Maksim Rudnev 🇺🇦
@maksimrudnev.com
📤 63
📥 178
📝 571
Values, morals, SEM, and fun. Formerly
https://twitter.com/MaksimRudnev
not just arxiv
12 days ago
0
0
0
"when directly asked if it was an AI or a human, [AI] claimed to be human 100% of the time" "when asked to report a ZIP code, over 80% of [AI]’s responses were 90210 (Beverly Hills)" :-D
www.surveypractice.org/article/1460...
loading . . .
How to Detect AI-assisted Interviews in Online Surveys | Published in Survey Practice
By James Martherus, Alexander Podkul & 2 more. The emergence of agentic artificial intelligence (AI) tools capable of autonomously interacting with web interfaces presents new challenges and opportuni...
https://www.surveypractice.org/article/146046-how-to-detect-ai-assisted-interviews-in-online-surveys
19 days ago
0
0
0
reposted by
Maksim Rudnev 🇺🇦
koenfucius
about 2 months ago
Sure, correlation does not mean causation, but that doesn’t mean it’s not interesting. This site maps how a wide range of tastes and preferences correlate with personality traits (and vice versa), not all of which are unsurprising:
buff.ly/YwPu1TY
via
@spencrgreenberg.bsky.social
0
1
1
mkey
about 2 months ago
0
0
0
How to make an echo in your echo chamber sound even more like you? "Value Alignment of Social Media Ranking Algorithms"
arxiv.org/pdf/2509.14434
"users can use [basic value] controls to architect feeds reflecting their desired values. [such] feeds diverge.. from engagement-driven feeds."
loading . . .
https://arxiv.org/pdf/2509.14434
about 2 months ago
0
1
0
AI doesn't take away your dirty work, AI redefines it.
about 2 months ago
0
0
0
reposted by
Maksim Rudnev 🇺🇦
Yulia Chentsova
2 months ago
Jinli Wu's paper is out. "The mind carries greater significance in Western White cultural contexts (i.e., a mind or mentalist focus/orientation) and that behavior carries greater significance in East Asian contexts (i.e., a behavior focus/orientation).
journals.sagepub.com/doi/10.1177/...
loading . . .
Why the West Minds and the East Behaves: An Integrative Review of the Cultural Evolution of Mind–Behavior Orientations - Jinli Wu, Yulia Chentsova-Dutton, 2025
One fundamental characteristic of humans is that we have both “exteriors” (i.e., behavior) and “interiors” (i.e., mental states). This distinction between the m...
https://journals.sagepub.com/doi/10.1177/00220221251362805
0
0
1
Rethinking measurement invariance causally by
@dingdingpeng.the100.ci
It is preferable to work with a causal definition of measurement invariance. A violation of measurement invariance is a potentially substantively interesting observation. Group differences can be thought of as descriptive results
loading . . .
Rethinking measurement invariance causally
Measurement invariance is often touted as a necessary statistical prerequisite for group comparisons. Typically, when there is evidence against measur…
https://doi.org/10.1016/j.cresp.2025.100241
about 2 months ago
1
13
6
‼️
doi.org/10.1177/0022...
about 2 months ago
0
0
0
reposted by
Maksim Rudnev 🇺🇦
Ian Hussey
about 2 months ago
Proposal for how to fix family wise error rates. For every uncorrected p value you must add an extra letter to the claim. “Eating chocolate maaaaaaaaay be associated with lower rates of stroke”
0
178
51
"war exposure significantly predicted increases in religiosity"... p = .034
doi.org/10.1080/1050...
about 2 months ago
0
0
0
reposted by
Maksim Rudnev 🇺🇦
Kendra Albert
2 months ago
Very glad to see that someone is doing the important work of aligning AI alignment.
alignmentalignment.ai
loading . . .
Center for the Alignment of AI Alignment Centers
We align the aligners
https://alignmentalignment.ai/
22
409
154
Arguably, the most annoying way to name your paper: make it a question which you don't answer in the paper.
2 months ago
0
1
0
It's not around mainstream media, but right now in Serbia academics are blocked out of universities, and their students are beaten up by a police for just trying to keep up with their duties. Video from the Department of Philosophy, University of Novi Sad, Serbia
loading . . .
Suzavac i sukobi u Novom Sadu
YouTube video by Portal Mašina
https://www.youtube.com/watch?v=zrIWgmWbuKU&t=78s
2 months ago
0
2
0
More and more positions like that (not just in the U.S.)
