1/ I'm delighted to finally share some of the work I've done since joining @fsuigc.

Last month, we published a report about political tolerance based on national survey data (N=1,004) we collected during the summer. In this study, we measure political tolerance as the willingness to interact with or accept people with opposing political views across different relational contexts. For comparison, we also measured tolerance towards ex-felons and flat earthers. 2/ As shown in the table below, across all contexts, people are much more willing to engage with people with opposing political views than the other two target groups. Regardless of the target group, though, openness tends to decline as the intimacy of an engagement increases. For instance, whereas 73% would engage in a social/recreational activity with political opponents (ex-Felon: 54%, flat-earther: 49%) without reservation, just 41% would be willing to date them (ex-Felon: 22%, flat-earther: 19%)

1/ Updated racial ingroup vs. outgroup feeling thermometer differentials from the ANES. In sum, while the 'curve has flattened', the attitudinal effects of the Great Awokening persist (at least wrt race). If you thought or hoped otherwise, sorry to disappoint.

https://twitter.com/ZachG932/status/1375198311355969542

2/ Average of all 3 differentials by race/ideological self-placement

1/ Needless to say, the recent surge in New York Times articles pairing the word "genocide" with "Israel" is driven by the war in Gaza.

But it would be wrong to single out the Times.

The same pattern shows up across every other outlet I looked at.

https://x.com/ZachG932/status/1925376779742564459

2/ In fact, coverage linking Israel to “genocide” now exceeds that of every actual or widely recognized genocide of the last 40 years, including:

Rwanda (1994)

Darfur (2003–2008)

Bosnia (Srebrenica, 1995)

Myanmar (Rohingya, 2017–Present)

Yazidis (ISIS, 2014–2017)

🚨1/ Just released what may be my most significant project to date:
The first in-depth, data-driven account of how racial preferences actually operated at a U.S. service academy (the U.S. Naval Academy)—and the recent federal court case that challenged them (and lost).

zachgoldberg.substack.com/p/after-harvar…

Published on my Substack, the report draws on thousands of pages of filings, depositions, internal admissions data, and expert reports from Students for Fair Admissions v. USNA—a case that flew under the radar, despite its profound constitutional and institutional stakes. 2/ The full report runs ~115 pages—so there’s a lot to unpack. Too much for a single thread.

Thus, over the coming days, I’ll be posting a series of threads, each walking through key sections.

Note that these threads are high-level summaries—many important details are left out. I strongly encourage reading the full report for the complete picture.

Here’s the outline for this series of threads:

Thread 1 (this one): What the internal data reveal about USNA’s use of race—and how the Academy tried (and failed) to discredit the revelations.

Thread 2: The government’s sweeping (and evidence-free) justification for race-based admissions—and why it collapses under scrutiny.

Thread 3: How a federal judge upheld the policy—and why his ruling still matters, even after Trump’s executive order rescinded the policy.

Thread 4: What can be done to permanently outlaw race-based admissions at the service academies—or at least make it far harder for future administrations to reinstate them.

1/3 Most studies of the relationship between big 5 personality traits and political ideology examine the effects of broad or domain-level traits (e.g. 'Conscientiousness', 'Agreeableness'). However, the results of such studies can be misleading insofar as groups may have different or opposite scores on facets within those domain-level traits. For example, if liberals score higher than conservatives on the 'compassion' facet of agreeableness, and conservatives higher than liberals on the 'politeness' facet, these differences are likely to offset, resulting in a domain-level correlation with ideology that is small or indistinguishable from zero.

Given this issue, a relatively recent paper opted to examine the domain-level AND facet-level relationships with ideology. The chart below (you'll probably have to zoom in) visualizes/summarizes the estimates of these relationships. 2/3

Together, studies 1 and 2 found that A) the Openness (vs. Intellect) facet of Openness/Intellect (aka 'Openness to Experience) to be the strongest predictor of ideological liberalism. In fact, net of Openness, Intellect had no influence on ideology; B) after Openness, the 'Withdrawal' facet of Neurotism and the 'Compassion' facet of Agreeableness emerged as the 2nd and 3rd strongest positive predictors of ideological liberalism; and C) all facets of Conscientiousness were positively predictive of ideological conservatism, though the facet of 'Orderliness' showed a stronger relationship than that of 'Industriousness'; D) Some facets of extraversion appear to be mildly predictive of conservativism, which is interesting given that past studies have not found domain-level extraversion to be an ideologically-relevant trait.

Study 4, which utilized a different personality inventory, provides a substantive replication of these results: A) Aesthetic sensitivity, which is part of the 'Openness' facet, emerged as the strongest positive predictor of left-wing/liberal ideological orientations. Intellectual curiosity also has an independent positive influence, but one that is half the size of Aesthetic sensitivity; B) the Compassion and Respectfulness facets of Agreeableness were again positively predictive of left-wing/liberal and right-wing/conservative ideology, respectively; C) while comparatively more weakly, the Depression (and, to a smaller extent, Anxiety) facet of Neuroticism showed a significant positive association with left-wing/liberal ideology; D) All facets of consciousness were positively predictive of right-wing/conservative ideology, though 'Productiveness' was relatively more influential; E) Facets of Extraversion were again weak if significant predictors of ideology overall, though certain facets (e.g. Assertiveness) appear to point in the right-wing/conservative direction.

1/Strange. Palestinian Christians face the same oppressive occupation as their Muslim peers, and yet they are consistently vastly less likely to support acts of terrorism against Israeli civilians. Such a mystery. 2/2 Socioeconomic, geographic, and other demographic variables hardly explain the difference. Again, such a mystery. Wonder what it could be.

