A chart showing the Fidesz margins in the single-member-constituency vote in the April 2026 Hungarian parliamentary elections.
Fidesz ended up winning only 10/106 single-member constituencies:
• 3 with the narrowest of margins
• A further 4 with small margins (at least compared to Tisza's)
Congrats! They're lucky to have you.
For the cost of one Harvard Professor of the Constitution Only Applies to White People, you could endow 10 professorships of any less commonly taught language at any university in the USA
Ever felt uneasy about how concepts & measures of "grey zone" autocracies are used interchangeably across the literature? Then check out my 🚨new paper🚨 in @democratization.bsky.social and get some substance to back up your gut feeling: tinyurl.com/ypdm2kna Or at least read the main points below ⬇️
This is very interesting - how did you estimate the sensitivity and subjectivity scores?
My dad died a few weeks ago. He was a good dad and funny and interesting and lived a remarkable life. He was a journalist, and, as a result, his friends tend to write wonderfully. I want to share this delightful remembrance by his friend John Jeansonne:
Why I don't expect liberal democracy to go away any time soon: My article on the resilience of liberal democracy, published in @jeppjournal.bsky.social.
Perfectly timed to the day after Orbáns defeat and a small spike in "democracy-optimism".
There are, I argue, good reasons for such optimism.
Wednesday we hosted Maria Pevchikh, chief investigator of the organization that Alexei Navalny founded. It was a very special event. A few thoughts.
Three 2-year full-time postdoctoral positions in autocratic politics at the Department of Political Science at Aarhus University (in connection to projects headed by Jakob Tolstrup and Alexander Baturo). Flexible start date. Application deadline: June 1, 2026.
international.au.dk/about/profil...
But shall this crazed old man be tamely suffered to drag a whole ship’s company down to doom with him?
This debate exists because of the way we teach statistics in the social sciences--we focus on *linear models* instead of *distributions.*
From a distributional perspective, the answer is easy. The Bernoulli distribution is the maximum entropy distribution for *any* binary variable. (1/2)
For anyone who is in the throes of grief a robin to remind us that love itself never dies only the body
The world's academic systems are in some bizarre competition regarding which ones can, structurally speaking, feed the most young scholars' lives into the woodchipper.
💡What is a V-Dem Country Coding Unit?
A V-Dem “country” is a political unit enjoying at least some degree of functional and/or formal sovereignty. Watch the video or read in the thread to find out more. 👇
🧵1(8)
#PoliSky #PoliSciSky #PolDataSky #PoliticalScienceData
#VDemMethodology #VDemDataset
Very interesting reflections on interpreting trends in V-Dem data for a specific country (Mongolia) from an expert on that country.
The sharp and persistent decline in expert ratings we observe corresponds with declines reported by @vdeminstitute.bsky.social and @freedomhouse.bsky.social, but our more frequent surveys show the decline was concentrated in early 2025 and ratings have largely stabilized since.
New Bright Line Watch report!
-Expert ratings of U.S. democracy largely stable at lowest levels measured (since 2017)
-Record partisan divide in public ratings of democracy driven by Republicans whose evaluations are highest since 2017
-Experts near-unanimous on many threats to democracy
🧵 below
In 2010, Mr Orbán's Fidesz won power pledging to build up a "national bourgeoisie". They have since fulfilled it, and then some, with their hierarchical clientelistic project.
Today's alternative to liberal democracy isn't "national" or "conservative" democracy. It's clientelist neopatrimonialism.
💡 What does V-Dem Data not cover?
Watch the V-Dem Methodology video or read in the thread to find out! 👇
🧵1(9)
#Polisky #PolDataSky #PoliSciSky #Academia #VDemMethodology #VDemDataset
@vdemamlat.bsky.social
Coming very soon: DEMSCORE Data version 7.0 released March 25th!
The DEMSCORE v. 7.0 release includes the most recent data updates from @viewsforecasting.org, @repdem-org.bsky.social, @qoginstitute.bsky.social, @vdeminstitute.bsky.social and #COMPLAB.
#PolDataSky #SocialScienceData #Polisky
I've been asked over the last few days/weeks what I think will happen on April 12th in Orbán's 🇭🇺 (and the aftermath), so here's my assessment, looking at everything as it stands now 🧵
tldr: It's complicated, and no one really knows (and those who say they do are either lying or overly confident) 1/
The claim in the new V-Dem release that the US now has the same liberal democracy score as in 1965 underscores the challenge of reducing a multidimensional concept to a single index. This graph compares the US in 1965 and 2025 on key dimensions V-Dem codes.
The Academic Freedom Index Update 2026 is out. This year, we focus on institutional autonomy and compare the decline in institutional autonomy in the United States with that in other major autocratizing countries.
academic-freedom-index.net #academicfreedomindex @fau.de @vdeminstitute.bsky.social
Two versions of the same aerial photo above each other. The photo shows a parking lot with many cars. The top version is high-res, with many colors, the bottom low-res, with 8 colors.
“Is it *really* a democracy?”—the type of question political scientists try to answer by converting fine-grained or multi-dimensional measures into few categories. And miss how such coarsening works. Convert a hi-res photo into a lo-res 8-color one. There’ll be information loss, possibly with …
/n
Definitely agree about the need for continuous reflection - I'd just highlight that there's robust (and open! see the Co-PI commentary in the DR, pgs 42-43) debate across the broad V-Dem network about the interpretation of the data and yearly updates, especially at the highest levels of aggregation.
A chart with V-Dem's Academic Freedom Index 1945-2025 for China, Hungary, India, Poland and Russia.
In states with R&D-intensive sectors, the challenge the polit elites curtailing acad freedom face is doing it w/o losing R&D's econ benefits. The PRC has innovated in this in an autocratic context, Hungary in an originally democratic one. Learning the effects however needs counterfactual estimation.
...while debates about expert-coded data (e.g. the V-Dem indicators, which are the direct aggregations of expert evaluations and measure relatively low-level concepts) are a whole other type of debate.
High level indices like RoW or the electoral democracy index rely on very specific conceptualizations about what democracy is and what indices and indicators best reflect the component concepts....
Each level of this process is subject to different types of judgments and critiques.
One other thing to consider is that there are many different levels of V-Dem coding. An index like RoW is a theory-derived aggregation of theory-derived aggregations of algorithm-derived aggregations of algorithm-derived aggregations of expert coded data...