2 months ago
0
0
0
A random paper, just—the art of science "In preliminary analyses, ipsative value scores showed almost no association with the outcome measures, whereas the raw mean scores showed robust associations. The raw mean scores were therefore used..."
doi.org/10.1007/s109...
3 months ago
0
0
0
reposted by
Maksim Rudnev 🇺🇦
Igor Grossmann, PhD
3 months ago
New paper in Proceedings B, forcing us to rethink the canonical understanding of cultural differences in independent and interdependent agency.
add a skeleton here at some point
0
3
1
"the mind carries greater significance in Western White cultural contexts (i.e., a mind focus) and behavior ... in East Asian contexts (i.e., a behavior focus)."
@yuliacd.bsky.social
I like this idea very much; explains all that Western fuss about "being a good person".
osf.io/preprints/ps...
loading . . .
OSF
https://osf.io/preprints/psyarxiv/gs2q4_v2
3 months ago
0
1
0
"Sage against the machine" — is a behind-the-paywall title of a Hindustan Times article reviewing our paper (
doi.org/10.1038/s414...
) on perception of wisdom in people across cultures (thanks to
@igi.bsky.social
) there's a paywall but just in case
www.hindustantimes.com/lifestyle/ar...
4 months ago
0
1
1
ok - might be sending all my *best* ideas there
imsj.org
4 months ago
0
0
0
Just like the ego in Freudian theory acts as a negotiator between the id and the superego, personality is a battleground where culture collides with situational demands.
#nocontextdrops
4 months ago
0
0
0
Values are not what people *are* (hence the weak association with behavior) Values are what people *aspire to be*
#nocontextdrops
5 months ago
0
0
0
Just a Cattell's box - good for teaching and reflecting on psychological data
5 months ago
0
0
0
I like Ellen Hamaker's articles, how very simple things such as a correlation coefficient may turn out so incomprehensibly complicated.
doi.org/10.1080/0027...
5 months ago
0
0
0
reposted by
Maksim Rudnev 🇺🇦
Julia M. Rohrer
5 months ago
Now officially out with nice formatting and all 🥳 "Thinking clearly about age, period, and cohort effects" -- a gentle introduction to the age-period-cohort problem and how to "solve" it through various types of assumptions.
journals.sagepub.com/doi/10.1177/...
12
276
76
wow they reached inside the LLM and identified specific value-mandating neurons in a neuron net. Sounds like a neuroscience but with LLM's brains
arxiv.org/pdf/2505.17712
5 months ago
0
0
0
Timothy Leary's wild bio came across, and GenAI seems to be our version of 60's LSD 1. both are acronyms 2. nobody knows how it works 3. feels sort of dangerous/illegal 4. uni's are excited AND terrified 5. the bravest of professors give it to students 6. everybody uses it anyways 7. very trendy
loading . . .
Timothy Leary - Wikipedia
https://en.wikipedia.org/wiki/Timothy_Leary
5 months ago
0
0
0
from Leary's handbook
5 months ago
0
0
0
what could be more natural
6 months ago
0
0
0
what "slider" and "measurement scale" actually mean Source:
www.nature.com/articles/s41...
6 months ago
0
0
0
indeed! should have been an epigraph to my blog
maksimrudnev.com/2025/04/24/n...
Source:
doi.org/10.1080/0022...
6 months ago
0
0
0
what a strange idea to measure values (=conscious beliefs) through implicit associations
doi.org/10.1016/j.pa...
www.sciencedirect.com/science/arti...
loading . . .
Development and validation of an implicit association test (IAT) for Schwartz's value dimensions
Research on personal values often relies on self-report measures, which can be subject to response bias and may not adequately capture the competing n…
https://www.sciencedirect.com/science/article/pii/S0191886925001448
6 months ago
0
0
0
timely
6 months ago
0
1
0
Next step towards making sense of non-invariance An effect size measure to quantify the degree of non-invariance and make sense of it, now with multiple groups It doesn't have cutoffs yet but it's for the best because invariance is continuous not binary
doi.org/10.1080/1070...
loading . . .
https://www.tandfonline.com/doi/full/10.1080/10705511.2025.2484812
6 months ago
0
0
0
"The utilitarian motivation for the social sciences is they can protect us from bad social-science reasoning."
