Christian Pieter Hoffmann

Prof @UniLeipzig @ifkmw & Institute for PolSci @crifc_de @cdp_org (Private account & personal opinions, of course.)

[Christian Pieter Hoffmann @cphoffmann]

Angesichts des IT-Ausfalls der Uni Leipzig haben wir eine kleine Info-Seite ins Netz gestellt. Hier erklären wir, wie man uns in nächster Zeit erreichen kann 🙏🏼
https://t.co/PILRa7F2bX

[Jacob Mchangama @JMchangama]

The default position in democratic Europe is now to blame every bad thing on the Internet and insist that government regulation of speech is the answer. This will not end well.

[Rob Sica @robsica]

"misinformation is a symptom, not the disease" https://t.co/oQHYvyYUU4

[Jan Zilinsky @janzilinsky]

He says many 2024 Trump/Vance voters switched or abandoned the GOP because of cost of living/inflation, citing polling (what people said mattered to them).

I might have a better measure - let's tally up the actual economic hardships people experienced (losing a job, borrowing money to pay the bills & more) and see how they voted.

Trump voters who reported four or more economic hardships were nearly three times as likely to back a Democrat for Senate as those who reported none.

As @BrendanNyhan has been saying there's no reason to take stated reasons for voting a certain way at face value https://t.co/9XtsK5Hcwr; and @rp_griffin, @johnmsides have documented that asking detailed questions about people's personal finances is the way to go: https://t.co/WdgLVjKm7B

[Jan Zilinsky @janzilinsky]

NORC just released an amazing dataset (its Civic Health & Elections Project) so we can test, again, whether it's really true that (young) men are massively more conservative than women

The ideological gender gap exists but...

[Christian Pieter Hoffmann @cphoffmann]

@acerbialberto I still remember the glowing coverage of social media when Obama was elected…

[Jay Van Bavel, PhD @jayvanbavel]

Explicit measures of prejudice are much stronger predictors of discrimination than implicit measures.

A new adversarial collaboration found that self-report measures of racial bias exhibited greater predictive validity of discrimination than indirect measures (explaining ∼45% vs 3% of the variance).

The zero-order correlations between each indirect measure and behavioral measures of discrimination were quite low ( r = .00 to .19) By comparison, correlations between direct measures and discrimination were all statistically significant (r = .099 to .381).

Moreover, the behavioral discrimination measures generally produced effects favoring Black over White targets, which "complicates claims that anti-Black effects on indirect measures are a source of cumulative disadvantage."

https://t.co/dzi7s0nNW3

[Christian Pieter Hoffmann @cphoffmann]

“The surveys also revealed a persistent divide between experts who saw misinformation as widespread and consequential and those more skeptical, with sharpest disagreement on its behavioral impact and increase over time.” - @Sacha_Altay et al.

https://t.co/0hAp4JQCQ2

[John B. Holbein @JohnHolbein1]

It is deeply discouraging that “evidence is the bottom line in any empirical inquiry” needs saying at all.

[Michael Gilead @Michael_Gilead]

🚩 When Polymarket or Kalshi says there's a 70% chance something will happen, should we believe it?

Excited to share our new paper, just published in PNAS, in which we examined more than ONE MILLION forecasts from Polymarket, Kalshi, Manifold, and Metaculus, spanning 12 domains and forecasting horizons ranging from hours to years.

The main finding is striking: Prediction markets and collective forecasting platforms are remarkably well calibrated. When they assign events a 70% probability, those events really do happen about 70% of the time. And this pattern holds surprisingly well across different platforms, domains, and forecasting horizons.
But there's an important twist. When we looked more closely, we discovered systematic biases in some of the domains where forecasting matters most (wars, elections).

For decades, psychologists have documented how poorly humans often reason about probability and uncertainty. Our findings offer an encouraging counterpoint: When judgments and information are aggregated, we can achieve a surprisingly accurate understanding of an uncertain future.

That doesn't mean we can predict what will happen next with certainty. But it does mean that we can be remarkably good at estimating how uncertain the future really is.

This work was a great collaboration. The data for this project was meticulously collected for years by Justin Dickerson, the project was led by Hilla Shinitzky and Maayan Peer, with @mattansb leading the analyses.

