iiinova — where your digital self keeps evolving

🧵4/4 3. Inspired by @nrennie.bsky.social's contribution: Using a similar approach, can you find evidence for Beatlemania?

000

🧵3/4 2. Inspired by @mickael.canouil.fr's contribution: define and defend a metric for "gender neutral" names. Based on your metric, what do you notice about changes in gender neutral names over time and across the three regions? Do those trends hold in the US data?

100

GitHub - hadley/babynames: An R package containing US baby names from the SSA

🧵2/4 1. Access, wrangle, and compare this data to the US baby names. Hint: The babynames package (github.com/hadley/babyn...) has this data through 2017, but students will have to go directly to the source to update to match this dataset's 2024 end date.

github.com

An R package containing US baby names from the SSA - hadley/babynames

200

Sara Stoudt @sastoudt.bsky.social · 15/06/2026

This week's #TidyTuesday has me thinking about classroom activities. 🧵1/4 Warm-up: Write a function to check the Rank column.

141

Creative writing meets data! Try out our app to find that data-driven detail you need to write about.... sharks! 🦈

000

GitHub - sastoudt/reading-for-humanness

And as an added bonus, what about speculative fiction and sci-fi for giving us new ways in to discussions of data ethics? Some examples of "data tales" here: github.com/sastoudt/rea...

github.com

Contribute to sastoudt/reading-for-humanness development by creating an account on GitHub.

000

#ecots26 continues...just finished my "Reading for Humanness" workshop! Reading guides help students with both comprehension & reflection on the human impact of algorithmic decision making, featuring articles from @themarkup.org that connect well to statistics and data science topics.

100

We collaboratively made some data art representing temperatures during past USCOTS and eCOTS.

Heat map of daily maximum temperatures during historical eCOTS and USCOTS events. We made a data mobile (color coded felt balls hang in strings representing each year).

000

Sara Stoudt @sastoudt.bsky.social · 11/06/2026

It was so fun! Students re-analyzed the data you all shared to try to find other stories in there and used your measurement guide to measure a pair of their own pants. Then, in class, we did some basic comparisons of the two datasets to see if anything has changed.

010

I'll be talking about measuring the old fashioned way. This past semester my intro stats students collected their own jean pocket data to compare and contrast with the iconic @pudding.cool pockets investigation (pudding.cool/2018/08/pock...).

pudding.cool

Women's Pockets are Inferior.

If you wear women's clothes, you already know this. But now we've got the data to show it.

220

Sara Stoudt @sastoudt.bsky.social · 11/06/2026

#ecots26 kicks off today with our Central Pennsylvania Regional Meeting at Bucknell. Our theme is Slow Data. I'm very excited to hear about and make some data art led by Kim Roth. (karoth.github.io)

karoth.github.io

About Dr. Kim Roth

About me

Sara Stoudt @sastoudt.bsky.social · 09/06/2026

This week for #TidyTuesday I'm doing a quick look at "grade" alignment, using dplyr's new filter_out and recode_values functions for the first time. Overall, CinemaScore had more higher grades while Metacritic and Rotten Tomatoes were harsher in their scores.

Clustered bar chart showing grade level of film scores across three sources (CinemaScore, Metacritic, and Rotten Tomatoes). Less than 20% of films had scores. Numeric scores were turned to grades. Minus or plus grades were aggregated. Overall CinemaScore had more higher grades while Metacritic and Rotten Tomatoes were harsher in their scores.

081

Sara Stoudt @sastoudt.bsky.social · 02/06/2026

It would be interesting to track down policy changes that correspond to big jumps. Know of any? Please share that context!

020

Sara Stoudt @sastoudt.bsky.social · 02/06/2026

Some of the key moves: used the countrycode package to get nice country names (cran.r-project.org/web/packages...) [eurostat and ecb columns work for the join except for Greece and the UK], pivoted longer, and used a wrapped labeller in my facet_wrap (ggplot2.tidyverse.org/reference/la...)

cran.r-project.org

countrycode: Convert Country Names and Country Codes

Standardize country names, convert them into one of 40 different coding schemes, convert between coding schemes, and assign region descriptors.

Sara Stoudt @sastoudt.bsky.social · 02/06/2026

A lot to unpack for #TidyTuesday. Some countries (e.g. Norway) always have equal leave policies per parent type. Others (e.g. Greece) have co-parent protections slightly lower than those of the birth mother. Others (e.g. Austria) show co-parent protections catching up to those of the birth mother.

Job protected leave duration in weeks between 1970 and 2024 across 21 European countries by parent type (birth mother v. co-parent)

170

Sara Stoudt @sastoudt.bsky.social · 28/05/2026

The align features of #cowplot just saved me (thanks a million @clauswilke.com)! See the capabilities for yourself: wilkelab.org/cowplot/arti...

wilkelab.org

Arrange multiple plots into a grid — plot_grid

Arrange multiple plots into a grid.

020

Am I finally going to be able to abandon Google Slides? This is a game-changer. 🤩

120

What about you? Share the backstory of your data story! 🤓 4/4

010

Then, I looked at the full time series for those countries (facet_wrap for the win), focusing on energy consumption/ supply & renewable capacity/distribution losses (an inefficiency angle to go with the initial rural/urban divide?) I'm still workshopping to see if anything interesting pops out. 3/4

I first looked at the ratio of fuel access between urban and rural by country (only one snapshot in 2010) and then picked a few countries with equal access, a few countries with really large discrepancies, and a few countries with a typical discrepancy level, trying to balance across continents. 2/4