Diogo Ferrari

190 posts

Diogo Ferrari

Diogo Ferrari

@DiogoFerrari

Political Scientist & Statistician. Assistant Professor at the University of California, Riverside @UCRiverside.

Katılım Temmuz 2009
491 Takip Edilen385 Takipçiler
Diogo Ferrari
Diogo Ferrari@DiogoFerrari·
@GerardoMunck Yes, but saying “properties that are not part of democracy,” suggests that there is one correct concept of democracy. I am not defending the index, but “what democray is” is contentious. It’d be better to say that the index does not match a very specific notion of democracy.
English
2
0
1
142
Gerardo L. Munck
Gerardo L. Munck@GerardoMunck·
These are the reasons why I share Przeworski's criticism of V-Dem's democracy indices. 👇 The same criticisms apply to the data from Freedom House and the Economist Intelligence Unit. The problems with these indices are significant.
Gerardo L. Munck tweet media
Gerardo L. Munck@GerardoMunck

Przeworski on V-Dem Data In a new interview, Adam Przeworski states that he does “not have much trust in V-Dem’s measurements of democracy” and that V-Dem seeks “media exposure by heralding crises of democracy.” I agree. See the full interview here: populismstudies.org/professor-prze…

English
5
14
65
7.5K
Diogo Ferrari
Diogo Ferrari@DiogoFerrari·
There is nothing deeply wrong with profession incentives, reviewing processes, and journal publication practices. It is just that political scientists are extremely good at only discovering and developing correct theories that are +90% of the time supposed by data.
Ryan Briggs@ryancbriggs

I have a new paper. We look at ~all stats articles in political science post-2010 & show that 94% have abstracts that claim to reject a null. Only 2% present only null results. This is hard to explain unless the research process has a filter that only lets rejections through.

English
1
0
1
158
Diogo Ferrari
Diogo Ferrari@DiogoFerrari·
This paper is analytically sophisticated and achieves a good degree of technical performance. Will it make theoretical progress?
Theory and Society (Springer Nature)@Theory_Society

Despite sophisticated analytic methods in social science, fields remain limited in their theoretical progress and consensus. Why? New in @Theory_Society, "The illusion of rigor: Why analytical sophistication cannot substitute for inferential coherence," by Aamir Rashid and Rizwana Rasheed. Check it out here: link.springer.com/article/10.100…

English
0
0
1
82
Diogo Ferrari
Diogo Ferrari@DiogoFerrari·
@yudapearl I completely agree. But structural assumptions for many social sciences problems are just really hard to believe. So, an alternative question is: Can researchers draw useful conclusions from assumptions they not even partially believe?
English
1
0
0
50
Judea Pearl
Judea Pearl@yudapearl·
I'm reposting this reply because many readers, primarily among economists, are raising similar questions about the relations between the Potential Outcome framework and the Causal Inference frameworks. The issue: Can researchers draw useful conclusions from assumptions they do not understand?
Judea Pearl@yudapearl

Here is my last post on the topic: My attention was caught by the paper on "conditional ignorability" by Elias Bareinboim and Drago Plecko causalai.net/r120.pdf. Why? Because I've been trying for years to convince Imbens that it makes no sense for researchers to articulate assumptions whose plausibility they cannot judge. Bareinboim and Plecko's paper explains formally why this is so. Another gem is this paper: ucla.in/3Rarp0C. Ask any PO economist if he/she can distinguish "good control" from "bad control" -- they can't, because conditional ignorability is cognitively formidable. @analisereal

English
1
1
19
4.2K
Diogo Ferrari
Diogo Ferrari@DiogoFerrari·
New publication at @The_JOP. Using experimental evidence, the study shows that a metacognitive intervention—informing partisans about how party cues influence policy opinions—can weaken partisan bias and reliance on party labels when evaluating policies. journals.uchicago.edu/doi/10.1086/73…
English
1
0
1
65
Diogo Ferrari
Diogo Ferrari@DiogoFerrari·
Studying a particular LLM under a specific evaluation setup and concluding that "LLMs exhibit motivated reasoning" is like running a linear regression on one dataset and concluding that linear regression generally produces large positive coefficients.
English
0
0
0
34
Diogo Ferrari
Diogo Ferrari@DiogoFerrari·
Models are developing fast, so these papers’ conclusions about “AI/LLM” may be gone and wrong next week.
English
1
0
0
40
Diogo Ferrari
Diogo Ferrari@DiogoFerrari·
One thing that bothers me about some current AI/LLM research in social sciences is sentences like “AI/LLM does this”, while in the paper they use a very particular model (eg. gpt-4o). So, their conclusions or title should not be about AI/LLM in general, but the model they used.
English
1
0
2
207
Diogo Ferrari
Diogo Ferrari@DiogoFerrari·
@arthur_spirling Have you used Claude code or codex? It may reduce it to 7 min and without the cutting and pasting part. Not perfect, but pretty decent.
English
1
0
0
42
Arthur Spirling
Arthur Spirling@arthur_spirling·
Folks ask me how new AI models will change what academics do. So this morning I vibe-coded (Opus) a set of fixes to an R package that's been broken and thus dormant for five years. Seems to be working perfectly. Took 15 minutes of cutting and pasting. github.com/ArthurSpirling…
English
2
0
1
589
Diogo Ferrari retweetledi
UCR Political Science
UCR Political Science@UCRPoliSci·
Check out this new publication from #UCR 's @DiogoFerrari ! Does feeling threatened lead voters to back more conservative candidates? A new experiment finds partisanship and policy matter more than status threat. Read in J. Exp. Pol. Sci: doi.org/10.1017/XPS.20…
UCR Political Science tweet media
English
0
1
3
176
Diogo Ferrari
Diogo Ferrari@DiogoFerrari·
Our graduate students will be better off in the long run if they are trained primarily in Python, and will navigate easier across fields. R just has accumulated more stats libraries, but it will likely change soon.
English
0
0
3
111
Diogo Ferrari
Diogo Ferrari@DiogoFerrari·
It's time for social scientists to start moving away from R and embrace Python. The entire AI community relies on Python, and most APIs are readily available in it. R falls short compared to Python (not real OOP, poor code encapsulation, and other structural drawbacks). 1/2
English
2
0
3
214
Diogo Ferrari retweetledi
Yann LeCun
Yann LeCun@ylecun·
SB-1047 would definitely have a chilling effect on open source AI and the entire AI ecosystem. Very much hoping that Governor @GavinNewsom will veto it.
Andrew Ng@AndrewYNg

A decision on SB-1047 is due soon. Governor @GavinNewsom has said he's concerned about its "chilling effect, particularly in the open source community". He's right, and I hope he will veto this. If you agree, please like/retweet this to show your support for VETOing SB-1047!

English
44
112
626
178.9K

Keşfet

@GerardoMunck @The_JOP @claudeai @yudapearl @arthur_spirling @elonmusk @BarackObama @taylorswift13