Michael Wallace

Hello new followers! You're probably here for the Roll Up stuff but some other things I talk about: 1. Stats! Contrary to some reports, winning free coffee isn't really my full-time job. I'm a biostatistician working on developing statistical methods to analyze health data.

Still getting things set up but here is where my occasional posts will mostly be from now on, hope to see you there!

Up in Ottawa for the second time in two months for @isi_wsc where literally every member of staff I have talked to was British and while I confess that's only an n = 3 situation it was still a bit unnerving

LRT it is definitely very normal when you first learn of a violent attack on your campus via the university subreddit which you only check every few days

My talk about my first experiences making general audience statistics videos has been uploaded - I'm by no means an expert so very much a 'here's what I learned', but hopefully some useful tips! outlook.office.com/mail/safelink.…

About to give a talk as part of the @michiganstateu Center for Statistical Training and Consulting data visualization seminar series, all about making introductory statistics videos - the talk is being recorded so hopefully online after! cstat.msu.edu My best slide:

@weepforsleep @NoContextBrits Excellent

@NoContextBrits @statacake 🤣

It's not just 'disabled people not seen in public', but also 'disability not seen in public'

A useful diagram, and while certainly true that an inaccessible physical environment limits the visibility of disabled people in public spaces, it's important to note societal pressures that encourage disguising disability also contribute to the 'disability is rare' misconception

Trying to find my way around @Carleton_U like

Whenever I see a conference uses an app my heart sinks. Not because I'm some luddite but because invariably these things are near impossible to use with low vision. Here are all the settings for this one - not even a dark mode :| @SSC_stat @WhovaSupport

It is time. Time for stats.

The thing I will never not find weird about staying in a hotel vs being at home is having to put shoes on before I eat breakfast

About to set off for the SSC annual meeting in Ottawa, which somehow takes 8+ hours to get to on trains for what would be a 5 hour drive 🙃

#UWaterloo is proud to announce that a full tuition waiver will be offered to all qualifying students from two First Nation communities on whose traditional territory the University is situated. More: bit.ly/431yA10 | #UWaterlooNews

@DylanSpicker Between that, and what you've observed above, I *think* that just about sorts it all out!

@DylanSpicker They *think* that during the earlier period there was a practice to record race information *if* no citation was issued, whereas in the latter period it's available for almost everyone, but there's no citation indicator for that time period

Twitter can you help with this data puzzle A dataset of police stops includes subject race and whether a citation was issued. Race is missing for ~30% Of those issued a citation, 3% have race recorded. Of those not issued a citation, 97% have race recorded. Any ideas why?

@DylanSpicker Oh, interesting - warnings presumably would be comparatively unlikely to record race (vs. citations), although still curious as to why race is being so reliably recorded for non-warnings/citations/arrests but then mysteriously missing for so many citations/warnings

@statacake Interesting. I wonder if it is a Chicago problem. The Chicago readme states "Data includes warnings and arrests, but is missing warnings". This appears to be a typo, but I wonder if Chicago data contains "warnings" under "citations" which could be the bulk of the stops?

@DylanSpicker I'm just looking at the Chicago dataset, which shows about 30% 'citation' - I'm now wondering if there's an issue with that variate since if across all locations it's a 5% citation rate, that would seem pretty odd for this location

@statacake Is this looking across all locations? And how much does the data skew towards "no citation"? The reported coverage rates (github.com/stanford-polic…) and your results seem to suggest that something like 95% of the total records would have to be "no citation". Is it that dramatic?

Obviously a big caveat emptor for anything too complex, and you've got to check their work, but I'm finding it's a real game changer for basic data cleaning/processing

Slowly training myself to remember that LLMs exist when I'm coding - little isolated tasks that might take me 5-10 minutes of googling/thinking take 1-2 minutes via an LLM and those time saves *really* add up