Jesse Onland

Figure 2 from the linked article.

Abstract of the linked article

New paper for anyone working with data:
Better data viz - for free, in few clicks.

Below, all N = 111, M = 0.04, SD = 0.27.
One-sided t-tests vs. 0 yield: t(110) = 1.67, p = .049.

Use raincloud plots. Or risk wrong conclusions!
Plot w/ @jaspstats.bsky.social today!

🧪 📊 #PsychSciSky #StatsSky

🧵👇

Will Posit's recent transition to hyping slop generation endanger its B Corporation status? Surely you can't be serious about climate targets while shilling this stuff, right?

Does this company only have "AI" in its name for marketing reasons, or is it actually engaged in slop generation?

The plot on the right is exactly what I had in mind!

Can you superimpose the exact and smoothed plots with differences in lighter grey?

Maybe just plotting a smoothed line on top would be more straightforward, I suppose.

Also formalized the frequentist interpretation of probability.

One of many line graphs in the NASA Mariner-Venus Report from 1962. Due to the rather black and bold and many grid lines, the straight lines that show the results (which are also black) are almost invisible.

Find the data 🧐🤓
#dataviz #datatoinkratio

Would the "The Lady Tasting Tea" be too breezy?

Not certain. I don't want a facsimile edition but other than that I'm not too picky.

Embarrassing. Why not just train a slop generator on all the conference submissions and then skip actually running it?

Maybe I'm too ggplot2-brained, but I wouldn't have guessed that "make one line in this chart dashed and the rest solid" is advanced!

Every time I can't figure out how to get what I want, the solution always seems to be some zany kludge.

My struggles with Tableau 📊 seem to boil down to two issues:

1. Too little control over the aesthetic mapping. (E.g. cannot map to position.)

2. Dimensions and measures are treated radically differently (e.g by filtering) in ways that are unintuitive and/or just never what one would want.

I've been on the hunt for a copy of this, but it seems scarcer than his other book, "Graphic Methods for Presenting Facts".

Watch the seismic waves from today's magnitude 7.3 Alaska earthquake ripple across seismic stations in North America.

More ➡️ loom.ly/kaJp5I0

(After live-generating 80 lines of SQL with five paragraphs of pre-written prompts)

"This would have taken hours!"

Maybe for *you*.

I'm sorry, but if the image slop generator you used for your slides can't produce a world map with South America in it, I'm not buying your data engineering pipeline slop generator.

Pretend tools for pretend work.

Plot by distance to an ocean port and you get a different set of outliers.

Data Strips Experiment I built a “Data Strips” app to experiment with new ways of graphically summarizing the distribution of a single variable.. You can try it out or access the code on GitHub. This post wil…

I've written an explainer blog post for my "data strips" #dataviz workbench web app, trying alternative 1-D distribution summary plots. rawdatastudies.com/2025/07/05/d...

I wouldn't want to be associated with that creep either, but I think if you open-source your #dataviz 📊 theme, you have to accept the possibility of creeps using it.

In any case, I'd never assume that two chart authors were associated just because they use the same style, and you shouldn't either.

Vertical axis trickery in Bigalke et al. (2021), via dynomight.net/blue-light/. 📊

A #dataviz adjacent post from Andrew Gelman about which Venn diagrams can be drawn with rotational symmetry. The same Venn after whom the diagrams are named originated the frequency interpretation of probabilities, incidentally.

(At least for that aspect ratio.)

Slicing it into a stacked bar chart would help, but 200 bars is too many.

Some hard-hitting #dataviz coming out of New York.

Typo: "ertility" in the fill scale label.

Does Posit develop open source data science tools, or just LLM plug-ins? Based on their social media presence, you'd assume it were the latter.

Screenshot of a web app displaying 13 one-dimensional horizontal views of the same data set. Summary of each: 1. Heatmap of fixed-sized intervals colored by relative count. 2. Kernel Density Estimation using a gaussian kernel and multiple bandwidths. 3. Rug plot with a vertical line at each data point, up to 2000 points. 4. Density Strip: KDE using color and a bandwidth multiplier of 50%. 5. Highest Density Regions plot with cut-off points at 0.0, 0.5, 0.95, and 0.99. 6. Highest Density Regions plot with cut-off points at 0.0, 0.5, and an adaptive outlier threshold based on a gaussian extrapolation of the middle region. 7. The shortest contiguous half that contains at least 50% of the data, applied iteratively to show 1/2, 1/4, 1/8, 1/16 and the shortest half mode. 8. Quantile regions at percentiles that correspond to equal intervals if the data is Gaussian. Each region would be one standard deviation wide. 9. Density Rug: Shortest regions of 20%, 50%, and 80% of the data, allowing for split regions penalized according to the Split Penalty. Other values are shown as a rug plot, except touching values are connected as a single region to avoid looking more dense than the shortest regions. 10. Shorth and Mode: Shortest half of the data as one or two contiguous intervals; any split is penalized according to the Split Penalty. A vertical line shows the "half sample mode" which is the iteratively applied shortest half. 11. Inner region shows Interquartile Range (IQR) of the data with a line at the median. 12. Inner region shows Interquartile Range (IQR) of the data with a line at the median. When the outer region extends beyond the common box plot whiskers (1.5 IQR), the endcaps are shown as arcs. 13. Box Plot: Inner region shows Interquartile Range (IQR) of the data with a line at the median.

6 old and 7 new 1D #dataviz. Trying Shortest Half with a twist: one break is allowed. Also half-sample mode & count-adaptive outlier thresholds. Here's 1000 random normal points plus two outliers. The green ones are new. Try it at xangregg.github.io/data-strips/

In short, we *are* reaching new audiences!

blank_theme_elements <- function(theme_elements) { rep(alist(element_blank()), length(theme_elements)) |> setNames(theme_elements) |> do.call(theme, args = _) } data(mtcars) mtcars |> ggplot(aes(x = as.factor(cyl), y = mpg)) + geom_boxplot() + theme_classic() unwanted_theme_elements <- c("axis.title", "panel.grid") mtcars |> ggplot(aes(x = as.factor(cyl), y = mpg)) + geom_boxplot() + theme_classic() + blank_theme_elements(unwanted_theme_elements)

A little trick to set unwanted #ggplot2 theme elements to `element_blank()` en masse. 📊

(Copyable code in alt text.)