 46% of Americans for 55% vs. 45% support. However, even in a survey where 90% of people in one group and only 10% in another group support a policy, 22% of Americans say they do not think the groups disagree on the policy; 11% are unsure.
Three other policy support questions presented scenarios in which majorities of two groups support a policy, but one group is more supportive than the other. In these hypothetical surveys, 90% of people in the first group support the policy, and 50%, 65%, and 80% of people in the second group support it.
When a survey shows that 90% of people in one group and 50% of people in another group support a policy, 52% of Americans think the groups disagree. Only 28% think they agree on the policy in this scenario. When support in the second group is 65% and support in the first group remains 90%, about as many Americans say the groups agree on the policy (44%) as say they disagree (38%). In this situation, Americans are more likely to think the groups are divided than to think they are united. However, if a survey shows 90% and 80% support among two groups, most Americans think the groups agree on the policy (70%), and a smaller majority see the groups as united (61%). Even in this situation, 19% of Americans think the two groups disagree on the policy.
Americans with college degrees are more likely than those without college degrees to think a survey where 90% of one group and 80% of another group support a policy represents agreement and less likely to think it represents disagreement. 80% of Americans with college degrees think the two groups agree on the policy in the 90% vs. 80% scenario, compared to only 64% of Americans without college degrees.
Only 15% of Americans with college degrees think the groups are divided in the 90% vs. 80% scenario and 13% think they disagree. On the other hand, 28% of Americans without college degrees see the two groups as divided and 23% think they disagree. Americans with college degrees are also less likely than those without college degrees to be unsure about whether each characterization of the results polled about applies to the situation.
Election polling
YouGov also asked about election polls showing support for two hypothetical political candidates in an upcoming election. There were three scenarios. In each one, Candidate A was described as having more support. The first showed the two candidates with nearly identical support: 44% support for candidate A compared to 43% support for candidate B. The second showed 46% vs. 42% support — a larger difference, but still within the margin of error for most public polls. The final scenario showed 52% vs. 36% support, a clear advantage for candidate A. To figure out how Americans interpret these results, the poll which of the following statements apply to each scenario: “Candidates A and B have about equal support,” “it’s hard to tell which candidate is more likely to win the election,” “Candidate A has a slight lead,” “Candidate A has a large lead," and “Candidate A is more likely than Candidate B to win the election.”
80% of Americans think the two candidates have “about equal support” when a survey shows Candidate A has 44% support and Candidate B has 43%. A smaller majority (65%) think so when a survey shows 46% vs. 42% support. Only 5% think the candidates have equal support when 52% of people support candidate A and 36% support candidate B. Similarly, large majorities think “it’s hard to tell which candidate is more likely to win the election” in the 44% vs. 43% and 46% vs. 42% scenarios. However, only 20% think so when a poll shows 52% vs. 36% support for the two candidates.
A majority (58%) of Americans think candidate A has a slight lead with 44% support to Candidate B's 43%, and a much larger majority (76%) see 46% vs. 42% support as a slight lead for Candidate A. Only 39% see 52% vs. 36% support as a slight lead, but that’s because Americans are much more likely to think Candidate A has a large lead (64%) in that scenario. Very few Americans think Candidate A has a large lead in either of the other two scenarios.
When Candidates A and B have 44% and 43% support, respectively, only 15% of Americans think Candidate A is more likely than Candidate B to win the election. In the scenario of 46% and 42% support, Americans are about as likely to say Candidate A is (39%) or is not (36%) more likely than Candidate B to win the election. Most Americans (77%) think Candidate A is more likely to win in the 52% vs. 36% scenario.
Republicans are more likely than Democrats and Independents to characterize Candidate A's lead as slight when it's 44% to 43% or 46% to 42%. For example, when Candidate A leads by 44% to 43%, 69% of Republicans think Candidate A has a slight lead, compared to 56% of Democrats and 51% of Independents. Democrats are more likely than Republicans to call these leads large, though a small percentage of Democrats say this. Republicans are more likely to think Candidate A has a large lead when the lead is 52% to 36% (77%) than are Democrats (64%) and Independents (53%).
When it comes to whether Candidate A is more likely than Candidate B to win the election, Democrats (18%) and Republicans (17%) are more likely than Independents (11%) to think so when Candidate A's lead is 44% to 43%. In the other two scenarios, Republicans are likelier to think Candidate A has a better chance of winning. For example, when the two candidates have 52% support and 36% support, 87% of Republicans think Candidate A is more likely to win, compared to 75% of Democrats and 72% of Independents.
— Taylor Orth and Carl Bialik contributed to this article
Related:
See the results for this YouGov poll
Methodology: This poll was conducted online on April 10 - 18, 2024 among 1,151 U.S. adult citizens. Respondents were selected from YouGov’s opt-in panel using sample matching. A random sample (stratified by gender, age, race, education, geographic region, and voter registration) was selected from the 2019 American Community Survey. The sample was weighted according to gender, age, race, education, 2020 election turnout and presidential vote, baseline party identification, and current voter registration status. Demographic weighting targets come from the 2019 American Community Survey. Baseline party identification is the respondent’s most recent answer given prior to November 1, 2022, and is weighted to the estimated distribution at that time (33% Democratic, 31% Republican). The margin of error for the overall sample is approximately 4%.
Image: Getty (Andrii Yalanskyi)
