Dave Hauser

The Sora AI disinfo nightmare is here

For more like this:
tiktok.com/@drewharwell
instagram.com/bydrewharwell

When I was a younger faculty, I kept wavering on whether I should apply for the Rising Star every year. And then when I finally worked up the courage to apply, I was no longer eligible. Lesson here is that you should not miss your shot and just apply! Don’t pray for someone to secretly nominate you!

Excited to share that I’ll be the incoming Editor of AMPPS. My first priority is building a diverse team of Associate Editors and Editorial Board members. If you’re interested, DM me or add your name via this super simple survey.
cornell.ca1.qualtrics.com/jfe/form/SV_...
Please share!

The threat of analytic flexibility in using large language models to simulate human data: A call to attention Social scientists are now using large language models to create "silicon samples" - synthetic datasets intended to stand in for human respondents, aimed at revolutionising human subjects research. How...

Can large language models stand in for human participants?
Many social scientists seem to think so, and are already using "silicon samples" in research.

One problem: depending on the analytic decisions made, you can basically get these samples to show any effect you want.

THREAD 🧵

Summary of design and results from our three studies. (A: Design) Each study used a similar experimental design, measuring both positive and negative demand in an online experiment, with three commonly-used task types (dictator game, vignette, intervention). Our experiments had ns ≈ 250 per cell. (B: Results) Observed demand effects were statistically indistinguishable from zero. The plot shows means and 95% confidence intervals for standardized mean differences derived from frequentist analyses of each experiment and an inverse variance-weighted fixed-effect estimator pooling all experiments (solid bars). Prior measurements of experimenter demand from a previous dictator game experiment (de Quidt et al., 2018; standardized mean difference from regression coefficient) and a meta-analysis primarily including small-sample, in-person studies (Coles et al., 2025; Hedge’s g statistic) are also shown for comparison (striped bars). The main text includes Bayesian analyses that quantify our uncertainty.

We often hear from reviewers: "what about demand effects?" So we developed a method to eliminate them. Something weird happened during testing: We couldn’t detect demand effects in the first place! (1/8)

I’ve noticed this too. If you are a Qualtrics user, you can discourage this by disabling copy+paste. Details below

bsky.app/profile/dave...

instructors! today is a good day to ask yourself this question again

Pseudo Effects: How Method Biases Can Produce Spurious Findings About Close Relationships - Samantha Joel, John K. Sakaluk, James J. Kim, Devinder Khera, Helena Yuchen Qin, Sarah C. E. Stanton, 2025 Research on interpersonal relationships frequently relies on accurate self-reporting across various relationship facets (e.g., conflict, trust, appreciation). Y...

In a new paper, my colleagues and I set out to demonstrate how method biases can create spurious findings in relationship science, by using a seemingly meaningless scale (e.g., "My relationship has very good Saturn") to predict relationship outcomes. journals.sagepub.com/doi/10.1177/...

Kingston folks, cute lemonade stand alert now (Mon Aug 18) on Napier between Brock and Johnson. Kids raising money for refugees. Show your support!

Are there reasons this would be a bad idea? Rather than tell me, give me the time to figure them out on my own, it’s the best way for me to learn

Caveats:
-must disable "new survey taking experience"
-survey layout must be flat, modern, or classic
-you should probably tell Ps you are disabling copy & paste just in case they are planning on drafting their responses in a word doc and pasting it in later

To prevent Ps from pasting (GPT-generated) responses to your text entry Qualtrics question, disable pasting.

Add this code to the OnReady section of your question's javascript:

jQuery("#"+this.questionId+" .InputText").on("cut copy paste",function(e) {
e.preventDefault();
});

Enjoy!

ChatGPT shows signs of the same biases that arise in audit studies of human beings.

When you give ChatGPT resumes, it's biased in how it evaluates minorities.

