This paper is being honored with a Best Short Paper Award at IEEE VIS 2025! I'm truly grateful for the recognition and looking forward to discussing this work at the conference. #ieeevis

Alex Kale@kalealex.bsky.social · 12mo ago

Announcing my forthcoming short paper at #ieeevis! In this solo-author work, I examine and critique the logic of generalization about decision support in visualization research. arxiv.org/abs/2508.06751

Announcing my forthcoming short paper at #ieeevis! In this solo-author work, I examine and critique the logic of generalization about decision support in visualization research. arxiv.org/abs/2508.06751

Toward a Logic of Generalization about Visualization as a Decision Aid

Visualization as a discipline often grapples with generalization by reasoning about how study results on the efficacy of a tool in one context might apply to another context. This work offers an accou...

arxiv.org

I'm extremely excited to share this new paper out of the Data Cognition Lab! In it @krisha-mehta.bsky.social, Gordon Kindlmann, and I reframe visualization as a mechanism for data disclosure and develop a vocabulary for how visualization design induces loss on underlying data signals.

Krisha@krisha-mehta.bsky.social · 12mo ago

I am excited to share a new paper titled “Designing for Disclosure in Data Visualizations” set to appear at IEEE VIS 2025 (link below). 1/n

Today, along with 2,000 other NIH employees, I had to clear out my office 😭 It was truly the honor of my life to work with such incredibly passionate people focused on improving human health. I’ve never experienced a more positive culture where *everyone* cared about their job and serving others.

I wish more HCI research was this careful about measurement. If we do turn the corner, I think we’re gonna find out that a lot of our study results don’t mean what the authors originally said they meant.

George Wu@geowu.bsky.social · last yr.

A new working paper with Daniel Banki, @urisohn.bsky.social and Robert Walatka, just submitted to SSRN. The paper is comment on Ryan Oprea's recent AER paper. The paper is processing, but you, my friends, get early entry. papers.ssrn.com/sol3/papers....

Thanks to the wonderful folks at @dsi-uchicago.bsky.social for making this video!

Data Science Institute@dsi-uchicago.bsky.social · last yr.

How can helping people think with data benefit society? Explore the research of Alex Kale (@kalealex.bsky.social), Assistant Professor of Computer Science and Data Science at the University of Chicago! www.youtube.com/shorts/8SkXI...

What tech company will be the first to stand up and say that gutting the NSF is bad actually? The economic argument seems pretty clear Less NSF --> fewer PhD students --> fewer researchers --> smaller AI tech pipeline --> slower progress --> less competitive globally

Most of your life won’t go as planned. Stop placing your happiness, health, and self-actualization on the horizon.

luna@imlunahey.com · 2y ago

people over 30 quote this with some life advice for the rest of us?

“The executive orders make it quite clear that certain DEI activities are considered to violate current law, and we cannot fund anything that includes activities deemed unlawful.” Okay, but until a few weeks ago, promises about DEI were required. Idk how any of our grants survive this litmus test.

Somewhere in the U.S., there’s a scientist staring at their NSF/NIH grant application wondering why they bother. This post is for you. Science and society both need you. Hang in there and know there is a whole community supporting you.

A great thread if you’re into visualization for model interpretability and Bayesian stats!

Michael "Shapes Dude" Betancourt@betanalpha.bsky.social · 2y ago

Problems arise, however, when we try to communicate the structure of p( theta | tilde{y} ) when the parameter space contains more than a few dimensions.

Anyone know how to turn off the AI assistance feature in Overleaf? I didn’t want this. I didn’t consent to this. I find the suggestions distracting in a way that basically makes the tool unusable for deep work. Am I alone in feeling this way?

This is so upsetting. I have two grants out for review, basically my whole summer worth of effort scuttled. Multiply this by thousands of scientists whose time and ideas are being thrown away. It’s especially concerning for early career researchers whose futures depend on securing funding.

Carl T. Bergstrom@carlbergstrom.com · 2y ago

I don’t have a definitive source on this yet but it looks like NSF review panels have been canceled. Sixty-some years ago, Eisenhower transformed higher education because he viewed a science education gap as the single greatest threat to national security. This is absolutely an act of sabotage.

I’m feeling for the researchers who are losing access to funding.

