Embracing a Unix-like philosophy toward data After quite a long hiatus, we’re back! My goal is to write more regularly, but, you know, we’ll see. When I was younger, I used to regularly watch Alton Brown’s Good Eats with my family. And, as any ...

New edudata blog post: embracing a Unix-like philosophy toward data

www.edudata.blog/em/

#edusky #databs

Counselor: These feelings of "everything sucks," how long have you had them? Butt-head: Uhh, since everything started to suck, I guess.

Here it is, your moment of zen.

a man in a striped shirt is holding a cup ALT: a man in a striped shirt is holding a cup

No matter how good I think my code is at the time of writing, I’m almost always repulsed by it 6 months later

The more I code in languages other than #rstats, the less I like R’s preference for implicit returns. Sure, you can save yourself a line of code, but the explicit return just feels better

Being Data-Informed vs Data-Driven Within the past week or so, I’ve encountered two people who have purposefully distinguished between being “data driven” and “data informed.” First, Rick Saporta differentiates these two approaches in ...

New edudata blog post: being “data driven” vs being “data informed”

#edusky #databs

www.edudata.blog/being-data-i...

Most data science work is just asking SMEs to help me understand their data/help me make sure I’m filtering the data correctly.

I just pushed v0.1.0 of my {blueycolors} #rstats package to GitHub.

Check it out if you’re interested in using Bluey-themed colors in ggplot:

ekholme.github.io/blueycolors/

Virginia's VALLSS 2024-25 Fall-Spring Growth Report Programming note: I’m back after a long summer hiatus. I honestly just kind of lost motivation to do weekly posts. So, moving forward, I’m not going to promise weekly posts. I’ll shoot for a 1 or 2 ti...

After a few week hiatus, edudata is back with another post. This one reviews Virginia’s fall-to-spring VALLSS growth report and offers some critiques about measuring student growth using ordinal categories:

www.edudata.blog/virginias-va...

#dataBS #edusky

I just learned about “doubleML models.” Is this just data science/ML people trying to reinvent causal inference with the worst name possible?

#rstats #datascience

The narrative fallacy and education I recently read this article on the Ludicity blog about the narrative fallacy, and it got me thinking about how this applies to education. Just to make sure we’re all on the same page, the narrative ...

This week’s Edudata blog post is on the narrative fallacy, school improvement, and root cause analysis

www.edudata.blog/the-narrativ...

#education

I’ve been doing stats/data stuff “seriously” for about 10 years, and I still think bootstrapping is voodoo.

It’s the “ONE WEIRD TRICK” of statistics.

Casio

Tweet: Stats twitter: “All statistical methods fail horribly without telling you and you shouldn’t trust any research ever. fml.” Data science twitter: “Here are NINE SIMPLE WAYS to convert your data into ACTIONABLE INSIGHTS using POWERFUL AI. Number 3 will astound you!! I’m on a boat”

Among my favorites, from @cameronpat.bsky.social

Stacked bar charts and data viz tradeoffs Stacked bar charts aren’t good plots, and this is a hill I will die on. Even though everyone over the age of, say, 7, has seen a stacked bar chart at some point in their lives and probably intuits wh...

New edudata blog post: alternatives to stacked bar charts, and some thoughts on tradeoffs when visualizing data

#dataBS #rstats #edusky

www.edudata.blog/stacked-bar-...

Yeah true. I put a literal bet on pope pizzaballa, so that’s just my copium

Here’s my copium: if he had been elected pope, he’d adopt a papal name that wasn’t pizzaballa. At least now, we still have cardinal pizzaballa

Some hopefully useful questions to guide data-driven decisions I’ve mentioned in multiple posts on this blog that the primary “point” of data is to help inform decisions. There are probably occasions where it’s useful to just report out on broad compliance-y thin...

New edudata blog post: In which I self-consciously offer a framework for data-driven decision making

#datascience #edusky

www.edudata.blog/some-hopeful...

Maybe plotnine is more suspect. As a primarily #rstats user, seaborn feels better than matplotlib but still mainstream enough in the Python community. But I also only use it for eda stuff, and when I need to make something polished, I use ggplot

I don’t think so. Polars seems to be gaining “share” in code, and sklearn supports polars df’s for all (most?) operations. And if the concern is learning from others’ code, LLMs give an alternative - my experience is they’re great at producing polars code.

a man and a woman are dancing in front of a sign that says fast lake spartans chess team ALT: a man and a woman are dancing in front of a sign that says fast lake spartans chess team

TFW the Google Gemini VSCode extension just writes your #rstats
function documentation for you

Luxury goods and non-negotiable values I’ve been very much into the Acquired podcast lately. The premise of the podcast is that each episode describes a single company, usually beginning at some pivotal pre-company occurrence (the birth of...

New edudata blog post: what can educators learn from luxury brands like Hermes?

#edusky #dataBS

www.edudata.blog/luxury-goods...

I keep coming across new lists of education research (and adjacent) job opportunities so if you are looking for work, check out these links. 👇

docs.google.com/spreadsheets...

docs.google.com/spreadsheets...

docs.google.com/spreadsheets...

www.purposephilcareer.com

Most importantly, you’re using the correct (ISO 8601) date format 🙃

I Feel Like I'm Taking Crazy Pills I recently stumbled onto a preprint research article examining the writing quality of college students before and after the introduction of ChatGPT, as well as some reporting on the same article publi...

New edudata post: do we really need “big data” to tell us obvious things?

www.edudata.blog/i-feel-like-...

Yeah, good point. Having to dig in a level to pull out the dataframe could be tedious for end users

Thanks!

So I can ensure that the objects passed to my functions are appropriate. But it also feels like I’m adding an extra layer that’s maybe not necessary? Thoughts on this vs just passing in a dataframe and checking against, say, column names?

I’m writing an #rstats package to work with data extracts (csvs) from a specific source. I mostly want a set of functions that will do data-frame-like operations on these extracts. My inclination is to create a new class that’s a thin wrapper around the data (formatted as a dataframe)…

For my educator friends in the RVA area, VCU is hosting (and I’m helping run) a free event on AI in education on 4/16.