You can add location information to your Tweets, such as your city or precise location, from the web and via third-party applications. You always have the option to delete your Tweet location history. Learn more
I’ve spent the last couple days reading RCT studies from criminology. Almost all restrict analysis to compliers & program-completers, instead of using random assignment as treatment or IV. 
Remember: just because it’s an RCT doesn’t mean results are unbiased. Read the paper.https://twitter.com/amitabhchandra2/status/1075931428322492417 …
![🏴{ [ RJ Miller ] }🏴](https://pbs.twimg.com/profile_images/693364505007382529/e9bdtIyj_normal.png)
Related, I’ve read several papers that compare “random samples” from treatment and comparison groups. The only reason I can think of to do this (rather than using the full samples) is to try to fool readers into thinking you ran an RCT.
@jenniferdoleac are you thinking of doing a meta-analysis if the RCTs in crim lit that corrects these issues?
Some students and I are working to re-analyze data from a few studies for the PELS relication conf this spring. But it’s been challenging to get data from study authors — unlike in econ, data from criminology papers often isn’t online.
That was my guess (about data availability). If you need any help, I have a couple students that would be happy to help with the data search.
Thanks! I’ll keep you posted.
@KoehlerJA Apropos your question about the replication crises and criminology. Might be that it just hasn't hit yet....
Criminologists have been able to differentiate meaningful RCTs from the kinds described here for a long time, so low-quality work getting published in the first place neither bothers me as much as it seems to bother @jenniferdoleac, nor do I think it prefigures a crisis per se
Seconded
I know lots of criminologist do this better —- it’s concerning mainly bc these studies still feature prominently in “what works” databases.
Oh yes that can def be concerning
Agreed. And having reviewed studies from other disciplines it's 
RCTs that restrict analysis to compliers are the Fools Gold Standard of causal inference..
It seems wrong to apply randomization to tthe criminal justice system, but without it we will never get the data we need to decide which policies really work
So public health researchers are advocates and criminologists are lousy trialists? 
I know stata gets halfway there, but I’d really love to see a package / checklist that says “put your data into this shape and I’ll run all the things you should run.” No need to reinvent the wheel every time.
The last few years I started talking with (non-economist) criminologists about various projects and it’s honestly been eye opening how few knew about things like this (and were pretty dismissive about it, alas).
You are getting to be a good candidate for Law School...
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.
