@max_haag @steffen_hurka & @CN_Kaplaner analyze EU directives &find complexity hinders implementation. States with higher administrative capacity cope better with complexities &Eurosceptic states face fewer infringements. Open access https://t.co/cQRRBrnv9R #RegGov @WileyPolitics
1/2 🔊Please share widely🔊
From July 24 - August 4, 2023, the University of Munich will host a Summer Institute in Computational Social Science. I'm super thrilled to co-organize this event together with @DrFollowMario & @Kunkakom.
#sicss2023
This follows a recent EUR-Lex update that saw the deletion of many 'deprecated' procedures from the search index. Most of these procedures are still available in CELLAR but cannot be accessed via the procedure search. [3/3]
EUPLEX dataset update
(available on https://t.co/AyqDCT9Aac)
We have just released a new version of the EUPLEX dataset. In addition to some bug fixes in the code base, the dataset now contains new legislative procedures up until 2023-01-03 and a new variable! [1/3]
Interview with Gary King, Professor of Political Science at Harvard: Listen now (57 min) | On Science, Data, Inference and More! https://t.co/ajbeqsmoIh
We present 'Neural String Edit Distance', a generalization of Levensthein distance that uses neural representation:
✏️ @jlibovicky, Alex Fraser
📎 https://t.co/52JqHIfjNo
🗣️Workshop on Structured Prediction
To all causality-minded readers who have been using Primer in the classroom and had trouble solving any of the homework problems - rejoice! We are now making the Solutions Manual available to all: https://t.co/Jx0L8KCsq7
So, let wisdom light the darker corners of our world.
📢📢📢Dear #TextAsData nerds in the social sciences, we have something to announce: The Text-as-Data Reading Group for the coming Spring term!
Eight (!!) amazing speakers from EU & US will present their outstanding text-as-data work to us and we'll discuss their papers in depth.
Very happy to teach "Using Directed Acyclic Graphs for Causal & Statistical Inference" again with @MarkusGFreitag :)
Expect: choosing control variables, post-treatment bias, sensitivity analysis, causal mediation, non-response bias, instrumental variables, and more!
For all the NLP/legal data/EU law nerds:
Looks like @eurlex has implemented some form of regular expressions/entity recognition to automatically create links in EU official documents whenever another EU act is mentioned in the text.
You can now dig into the EU law rabbit hole!
🆕 Now on our YouTube Channel:
📺 "Whose scale is it anyway?" by @conjugateprior
📄 You can also find the workshop materials here: https://t.co/eHeEPeCwPF
https://t.co/QlZk5LA4QV
🇨🇳 Social media censorship in China is the topic of the next #CIVICADataScience seminar.
We are proud to welcome Prof Margaret Roberts to share how her latest research makes use of the method of text matching.
📅 26 January
🔗 https://t.co/coKqdDM1vp | @CIVICA_EU#DataScience
New article out (online first, open access) in @EUP_TheJournal looking at the bargaining success of the European Parliament and the Council in EU informal trilogues using #textasdata: https://t.co/TaAtTrVA7M
🧵
Why do we observe this pattern for the EP and not for the Council? Reasons could be manifold, but may include the (relative) transparency of EP intra-institutional proceedings vs the opaqueness of the Council and "teenage" EP vs "adult" Council negotiators.