Who are “Trump Republicans?”

YouGov Blue
15 min readJan 25, 2021

--

Throughout the 2019–2020 election cycle, YouGov Blue fielded a regular survey of about 1,000 US registered voters every week. Each wave of this project, the YouGov Blue Core, included a set of common political, demographic, and policy-related items, as well as items relevant to the particular news going on around a wave’s fielding period.

One of the common items we included pertained to political identity. We asked voters,

People often use labels to identify their political beliefs. Which of the following do you identify with? Select all that apply.
1. Moderate
2. Progressive
3. Liberal
4. Socialist
5. Conservative
6. Libertarian
7. Swing Voter
8. Nationalist
9. Obama Democrat
10. Democrat
11. Republican
12. Trump Republican
13. Other (please specify):

Voters could choose as many labels as they wanted, and labels were shown in a random order.

Here, we focus on Republican responses from the height of the campaign — from the end of June through election day, plus some supplemental data we ran in November 2020 and in January of 2021. This includes data on about 12,000 Republican voters, about 5,000 of whom identify as Trump Republicans. Overall, Trump Republicans comprise about 18 percent of the electorate overall, and about 42 percent of the Republican electorate.

We find:

  • Trump Republicans are significantly more ideologically conservative than the rest of the party, are more politically active, watch more Fox News, and score higher on measures of racial animus and hostile sexism
  • Trump Republicans and other Republicans use social media at about the same rates, but Trump Republicans are significantly more likely to do so to engage with politics and political causes
  • Trump Republicans have more intense dislike for Democrats than other Republicans, and are significantly more negative toward Republicans Trump has attacked in the past, like John Roberts and Mitt Romney
  • Trump Republicans are older than, but are not otherwise demographically distinct from, other Republican voters
  • Trump Republicans hold similar policy attitudes to other Republicans on most issues that we surveyed
  • The share of the party that identified as a “Trump Republican” was stable over time at about 42 percent of the Republican Party, starting to tick downward only once it was clear Trump had lost the election

Throughout this analysis, we will be referring to “Trump Republicans” and also to Republicans who did not choose the “Trump Republican” label. For simplicity and brevity, we will use the acronym “TR” to refer to “Trump Republicans” and “NTR” to refer to “Non-Trump Republicans,” which means Republicans who said they preferred ideological labels besides “Trump Republican.” In our sample, TRs represent about 42 percent of Republicans, and NTRs represent about 58 percent of Republicans.

Trump Republicans are more ideologically extreme, and watch more Fox News

The biggest differences between Trump Republicans and non-Trump Republicans show up when it comes to the ideological preferences and media consumption habits of each group.

Trump Republicans are about as likely to identify as ideologically “conservative” as other Republicans, but are much more likely to identify as “very conservative” than the rest of the party. Trump Republicans are only about half as likely to identify as “moderate” as the rest. About 13 percent of TRs compared to 25 percent of NTRs identify as “moderate.” Fully 43 percent of Trump Republicans identify as “very conservative” compared to 27 percent of the rest of the party.

On the subject of political ideology, our data suggest Trump Republicans are significantly more conservative than the rest of the party. On the subject of political labels, we let respondents select multiple options. There, we found that Trump Republicans were more likely to select additional political labels associated with political conservatism than other Republicans.

Perhaps not surprisingly, Trump Republicans and NTRs were about as likely to select the label “conservative.” The statistically and substantively significant differences emerge among the other labels popular among Republicans. About 18 percent of NTRs chose the “moderate” label, compared to just 8 percent of Trump Republicans.

Trump Republicans were about three times as likely as the rest to identify as a “Nationalist.” We did not ask respondents to define these labels, only to select them. Whatever Trump Republicans think “nationalism” means, they are more likely to take on that label than are other Republicans.

Later in the survey, we asked a variety of items designed to measure hostile sexism and racial animus. These items come from canonical academic measures of these constructs, and are enumerated at the end of this article. From responses to these batteries, we constructed scales of hostile sexism and racial animus, where values of 0 represented the lowest level of sexism or racism and values of 100 represented the highest level of sexism or racism.