📄 Read the paper:
https://t.co/CJVEgcWIp7

#Forecasting #Polymarket #Kalshi #DecisionScience #PNAS

[Laurenz Guenther @GuentherLaurenz]

France currently experiences large-scale rioting, mostly by (former) immigrants. Such events are often treated as isolated incidents. They are not.
As I show here, immigration-related riots are becoming increasingly frequent and intense over time:
https://t.co/UKrfFfC430

[Christian Pieter Hoffmann @cphoffmann]

„Given the very weak relationships between SMU and the severity of anxiety and depression symptoms, SMU is unlikely to account for the contemporary youth mental health crisis.“

https://t.co/1p9VHRL0Xc

[Rolf Degen @DegenRolf]

First meta-analysis of the most authoritative studies: Adolescents' use of social media accounts for less than 1% of their subsequent symptoms of depression and anxiety.
Most of the empirical investigations into social media use and mental health symptoms have utilized cross-sectional designs or have shown small effects, leading to scepticism in the field. More conclusive evidence could be synthesized from the full range of more robust longitudinal and experimental designs. Experimental designs are required to draw conclusions regarding causal relationships. Longitudinal designs afford conclusions regarding the direction of effects over time.
To the best of our knowledge, this is the first review [and meta-analysis] to comprehensively synthesize longitudinal and experimental evidence on the association between time and frequency of social media use and mental health symptoms. We identified a considerably higher number of longitudinal papers than in a previous review.
The key findings of this review and meta-analyses were that the relationships effects between time and frequency of social media use and depression and anxiety symptoms were statistically significant but relatively weak The time spent by young people using social media accounts for less than 1% of their subsequent symptoms of depression and anxiety. There is no evidence that this effect increases as age decreases. Reducing time spent on social media by young people is unlikely to be effective in preventing the worsening of their symptoms of depression and anxiety.
Given the very weak relationships between social media use and the severity of anxiety and depression symptoms, social media use is unlikely to account for the contemporary youth mental health crisis. In relation to policy implications, a relationship between duration and frequency of social media use on subsequent anxiety and depression is not a solid basis for the rationale for age-related legal restrictions on social media.

[Jay Van Bavel, PhD @jayvanbavel]

Liberals and conservatives behave differently in everyday life--conservatives seem to be out with others at lot more while liberals stay home and doomscroll.

This study analyzed activity using smartphones to track college students and found that conservatives were more likely to be at parties, noisy places, fraternities and religious facilities. Liberals were at home, resting, doing chores, and browsing social media.

This helps explain why conservatives are, on average, happier than liberals.

The behavioral differences were small but robust, not accounted for by other plausible factors (e.g., demographics), and most pronounced in the leisure domain. https://t.co/73xqgsJ8pg

[Steve McGuire @sfmcguire79]

If the trend holds, academic journals will soon be receiving twice as many submissions per year as they did before the introduction of ChatGPT.

AI will write them. AI will review them. And AI will read them.

[Dina D. Pomeranz 🟣 @DinaPomeranz]

Bias against boys in education

[Jay Van Bavel, PhD @jayvanbavel]

Cornell just produced a massive new report on higher education noting that companies have a “jarring” skepticism of higher ed thanks to a new generation of degree holders who lack key skills like handling uncertainty, compromising, and absorbing feedback.

Faculty also report that incoming students, despite four years of undergraduate education, are “underdeveloped” in these relational or interpersonal skills.

This is a serious crises in higher education and the report calls for a greater need for conflict-resolution training. We need to do a better job training our students to deal with these issues.

These are traditionally the skills students developed in higher education. I loved college because it challenged me, sparked debate, and required me to grapple with complexity. This should be an essential part of college--inside and outside the classroom.

https://t.co/qvTg2hB7HO

[Jay Van Bavel, PhD @jayvanbavel]

Toxic workers impose a huge cost on organizations.

Avoiding a toxic worker (or converting them to an average worker) provides far more

benefits than retaining a superstar.

Assuming that it is no more costly to avoid a toxic worker (or replace them with an average worker) than it is to find, hire, and retain a superstar, it is also more profitable to do the former over the latter. https://t.co/ZqBnbupu2t

[Mohsen Mosleh @_mohsen_m]

1/ New paper led by @yuweichuai w/ @captaineco_fr @DG_Rand. In Nov 2025, X began showing where accounts actually operate from. We studied 8,200 politically engaged accounts claiming to be in the US; half were revealed to be elsewhere. What happened when the mismatch went public? 🧵
Pre-print: https://t.co/kw53EaciHT

[Sacha Altay @Sacha_Altay]

With Claude I did a systematic literature review of misinformation prevalence, exposure, and concentration.

There is a companion website that allows you to play with the data and signal me if an estimate is wrong or missing: https://t.co/Vf0kLIcBm3

[Sean Westwood @seanjwestwood]

"It's not that the public does not care about democracy. But Americans' understanding of the democratic way of doing politics is one of competitive parties, robust public participation, and open and free debate about the paths their communities should take. They see it differently from Democratic elites and progressive activists." Incredible new article: https://t.co/Y0f1vWRuaK