Skip to content
ThreadOut
Crémieux@cremieuxrecueil · Sep 12, 2026

A few years back, a friend of mine looked into the 2018 American National Election Study pilot, and he found that…

10 tweets1 min read♥ 2.7Koriginal

Summary

Analysis of National Election Study data shows that white liberals are the only racial/ideological group that report lower warmth toward their own race compared to other groups, a phenomenon that emerged around 2012 during the 'Great Awokening' and correlates with liberal ideology but not conservative ideology or party affiliation. This self-directed racial coldness is historically unprecedented and doesn't follow a simple dose-response pattern with partisan strength.

Summarized by ThreadOut AI from the full thread. May miss nuance — read the thread below.

  1. #1

    A few years back, a friend of mine looked into the 2018 American National Election Study pilot, and he found that every racial group felt warmth towards members of their own race, except for White liberals I wanted to see if that replicated and expand on it🧵 #1: It replicates

  2. #2

    The next question is: 'OK, but isn't this just a liberal thing? Wouldn't we see this among other races if we split them by ideology, too?' #2: This self-hate phenomenon is specific to Whites.

  3. #3

    The next question is: 'But has it always been this way? Have liberal Whites always hated themselves?' #3: This phenomenon is timed to the Great Awokening, Obama's re-election.

  4. #4

    The next questions are: 'Was there a parallel for conservatives, a sort of anti-awokening?' and 'Is there an ideological dose-response?' #4: Not even a hint. Conservatives are basically as they've always been with respect to one another. #5: More liberal = more self-hating.

  5. #5

    The next question is: 'You've shown results for ideology, but does this follow with respect to party, too? That is, did White Democrats start to hate Whites?' #6: Yes, the results by ideology are transitive and also apply to the liberal party.

  6. #6

    The next question is: 'Do the results for party transfer in a dose-response fashion? That is, are stronger Democrats more self-hating?' #7: They don't. Look:

  7. #7

    The next question is a methodological one: 'Does this hold if we use the old Civil Rights-era weighting and linking datasets?' Followed by: 'Has this happened before?' #8: It holds. #9: Not in any recorded era.

  8. #8

    In my new article on this, I've asked and answered some additional questions, and if you go over there, you can submit more questions for me to look into. For example, I look into trends for other races, international results, replication in other cohorts, etc.

  9. #9

    Here's the link to the article: cremieux.xyz/p/whats-up-wit…

  10. #10

    OP should say *net* warmth. When are we getting the ability to edit just one post in a thread?