Trump Republican voters reported higher levels of racism and sexism than did other Republicans. On every wave of our Core survey, we included canonical academic measures of hostile sexism and racial animus. On average, Trump Republicans were about 10 percentage points higher than NTRs on both scales. The average Trump Republican had about a 48/100 on the hostile sexism scale and about an 80/100 on the racial animus scale. The average non-Trump Republican had about a 38/100 on the hostile sexism scale and a 69/100 on the racial animus scale.

Perhaps not surprisingly, Trump Republicans are more likely to report Fox News is a primary news source than are other Republicans. About 49 percent of NTRs say Fox News is a primary news source for them, compared to about 66 percent of TRs. This result holds even accounting for Republicans’ demographic and political differences across the TR/NTR divide. Trump Republicans are significantly more likely to report that Fox News is a primary news source for them than are Republicans who don’t identify as Trump Republicans.

Notably, there is a substantively small but statistically noticeable interaction effect between “Trump Republican” status and age when it comes to Fox News viewership. While among non-Trump Republicans Fox viewership tends to drop off with age in a more or less linear fashion, younger Trump Republicans watch more than their peers one generation older. Trump Republicans aged 18–29 watch more Fox News than Republicans in the 30–44 age bracket. Both groups are pretty small — about 12 percent of Trump Republicans are aged 18–29, along with 17 percent of non-Trump Republicans.

Trump Republicans use social media at the same rate as other Republicans, but use it for politics much more

We asked voters a variety of items about their social media and news consumption habits. We asked about social media usage in a few different ways. One, we asked whether people used various social media platforms. Two, we asked if they used those platforms as news sources. On that item, we included additional news sources like the major cable news networks, newspapers, podcasts, and so on. Three, we asked if they used those platforms to talk about politics, as part of a question on how people participate in politics.

In the first case, we did not find major differences within Republican voters, whether they identified as a “Trump Republican” or not. Similar shares of Trump Republicans and non-Trump Republicans reported using various major social media platforms (Facebook, Twitter, and reddit), and using those platforms for news. About 76 percent of TRs and 75 percent of NTRs said they used Facebook. About 27 percent of TRs and 24 percent of NTRs said they used Twitter. About 7 percent of TRs said they used reddit and about 8 percent of the rest said they used reddit at all.

About 39 percent of TRs said they used Facebook as a primary news source, compared to 33 percent of NTRs. We find similar trends for other social media platforms. Similar shares of TRs and NTRs reported using Twitter as a news source — about 16 percent for TRs and about 13 percent for NTRs. About 4 percent of TRs and 5 percent of TRs reported using reddit as a news source.

Though Trump Republicans and other Republicans use social media at the same rate, Trump Republicans report they use social media for political purposes to a significantly higher degree. We asked respondents a battery of items about their participation in politics. We asked,

Some people participate in politics in various ways other than voting, while others do not. Which of the following, if any, have you done in the last 6 months? (Check all that apply)
1. Volunteered for a candidate, political party, or other political organization
2. Attended a rally or protest
3. Called or wrote to an elected official
4. Attended a town hall held by an elected official
5. Posted about politics on social media
6. Made a donation to a candidate, party, or other political organization
7. Other
8. I have not done any of the above

The largest gap we found among TR/NTR Republicans related to posting about politics on social media. Trump Republicans were fully 15 percentage points more likely to report posting about politics on social media, with 42 percent saying they had done so recently compared to 27 percent of non-Trump Republicans. While Republicans across the various ideological labels tend to use social media to a similar degree, Trump Republicans stand out in how much they use social media for politics. About 42 percent of Trump Republicans reported they had posted about politics on social media recently, compared to 27 percent of the rest of the party.

Indeed, across measures of political activism, we find that Trump Republicans are at least slightly more active than non-Trump Republicans on each of these items. Trump Republicans provide statistically and substantively significantly more donations, phone calls and letters, and social media posts than do other kinds of Republicans.

Trump Republicans intensely dislike Democrats and progressive groups — and even some Republicans

In each wave of our survey, we asked voters how much they favored or disfavored a variety of politicians and political groups. Here, we summarize some key differences and similarities between Trump Republicans and non-Trump Republicans in who in politics they favor and disfavor.