add a skeleton here at some point
6 months ago
0
0
0
Bookmark: a nice collection of video seminars on basic values from the key scholars
whatdowevalue.com.au/events/
6 months ago
0
0
0
there is no such thing as [see thread]
scholar.google.com/scholar?star...
loading . . .
https://scholar.google.com/scholar?start=30&q=%22There+is+no+such+thing+as%22&hl=en&as_sdt=0,5
6 months ago
1
0
0
"Claude expresses many practical and epistemic values, and typically supports prosocial human values while resisting values like "moral nihilism" ...unlike human value.. Claude’s trans-situational values are predominantly service-oriented, pragmatic and epistemic"
arxiv.org/abs/2504.15236
loading . . .
Values in the Wild: Discovering and Analyzing Values in Real-World Language Model Interactions
AI assistants can impart value judgments that shape people's decisions and worldviews, yet little is known empirically about what values these systems rely on in practice. To address this, we develop ...
https://arxiv.org/abs/2504.15236
6 months ago
0
0
0
reposted by
Maksim Rudnev 🇺🇦
Justin Wolfers
7 months ago
"I'm going to govern in econometrics" is a winning campaign slogan for the ages.
20
1385
321
How do you come up with the scale name? There are no rules how to make a good name and little thought is given to it. I reflect on that in a blogpost btw: "Onomastics is the study of proper names, including their etymology, history, and use. "
maksimrudnev.com/2025/04/24/n...
loading . . .
Naming measurement instruments
Imagine you are trying to test criterion validity of your new Eggplant scale using an allegedly different Aubergine scale. How frustrating it is to realize that it’s the same thing and your attempt to prove validity generally fails because of that.
https://maksimrudnev.com/2025/04/24/naming-measurement-instruments/
7 months ago
1
0
0
Is it unethical to make public an anonymous review which you received on your manuscript after submission to a journal? I am so often tempted to expose the quality of some reviews. (Even journals that practice 'transparent' peer-review don't publish reviews of the rejected papers).
7 months ago
0
0
0
reposted by
Maksim Rudnev 🇺🇦
PNAS Nexus
7 months ago
A study measures the performance of ChatGPT and open-source LLMs in annotating US media tweets for political content. All models were outperformed by a supervised classifier. The authors urge caution when using chatbots for social science annotation. In PNAS Nexus:
academic.oup.com/pnasnexus/ar...
1
2
1
something about this friday night R coding
8 months ago
0
0
0
Are there any null-result scale development studies? It feels like a study aimed at introducing a new scale can't fail, otherwise it's not publishable. Moreover, a scale could be disqualified only by being replaced or forgotten. Then, overall, how scientific is scale development?
8 months ago
1
0
0
Our article "Dimensions of wisdom perception across twelve countries on five continents" received 13464 article downloads in 2024, placing it as one of the Top 25 downloaded Social Science papers out of 415 published in Nature Communications last year.
www.nature.com/articles/s41...
8 months ago
0
0
0
onlinelibrary.wiley.com/doi/full/10....
loading . . .
On the importance of being clear about the level of analysis of interest: An illustration using the case of self‐compassion
Objective Theories about within-person (WP) variation are often tested using between-person (BP) research, despite the well-established fact that results may not generalize across levels of analysis...
https://onlinelibrary.wiley.com/doi/full/10.1111/jopy.12924?campaign=woletoc
8 months ago
0
0
0
My new favorite English verb: it-is-what-it-is-ing
10 months ago
0
0
0
the authors pay $2690.00 so the readers can read their paper in Scientific Reports (
www.nature.com/srep/open-ac...
). But why not showing some questionable ads on the same page, right?
10 months ago
0
1
0
Another evidence that a person-country congruence of values is associated with higher SWB, now with political values but unfortunate measurement decisions and uninsured implied causality
doi.org/10.1007/s109...
loading . . .
The Effect of Political Regime on the Association of Values with Subjective Well-Being - Journal of Happiness Studies
Previous research has often demonstrated that liberal values and democratic regimes are associated with higher levels of subjective well-being (SWB). However, consistent results are only found at the ...
https://doi.org/10.1007/s10902-024-00841-9
10 months ago
0
0
0
Rejected hypotheses require more effort (and drive p-hacking)
10 months ago
0
0
0
Load more
feeds!
log in