When you ask ChatGPT to generate resumes for women & minorities, it generates systematically different types of resumes

Cover page for the manuscript: Morey, R. D., & Davis-Stober, C. P. (2025). On the poor statistical properties of the P-curve meta-analytic procedure. Journal of the American Statistical Association, 1–19. https://doi.org/10.1080/01621459.2025.2544397

Abstract for the paper: The P-curve (Simonsohn, Nelson, & Simmons, 2014; Simonsohn, Simmons, & Nelson, 2015) is a widely-used suite of meta-analytic tests advertised for detecting problems in sets of studies. They are based on nonparametric combinations of p values (e.g., Marden, 1985) across significant (p < .05) studies and are variously claimed to detect “evidential value”, “lack of evidential value”, and “left skew” in p values. We show that these tests do not have the properties ascribed to them. Moreover, they fail basic desiderata for tests, including admissibility and monotonicity. In light of these serious problems, we recommend against the use of the P-curve tests.

Paper drop, for anyone interested in #metascience, #statistics, or #metaanalysis! @clintin.bsky.social and I show in a new paper in JASA that the P-curve, a popular forensic meta-analysis method, has deeply undesirable statistical properties. www.tandfonline.com/doi/full/10.... 1/?

FIGURE 7: LIFECYCLE PROFILE: FULL-TIME EMPLOYMENT AMONG PHD GRADUATES

FIGURE A3: SHARE OF PHDS WORKING AS PROFESSORS

FIGURE A5: TOTAL INCOME OVER THE LIFE CYCLE, BY HIGHEST DEGREE EARNED

TABLE B1: DESCRIPTIVE STATISTICS ON PHD COHORT: EARNINGS DURING THE PHD (BY GENDER, CITIZENSHIP, AND FIELD)

Is a #PhD worth it?

In data from #Canada, doctoral grads earned less at first because (A) they entered the job market later, but they surpassed others if they (B) got academic jobs and (B) kept them.

So a PhD's value is waning as B and C become harder.

econpapers.repec.org...

We’re hiring a tenure track assistant professor in our amazing Vassar cognitive science department! If you have any questions, let me know. I’ll be at Cog Sci in San Francisco all week if you’d like to chat in person

The word “Research” is doing way too much work. We need separate words for “creating new verifiable knowledge” and “looking shit up on the internet”

Lack Of Concrete Dinner Plans Leaves Power Vacuum Filled By Radical Pro-Tapas Fanatics

Lack Of Concrete Dinner Plans Leaves Power Vacuum Filled By Radical Pro-Tapas Fanatics theonion.com/lack-of...

Survey invitations often describe the topic of the survey. A simple, classic example of how this biases results: survey recruitment materials about bird watching led to higher estimates of the % of people who engage in bird-watching. doi.org/10.1093/poq/...

Dear Colleague It appears that your Wellbeing Awareness training has not yet been completed. This e-learning forms part of the University's strategy to improve health and wellbeing amongst its staff and is an important means of providing you with key information on how to manage work-life pressures.

MAYBE DON’T EMAIL ME AT THE WEEKEND, THEN.

Thinking today about how rich conservatives throw their money at 500 different people doing whatever they want and if 10 payoff they consider it money well spent while rich liberals will make you complete a 35 page proposal to hope to get $15k for certain prescribed activities.

Congrats, Bill!

screenshot of our public handbook

every year my lab does a re-read + edit of our Handbook, a documentation resource for how we do science

this year we also updated our Public Handbook, an open-access version for folks wanting to improve their own docs

it's at handbook-public.themusiclab.org and available for noncommercial re-use

I'm an Associate Editor at Current Directions in Psychological Science and I have a new Special Issue call on Intergroup Allyship. Proposal submissions are due by July 18. Please share widely!
www.psychologicalscience.org/publications...

Pro tip, folks: if you make a figure for a publication, don’t just hand the copyright over to the publisher. Post it on OSF first and then give yourself permission to use it in the first publication and then all future publications. That way, no permissions issues.

You wanna see something really scary? open.substack.com/pub/amandagu...

Social Psychology is calling for Special Issue proposals. We will be reviewing proposals on a rolling basis & foresee multiple Special Issues over the next year or more. For more information about what we are looking for, please see the call here:
osf.io/2f3vp
#socialpsych

It’s because of collocation (ie convention). I’m natural language they occur far less frequently in that order so it sounds weird. Like saying pepper and salt or quiet and peace. Collocations are learned thru experience so many ESL speakers might not notice the difference

It’s a big inferno from a big ol turtle

but if i'm not allowed to feed artists into my artist killing machine it's going to be basically impossible for the machine to keep killing artists. is that what you want?? and before you answer keep in mind this is a very expensive machine