Even among those who don’t explicitly adopt the “Trump Republican” label, Trump is overwhelmingly popular. His net favorability among non-Trump Republicans is +70 percentage points, up to +96 percentage points among Trump Republicans. In contrast, President Joe Biden’s net favorability is -75 percentage points among NTRs and -90 percentage points among TRs.

Notably, Trump Republicans hold strongly less favorable views toward two Republicans Trump has attacked in the past: Supreme Court Chief Justice John Roberts and Utah Senator Mitt Romney. At the same time, Trump Republicans are strongly more supportive of Supreme Court Justice Amy Coney Barrett, House Minority Leader Kevin McCarthy, and Senate Minority Leader Mitch McConnell.

The following chart summarizes the differences in net favorability between Trump Republicans and the rest of the party. Points to the right of zero on this chart represent groups or politicians who are more popular with Trump Republicans than NTRs, and points to the left of zero on this chart represent those less popular among TRs than NTRs.

Not surprisingly, for example, Trump and former Vice President Mike Pence are significantly more popular among Trump Republicans. Trump Republicans also hold significantly more negative views of groups like Black Lives Matter and Representative Alexandria Ocasio-Cortez than do other Republicans (though both are clearly unpopular among Republicans overall). At the same time, some Republicans like Mitt Romney and John Roberts are conspicuously less popular among Trump Republicans than among the rest of the party.

Trump Republicans are older than other Republicans but are otherwise similar demographically

Demographically, Trump Republicans look similar to the rest of their party. They have slightly fewer college degrees (27 percent of TRs compared to 34 percent of NTRs), but this result is explained by the fact that Trump Republicans are older than the rest of the party. About 35 percent of TRs are over the age of 65, compared to 25 percent of the rest of their party.

Trump Republicans are no wealthier or less so than other Republicans. Across the party, there are roughly the same number of Trump Republicans and not Trump Republicans in each income bracket, as the graph below shows. There is not an economic class divide in the Republican Party when it comes to Donald Trump.

Similarly, there is not a geographic divide between Trump Republicans and the rest of the party. Roughly equal shares of TRs and NTRs identify as living in urban areas (about 19 percent each), suburban areas (low fifties percent in both cases), and rural areas (high 20s percent in each case).

Accounting for other factors, the biggest demographic difference between Trump Republicans and the rest of the party is age. Trump Republicans are, on average, about 5 years older than the rest of the party. This age difference persists even accounting for a variety of other factors, and no matter how one slices the data, it is difficult to escape the conclusion that Trump Republicans are slightly but significantly older than the rest of the party.

Trump Republicans are mostly not distinct on policy issues

Despite being significantly more politically active and more ideologically extreme than other kinds of Republicans, when it comes to bread and butter conservative issues, Trump Republicans do not stand out. On subjects like abortion, gun control, and even the US-Mexico border wall, we find only small differences between Republicans by ideological label.

On abortion, for example, we find that about 24 percent of Trump Republicans want abortion to be legal in at least some circumstances, compared to 30 percent of non-Trump Republicans and 28 percent of the party overall. About 76 percent of TRs and 70 percent of NTRs want abortion to be illegal in most or all circumstances.

Trump Republicans are slightly more extreme on the subject of gun ownership. We asked respondents,

Do you think gun control laws should be made more or less strict than they are now?
1. Guns should be banned completely
2. More strict
3. Kept the same
4. Less strict
5. There should be no restrictions on gun ownership
6. Not sure

About 13 percent of Trump Republicans want more strict gun control, compared to 22 percent of NTRs. About 22 percent of TRs want unlimited access to guns, compared to about 15 percent of the rest of the party. This marginal difference persists even when accounting for the relative age difference between these two subsets of the party: About 16 percent of NTRs over the age of sixty-five want unlimited access to gun ownership, compared to 21 percent of TRs over the age of sixty-five.

