Post

Observable
@observablehq.com

Great investigation of a classic dataset. Observable bundles the similar (but later) “cars” dataset from the 1983 ASA Data Exposition, hosted here lib.stat.cmu.edu/datasets/ and described in lib.stat.cmu.edu/datasets/car....

Tristan A.F. Long@thelonglab.bsky.social

As I prepare for another early morning #BI393 Biostatistics lecture, came across this deep dive into the history (and errors ?!?) in the well-known "mtcars" dataset! #rstats jengianoulis.substack.com/p/secrets-of...

jengianoulis.substack.comSecrets of the classic dataset "mtcars"Back in time to Sunset Boulevard in the 70s
Observable
@observablehq.com

In general, the epistemics of “sample data” are interesting. Their primary value is pedagogical; they may have a nice spread, a mix of data types, the right level of data quality issues. But they represent something real, which can get lost in the demos.

1 repost0 likes

Observable
@observablehq.com

We made *thousands* of charts of the “penguins” sample dataset — we could probably sketch many of them from memory — before we got any clue why Dr. Kristen Gorman had actually taken those measurements, and what it said about evolutionary pressures of change in sea ice extent due to climate change.