Estimating important things with public data

Intro One of the neatest things I’ve learned about in the last few years is how to estimate effect sizes from coarsened data. The original article discussed using the method with publicly available data reported by statewide education agencies, and that’s the approach I’ll use here too. But more generally, the method could be applied in any situation in which a continuous distribution is reported out in ordinal bins.

Looking into #KeepFamiliesTogether

This week I’m at the Seattle branch of the Summer Institute on Computational Social Science. Today, we discussed digital trace data and spent some time thinking about different sources of digital trace data. One of the easiest sources is twitter, and thanks to Mike Kearney, it’s easily accessible in R via the {rtweet} package! I joined a group addressing a topic I find myself dwelling on throughout the day - the immigration crisis at the border.

Peeking behind the curtain with {slidex}

I gave a lightning talk (slides here) this past weekend at the second annual Cascadia R Conference that was focused on creating and contributing new themes to the {xaringan} package, which is essentially a really well thought out and well-organized R Markdown wrapper around the remark.js package for producing beautiful HTML slides. At the end of that talk, I announced the work-in-progress version of my new R package, {slidex}, for converting Microsoft PowerPoint slides to R Markdown, and specifically {xaringan}, with a single function.