Not surprisingly, Trump supporters overwhelmingly favor building a US-Mexico border wall — but Republicans overall support this idea. The share of TRs who feel this way is statistically significantly higher but only marginally higher than the share of NTRs who feel the same way. Support for a US-Mexico border wall is overwhelmingly high across the party.

The share of Trump Republicans in the party was stable — until he lost

Finally, we note that throughout the election cycle itself, the share of Republicans who identified as Trump Republicans was stable. In various regression settings we did not find “wave-fixed effects,” i.e., we did not really see evidence that the share of the party identifying as Trump Republicans jumped or dipped much over the course of the election.

Since then, however, some of our early 2021 data suggests the share of Republicans who identify as Trump Republicans is going down. From a cycle-wide average of about 42 percent, the average in January (which, we caution, only accounts for about 1,000 of the Republicans overall and thus about 350 Trump Republicans) is about 35 percent.

This is notable because, when we tried to correlate rising or falling shares of Trump Republicans in the electorate with this or that scandal or notable political event during the election, we did not find much. The share of Republicans identifying as Trump Republicans did not, for example, rise around the GOP convention, or any of the debates. The number did not fall during any of the President’s innumerable scandals. As we move on to speculate what Trump will do with his post-presidency career, our data suggest that, to a certain extent, the brand may be out of his hands.

But our data also show that, independent of any sort of labels, Trump’s ideas are extremely popular among Republicans. The “Border Wall” remains an extremely popular idea among Republicans, for example. Whatever the future of the Republican party, it is hard to imagine it will be fully excising itself of Trump any time soon.

Appendix A: Hostile sexism and racial animus scales

Our racial animus scales incorporated the following items:

Please indicate the extent to which you agree with the following statement: Generations of slavery and discrimination have created conditions that make it difficult for African Americans to work their way out of the lower class.
1. Strongly agree
2. Somewhat agree
3. Neither agree nor disagree
4. Somewhat disagree
5. Strongly disagree
6. No opinion

Please indicate the extent to which you agree with the following statement: Irish, Italian, Jewish, and many other minorities overcame prejudice and worked their way up. Blacks should do the same without any special favors.
1. Strongly agree
2. Somewhat agree
3. Neither agree nor disagree
4. Somewhat disagree
5. Strongly disagree
6. No opinion

Please indicate the extent to which you agree with the following statement: White people in the U.S. have certain advantages because of the color of their skin.
1. Strongly agree
2. Somewhat agree
3. Neither agree nor disagree
4. Somewhat disagree
5. Strongly disagree
6. No opinion

On this scale, responses four or five in the first item, one or two in the second item, and four or five in the third item were given values of 1 on this scale, and 0 otherwise. The findings in this post are robust to recoding this scale to assign responses 3 and 6 to either the higher or lower ends of the racial animus scale. The findings in this post are robust to recoding this scale such that the “highest racial animus” value is given a score of 4 and the “lowest racial animus” value is given a score of 1 with intermediate values given a 2 and three, and so on.

Our hostile sexism scale incorporated the following items:

Please indicate the extent to which you agree with the following statement: Most women interpret innocent remarks or acts as being sexist.
1. Strongly agree
2. Somewhat agree
3. Neither agree nor disagree
4. Somewhat disagree
5. Strongly disagree
6. No opinion

Please indicate the extent to which you agree with the following statement. Most women fail to appreciate fully all that men do for them.
1. Strongly agree
2. Somewhat agree
3. Neither agree nor disagree
4. Somewhat disagree
5. Strongly disagree
6. No opinion

Please indicate the extent to which you agree with the following statement. Women are too easily offended.
1. Strongly agree
2. Somewhat agree
3. Neither agree nor disagree
4. Somewhat disagree
5. Strongly disagree
6. No opinion

On this scale, responses one and two in the first item, one or two in the second item, and four or five in the third item were given values of 1 on this scale, and 0 otherwise. The findings in this post are robust to recoding this scale to assign responses 3 and 6 to either the higher or lower ends of the racial animus scale. The findings in this post are robust to recoding this scale such that the “highest racial animus” value is given a score of 4 and the “lowest racial animus” value is given a score of 1 with intermediate values given a 2 and three, and so on.

--

--

No responses yet