1/ I'm thrilled to release my latest @ManhattanInst report, which presents findings from the most comprehensive study to date of public perceptual accuracy with respect to the prevalence and racial distribution of police use of force manhattan.institute/article/percep… 2/Until now, pretty much the only data we had on this topic was a 2021 @SkepResCenter survey, which featured a question that asked people to estimate the number of unarmed black men killed by police + the black share of deaths by police in 2019.

https://twitter.com/ZachG932/status/1364024711592738817?s=20

1/ Similarly (and relatedly), while there are a few widening sex differences among 12th-graders on measures of depression, self-esteem, life satisfaction, self-derogation, and loneliness... https://t.co/GYpBJcbfCY

https://twitter.com/ZachG932/status/1679891549470044167

2/ ...the increasing ideological differences on these measures are far larger and more numerous

1/ Update of an earlier/2020 thread (

https://twitter.com/ZachG932/status/1270824521289924610?s=20

) with additional data+more groups: the median unarmed black police shooting victim now returns 21x the # of article results for a white victim, and 7-11x the # for Hispanic and 'other' (Asian, Native American) victims. 2/ 93% of black victims returned at least 1 article result vs. 60% of white, 70% of Hispanic, and 80% of 'other' victims. 40% of black victims were featured in at least 1 NYT article, as compared to 10% of white, 12% of Hispanic, and 10% of other victims.

Schools might be a good place to start. None of this appears inevitable, at least by the data showcased in my and @epkaufm's latest @ManhattanInst report

https://twitter.com/ImmuneHack/status/1630972999015727109?s=20

See also this chart from the appendix

1/ I'm excited to announce the release of my and @epkaufm's full @ManhattanInst report on the prevalence and effects of CSJ/CRT-related school exposure among recent high school graduates. manhattan-institute.org/school-choice-… 2/ For those that missed it, a preview of some of our main findings was previously published in @CityJournal (city-journal.org/yes-critical-r…), which I also summarized in this thread

https://twitter.com/ZachG932/status/1583231756748345345?s=20

1/ There's been some chatter of late (most notably from @Musa_alGharbi and @balajis) about a possible 'receding' of the Great Awokening. At least as far as attitudinal trends go, it may still be too early to know for sure (I'll have a better idea when more 2022 data drops). 2/That said, here are where things currently stand in my racial liberalism index (which I assembled as part of my dissertation work). First, slope among nonwhite Dems is leveling off. Slope among white Dems still pointing upward (hence the growing gulf between them and nw dems)..

1/The Dem Party is on the cusp of becoming majority-minority. However, in my first @ManhattanInst issue brief, I argue that if current trends persist, it is likely to become a maj-min. party that is inordinately and indefinitely governed by a highly educated white minority. 2/Researchers of American politics have noted a growing 'diploma divide', which is particularly pronounced among whites. Whereas a college degree (and vocab test scores) was traditionally positively associated with GOP identification among whites, this association has flipped.

1/n To continue with an earlier thread

https://twitter.com/ZachG932/status/1610771780653961216?s=20&t=a8OqSzygyfIbeaN6wzI8EA

, the idea that blacks and other 'PoC' suffer higher levels of stress (on account of racism/bias) is not borne out in the data either. 2/n Indeed, across 6 different datasets/measurements, levels of self-reported stress among blacks were consistently either significantly lower or statistically indistinguishable from those reported by white respondents.

1/n These patterns are quite remarkable when one considers the rampant white supremacy and oppression that PoC regularly endure. Or perhaps this speaks more to 'white fragility' (?) 😄. 2/n Results in terms of monthly or more.

1/ To what extent, if at all, is CRT being taught or promoted in America's pre-tertiary public schools? @epkaufm and I recently conducted a nationally representative survey of 18-20 year-olds to find out. While a full report of our findings will be published in the coming... 2/ ..months, we give you a preview of some of them in our latest @CityJournal article city-journal.org/yes-critical-r…

What follows is a summary thread.

1/n I'm pleased to announce the release of my first (of what will hopefully be many) @ManhattanInst research publications. The current report examines racial gaps in Dem pro-depolicing attitudes through the framework of 'luxury beliefs' theory. manhattan-institute.org/is-defunding-t… 2/n What follows is a 🧵summarizing notable findings, the most interesting of which, I think, is that support for depolicing policies among white Democrats either increases or shows no sensitivity to local violent crime levels. More on this later.

1/n If 2016 was the year white liberals started rating non-whites more warmly than whites, 2020 was the year they started.. a) rating whites as more violent than blacks ..b) rating whites as more lazy than blacks

1/n After trailing them somewhat in 2016, white Dems (51.8%) are now significantly more likely (p=0.001) than non-white Dems (43.6%) to prioritize the use of 'inclusive'/PC language over speaking frankly about issues and problems facing the country. Little change among white Reps 2/n Disaggregating the Dem racial/ethnic groups, whites are sig. more likely to give this response than blacks and hispanics, but are virtually tied with their Asian counterparts (who, as some contend, do their best to ape woke white elite sensibilities )

1/n 53% of white liberals (48% of white Dems) think that white people should feel guilty about racial inequality 2/n Needless to say, this is more or less the purpose of equity-oriented K-12 curriculums--i.e., cultivate white support for race-conscious/anti-white policies via the inculcation of collective guilt and shame.

1/n Very briefly (I'm vacationing in India atm), the 2016 CCES Team Module of American University shows that 30% of white liberals (19% white cons) don't think there ANY IQ diffs between wealthy and poor people. Among those that believe there *are* diffs..

https://twitter.com/RhiannonDauster/status/1468245826157821954

2/n 32.6% of white liberals (vs. ~18% of cons) indicated that genes were 'not important at all' in these differences, which was their modal response choice