WHEW it’s finally published!! New paper out led by the ever incredible Dr Anna-Sophie Jürgens, all about using #TextualAnalysis methods to study #science representations in #popculture 👩🔬✨🧐 We explore and describe ways of finding hidden meanings in images, comics, films etc. with scientific themes 🕵️♀️🔭 It’s a great intro guide for students & researchers interested in using these methods. https://jcom.sissa.it/article/pubid/JCOM_2303_2024_Y01/
Is it valid to say 40% of American police officers perpetrate domestic violence?
After half an hour digging through empirical #research I think I can confidently say "no." Why? Because we have no real idea how much DV is perpetrated by cops. Because nobody can really study it with solid methods. Because (IMO, anyway) a lot of scholars are kind of scared of #cops .
One guy testified for Congress in 1991, and he gets cited a lot, but his #ResearchMethods were not perfect. Then the rate of good research on this went down for the next 30 years.
Lots of opinions, "perceptions of...", etc., but `` numbers? I'm currently not finding any I trust a whole lot. One lit review (2016) found estimates of #DomesticViolence rates from 5% to 40%. Another article just said "Probably about the same as the general public," after clearly misreading a couple of previous estimates.
Anyway, nobody knows how much domestic violence police commit, and it sort of feels like the police want it to stay that way.
In case you missed it! Take a look at "Correcting Selection Bias in Big Data by Pseudo-Weighting" by An-Chiao Liu, Sander Scholtus, and Ton De Waal
"In this research, we extend the [Elliott/Valliant] EV method to be suitable for all ranges of inclusion probabilities, while retaining the attractive properties of the original study. Any model that is suitable for propensity estimation can be easily applied." #Statistics#ResearchMethods#BigData https://doi.org/10.1093/jssam/smac029
#DigitalHumanities and #ComputerScience intersect in the 21st-century transformation, challenging traditions.
This book explores combining qualitative and quantitative methods, providing insights from 10 research projects. It's a guide for cutting-edge research beyond conventional strategies. #openaccess 📖
Calling all researchers: McGill University in Montreal is running a free series of workshops on data anonymization in October. Registration information at the link below, hope to see you there! Please boost if you think you know folks this might be of interest to.
Just finished ‘Evaluating What Works’ by @deevybee & Paul Thompson. A really good primer on how to evaluate #interventions and, despite the speech and language therapy context, I found it highly relevant to #psychology & #metaresearch.
I’ll be adding this to my module’s reading list, but unfortunately I can’t put it on my Goodreads :ablobcatcry:
Google is getting a lot worse because of the Reddit blackouts
With Google’s generally poor search results nowadays, appending “reddit” has long been the default way I search for almost anything (and no, I’m not ready to get my info from an AI chatbot, either). But given the sheer volume of subreddits that are currently unavailable — including some of the most-subscribed subreddits — clicking through many Reddit links in search results takes me to a message saying the subreddit is private. ...
It's not often that I get to point out Internet Monopolies being sharecroppers.
The real dish here is that Reddit was one of the few domains in which the ad-fed #enshittification and #SidamTouch (ad-centric media turns everything to shit, reverse of Midas) wasn't ... overly dominant.
And now courtesy of mismanagement by #spez, #Reddit, #AdvanceMedia, and the Reddit board, #GeneralWebSearch which as been in a death spiral for years is suddenly getting far, far worse.
I've commented multiple times that I rely far more on traditional media (mostly books and magazine articles) these days than the Web. Sites/services such as #SciHub, #LibraryGenesis, and #ZLibrary have been absolutely vital for this, and despite much of the online world getting markedly worse, these are bright spots.
(Internet Archive, Wikipedia / Wikimedia, Project Gutenberg, and a handful of other sites/services are among the other bright spots which happen to operate inside the law, though the fact that useful sites have to violate law says a hell of a lot about how corrupt and societally-failing the law is these days.)
My #ResearchMethods for #ContentDiscovery now are based strongly on library research techniques I'd learned in the 1980s: research topics of interest, find major works and the authors of those works, read those, and if the same names or works keep turning up then find and read those. I'll also make heavy use of podcasts, especially those reviewing books and/or interviewing authors (particularly on academic topics), most notably the #NewBooksNetwork.
This may not lead you to truth, but it will virtually always point you to the foundations of present understanding and orthodoxy.
Truly principled authors will note conflicting / contradictory viewpoints --- #PatrickOphuls is excellent in this regard. Even unprincipled authors will often point out key voices in opposition to them, though usually by trash-talking and belittling them. (I'd found a wonderful example of this in a Reason review on Conway & Oreskes latest book The Big Myth.)