Michael D. Green, PhD@michaeldgreen.phd · 2y ago

The academic research environment felt stacked against us even before the new administration. My early career friends and colleagues are disheartened about our prospects and the downstream effects of less support on our careers and science. www.science.org/content/arti...

“Universal fruitstuff magnitudes” is now a term of art

Alex Holcombe@alexh.bsky.social · 2y ago

“A lot of psychology is refusing to admit we can’t measure something. Instead, we calculate a standardized effect size and claim we turned the apples and oranges into universal fruitstuff magnitudes that can be directly compared.” - me, trying to teach statistics to a research student here

Professors, don't be mean to job candidates, they remember it the rest of their lives. Source: remembered it the rest of my life.

📊 #dataviz #map Very cool interactive Voronoi diagram tessellation of the globe partitioned by nearest major airport www.jasondavies.com/maps/voronoi...

World Airports Voronoi

The major airports of the world and their closest regions.

jasondavies.com

Just because I keep seeing the "replications show which effects are real" or "effects that didn't replicate once are flukes" heuristics repeatedly showing up on my TL today, I'm re-upping this thread. I beg everyone to please stop searching for quick and conclusive judgments on what's "true".

Berna Devezer@devezer.bsky.social · 3y ago

1. Reproducibility is not a generally reliable indicator of particular issues in a field. The problem as we understand is to expect high levels of reproducibility unconditionally, regardless of the specifics of the scientific field. We need to calibrate our expectations.

The distinction @jessicahullman.bsky.social makes between 1st and 2nd order uncertainty is important to understanding prediction vs inference. Conditional predictive performance (1st order) can be checked, but whether we’ve learned the right distribution (2nd order) is what we need to generalize.

Calibration “resolves” epistemic uncertainty by giving predictions that are indistinguishable from the true probabilities. Why is this still unsatisfying? statmodeling.stat.columbia.edu/2024/12/31/c...

This made me laugh.

Arlen Parsa@arlenparsa.com · 2y ago

One of the proposed new Illinois flag designs, out of thousands submitted

How wrongheaded to think that “engagement” without communicative intent will keep actual people on a dying platform. Meta is phoning it in at this point. This bit about appealing to younger audiences is either wishful thinking or a blatant lie to assuage nervous investors.

Culture Crave 🍿@culturecrave.co · 2y ago

“We expect these AIs to actually, over time, exist on our platforms, kind of in the same way that accounts do" They think this and other AI products they are rolling out will help attract and retain a younger audience www.ft.com/content/9118...

Check out the earthquakes over the past year with this interactive globe: observablehq.com/@joewdavies/... You can toggle between country borders and tectonic plates. I quite like being able to gauge the depth of them. #earthquakes #seismology #cartography #globe #dataviz

“Where are the findings? You only provide quotes.” New article considers qualitative researchers’ experiences of methodologically incongruent peer review feedback Open Access: dx.doi.org/10.1037/qup0... Few quotes follow 🧵

This guy chased me off of the psych methods Facebook group about 10 years ago. I’m not surprised to hear others had issues with him. You really don’t want to be remembered by your colleagues primarily for verbal abuse. What an unfortunate way to live.

Jim Coan@jimmycoan.bsky.social · 2y ago

"Coyne’s blogs were verbose and erratic, often veering into personal attacks. He freely fabricated conspiracies and leveled baseless accusations that obscured whatever legitimate critiques he might have had."

Writing is thinking. It’s not a part of the process that can be skipped; it’s the entire point.

Anna E. Clark@annaeclark.bsky.social · 2y ago

This is what's so baffling about so many suggestions for AI in the humanities classroom: they mistake the product for the point. Writing outlines and essays is important not because you need to make outlines and essays but because that's how you learn to think with/through complex ideas.

I imagine this is how most statistical consulting turns out. There’s only so much you can do with “garbage in.”

Crystal Lewis@cghlewis.bsky.social · 3y ago

Super cool way to move between aggregate and discrete distributional representations! Reminds me of Steve Drucker’s work.

Joe Davies@joewdavies.bsky.social · 2y ago

If you're interested in animating gridded data into discrete bins using webgl feel free to reuse our library: github.com/eurostat/reg... The example shows a population grid of Europe being rolled out, and then groups the cells according to their populations. #stats #webgl #cartography