To Advance Racial Justice, We Must Mobilize The Base

By John Ray (@johnlray)

In this post we look at the 2016 and 2018 behavior of US voters, and compare some of their attitudes on race based on our What The Hell Happened project. While in previous posts we created full breakdowns by 2012-2016-2018 vote, here we simplify by focusing on who switched from 2016 to 2018 -- largely due to sample size concerns, but also for simplicity. In our survey (full dataset and codebook available here), including weighted counts, we had:

  • 1,343 voters who voted Democrat in both 2016 and 2018 (Dem-Dem)

  • 1,252 voters who voted Republican in both 2016 and 2018 (Rep-Rep)

  • 285 voters who didn’t vote or voted third party in 2016 and voted Democrat in 2018 (Other/NV-Dem)

  • 205 voters who didn’t vote or voted third party in 2016 and voted Republican in 2018 (Other/NV-Rep)

  • 89 voters who voters who voted Republican in 2016 and Democrat in 2018 (Rep-Dem)

Even considering only two cycles of data rather than three, we still don’t really have enough voters who flipped from Democrat to Republican from 2016 to 2018 (just 36, all told), so we’re going to exclude them for now.

This post focuses on certain measures of race and shows that voters who swung from the Republicans to the Democrats from 2016 to 2018 are much closer to the GOP on race issues than to the Democrats. This is important for Democrats to know because the Republicans will have an outspoken racist at the top of the ticket in 2020. We’ve proposed some policy-based ways forward here, but whatever the way forward is, Democrats must be prepared for a voting middle that is significantly worse on race than we might hope.

We included several measures in our survey that tap directly into racial animus, and several items that tap into so-called “cultural values” that we know correlate highly with views on race. One of those is views on the Confederate flag. Specifically, in our survey, we asked voters:

Which of the following is closer to your view regarding the Confederate flag?

<1> The Confederate flag is mostly a symbol of Southern heritage and culture

<2> The Confederate flag is mostly a symbol of slavery and white supremacy

<3> Don't know

Overall, voters were narrowly split on this: About 44 percent of respondents said it is mostly a symbol of slavery and white supremacy, 48 percent said it was a symbol of Southern heritage, and about 8 percent said they were unsure (For context, 8 percent is roughly normal for an item of this kind). But among voters who switched from Republicans to Democrats in 2018, their views were much closer to traditional Republicans: 72 percent of Rep ‘16 to Dem ‘18 voters felt it was a symbol of Southern heritage, which is statistically equivalent to the 62 percent of those who didn’t vote in 2016 or voted for a third party in 2016 to voted for a Republican in 2018, and slightly below the 90 percent of loyal Republicans who said it was a symbol of Southern heritage.

Democrats were the opposite: 81 percent of loyal Democrats recognize the Confederate flag as a symbol of slavery and white supremacy, as do 66 percent of those who didn’t vote in 2016 or voted for a third party in 2016 and voted Democrat in 2018. Those quantities are well outside the confidence intervals needed to assert that Democrats and non-voters or third party voters turned Democrats are different from Rep-Dem voters on the question of the Confederate flag.


One reason for this speaks for itself: Rep-Dem voters from 2016-2018 are significantly more white than loyal Dem and Other/NV-Dem voters, but also slightly less white than Other/NV-Rep voters and Rep-Rep voters. Fully 83 percent of Rep-Dem voters are white, compared to 67 percent of loyal Democrats and 90 percent of loyal Republicans.


Rep-Dem voters are much older than loyal Democrats and Other/NV-Democrats as well. The average Rep-Dem voter from 2016-2018 was about 58 years old, roughly the same age as the average loyal Republican. The average loyal Democrat was fully five years younger. Those who didn’t vote in 2016 or voted for a third party but then voted for the Democrats were younger still, as were non-voters/third party voters who voted Republican in 2018.


In addition to their views on the role of the Confederate flag in American history, Rep-Dem voters were closer to Republicans on measures of racial animus than they were to Democrats. We have explained our scale, based on current social science research on the subject, elsewhere, with details below.

On that scale, Rep 16-Dem 18 voters scored much worse than loyal Democrats and other/non-voters turned Democrats. Rep 16-Dem 18 voters averaged about 35 percent of the full racial animus scale, compared to just 12 percent for loyal Democrats and 48 percent for loyal Republicans. While still not as bad on race as Republican voters, those swing voters are plainly closer to Republicans on race issues.


There is no doubt that Trump’s re-election campaign is going to rely largely on racism. Absent any policy accomplishments except for a widely reviled tax cut and the least popular Supreme Court nominees in modern history, Trump’s general strategy will inevitably involve the same appeals to white supremacy as his campaign deployed in 2016 and 2018. Democrats should be prepared for what this means for their strategy going forward. We will have subsequent posts on this very subject in the coming weeks.

John Ray (@johnlray) is a Senior Political Analyst at YouGov Blue and a Senior Adviser to Data for Progress.

Between October 27 and November 7, 2018, YouGov Blue surveyed a nationally representative sample of 3,215 respondents. The overall survey margin of error was +/- 2 percent. The sample was weighted to the 2018 electorate.

In our survey, respondents were asked to “indicate the extent to which [they] agree with each of the following statements.” The items, presented below, were asked in random order.

  • Generations of slavery and discrimination have created conditions that make it difficult for African Americans to work their way out of the lower class.

  • I am angry that racism exists.

  • Irish, Italian, Jewish, and many other minorities overcame prejudice and worked their way up. Blacks should do the same without any special favors.

  • Racial problems in the U.S. are rare, isolated situations.

  • White people in the U.S. have certain advantages because of the color of their skin.

For each statement, respondents were asked whether they strongly agreed, somewhat agreed, neither agreed nor disagreed, somewhat disagreed, or strongly disagreed. For the “Slavery,” “Angry” and “White advantages” items, the “strongly disagree” position is the position associated with higher levels of racial animus, while the “strongly agree” position is associated with lower levels of racial animus. For the “Other minorities” and “Rare problems” items, the “strongly agree” position is the position associated with higher levels of racial animus. To create our scale, the “highest racial animus” position was given a score of 4, down to a 0 for the “lowest racial animus” position. For example, respondents who strongly agreed that “Blacks should do the same without any special favors” on the “Other minorities” item were given a 4, those who somewhat agreed a 3, those who neither agreed or disagreed a 2, those who somewhat disagreed a 1, and those who strongly disagreed a zero. The scale is a sum of these values for each item. For convenience of interpretation, we report the score as a percent below.

Subscribe to our mailing list

* indicates required