Mika Braginsky

made an #rstats pkg that calls #JuliaLang to fit mixed models with an interface consistent(ish) to lme4: github.com/mikabr/jglmm

How robust is your meta-analysis to publication bias, phacking, unmeasured confounding, and more? Me and Maya Mathur @stanford_qsu created a suite of R packages and Shiny apps to help you answer these questions, check it out at metabias.io

@mcxfrank @alephlove it's the tufte book webfont: edwardtufte.github.io/et-book/

@alephlove I think @mbraginsky knows…

Do you want to do a psychology experiment while following best practices in open science? My collaborators and I have created Experimentology, a new open web textbook (to be published by MIT Press but free online forever). experimentology.io Some highlights! 🧵

We are proud to publicly announce the MIT Graduate Student Union! MIT grad student-workers are unionizing to create a healthy and fair working and learning environment for all, by giving us a voice in the decisions that affect us. (1/4) #MITGSU

@bergelsonlab @tjmahr @mcxfrank @ev_fedorenko we're extracting the coefs df from the model and doing our own plotting, e.g. #L197" target="_blank" rel="nofollow noopener">github.com/langcog/wordba…

@mbraginsky @tjmahr @mcxfrank @ev_fedorenko cool, thanks @mbraginsky , looks like wordbank data:). are those plots some sort of ggplot + ggfortify from the lme4 output or something? if you had the repo/code link handy (email is fine) that'd be cool:)

halp! link me to ur fave #DataVisualization of REs (slopes & intercepts) in multilevel models, bonus if it's #ggplot & based on #brms or #lme4 output w/graph code, but i'm up even just pics from who knows what code. #RStats @tjmahr @mcxfrank @mbraginsky @ev_fedorenko others??

@bolkerb @tcarpenter216 @kleinschmidt @davidrfeinberg sounds great!

I used to use SSI's "HLM" software to run mixed models. Now I use lme4 but it's somewhat slow and the optimizer struggles to estimate at times. Without leaving R (or going bases), is there a better / modern solution?

@jmankewitz is she wrong

My sister describing my research: “It’s about acquisition I think. Yeah acquisition....or computers. Maybe both??”

@fobinf good idea, thanks!

@mbraginsky Great idea! I suggest to add geom_sina which will look nice inside the violins. Or alone.

made a pkg for pirate plots in ggplot: add any of points/means/bars/CIs/violins – better than ye olde bar/box plots github.com/mikabr/ggpirate

@Mika_Asaba @mollyllewis what if I'm bossy and grumpy, am I now mikæ

@mollyllewis the difference one letter makes! @mbraginsky

Curious about the semantic associations of your name? Check it out in this app! mlewis.shinyapps.io/namesemantics/

Pretty excited that our Wordbank book is out from MIT press - just got the hard copy! mitpress.mit.edu/books/variabil… Always free online at wordbank-book.stanford.edu

CHILDES is the premier resource for studying transcripts of children's language. We created childes-db.stanford.edu, a reproducible, versioned interface. Now childes-db has a 2020 update, with 1) updated R API, 2) new Python API, 3) phonbank, and 4) bugfixes.

The mystery of children’s first words. Great story in @TheAtlantic by @michaelerard - featuring my work on using larger datasets to study children’s language with Wordbank (wordbank.stanford.edu), and also a guest appearance by my son Jonah! theatlantic.com/family/archive…

@dzj_evalparse tends to be several orders of magnitude faster than lme4!

@mbraginsky What's the advantage of a Julia model

made an #rstats pkg that calls #JuliaLang to fit mixed models with an interface consistent(ish) to lme4: github.com/mikabr/jglmm

@LanguageMIT @kmahowald that's the same year I took it! wow I've gotten way less far in life

The black hole image algorithm boss lady - Katie Bouman - took my psycholinguistics lab class in 2013 (with @kmahowald TAing) cnn.com/2019/04/10/us/… She did a project on noisy channel interpretation, and now she is figuring out the origins of the universe

Work by Alessandro Sanchez, Stephan Meylan, @mbraginsky, @MacDonaldSt, and @danyurovsky.

Introducing childes-db: a new interface to CHILDES! Interactive plots of child language data + a reproducible API: babieslearninglanguage.blogspot.com/2017/10/introd…

@klmr you can just call it with bars=FALSE and presto!