siri show me where universities are

For science to be self correcting we need scientists to call out (and respond to) valid critiques of results no matter where they are in the publication process.

Lots of respect to the authors below for not just notifying the original paper's authors/publisher, but especially for sharing publicly

CALDER Webinar: The Four-Day School Week: New Evidence for District Leaders and Policymakers Featuring Emily Morton and Andrew Camp Wednesday, April 22 at 1:00 ET

📢 Happening TOMORROW!

Do four-day school weeks deliver on their anticipated benefits?

Join Emily Morton & @andrewmcamp.com for a deep dive into the realities of 4DSW implementation and its effects on teacher recruitment and retention.

Register ➡️ bit.ly/4rNbwP3

starting to think I must just be really good at finding null effects

CALDER Webinar: The Four-Day School Week: New Evidence for District Leaders and Policymakers Featuring Emily Morton and Andrew Camp Wednesday, April 22 at 1:00 ET

📢 Just TWO days away!

Join Emily Morton and @andrewmcamp.com this Wednesday, April 22 @ 1 PM ET for a webinar exploring the latest research on four‑day school weeks—what the evidence shows and how it can inform district decision‑making.

Register ➡️ bit.ly/4rNbwP3

It's not too late to sign up for this webinar! Really excited to summarize some of the work on 4DSWs and teachers + listen to Emily's latest research in this area.

Wednesday at 1pm ET

Registration link ⬇️

So yes, education research needs stronger norms around sharing data and code. But that doesn't happen by tsk-tsking folks.

That happens when the legal/ethical constraints surrounding sharing data are understood and solutions developed.

Also: we should call out bad research when we see it.

Instrumental Variables in Randomized Trials - Blueprint Labs Many randomized clinical trials fail to play out as intended: some participants assigned to the treatment group remain untreated, while others assigned to the control…

Instead, wouldn't it be better to learn across fields and improve research quality across all the board?

Example: blueprintlabs.mit.edu/research/ins...

Does this mean that education research is more rigorous than medical research? Of course not! That's a silly distinction to try and make.

But it does mean that research is complicated and assessing the credibility of research across fields is complicated.

AI stethoscope trial highlights the perils of implementation gaps When used as intended, an AI stethoscope enhanced the detection of cardiovascular disease — but low uptake and workflow challenges hampered effectiveness in a large pragmatic trial.

Recent example: www.nature.com/articles/d41...

This is something that I would have learned not to do in the first weeks of my ed school based quant methods course

Focusing on a single aspect of research (e.g. data availability) is a poor way to assess the credibility of a field or paper.

For example, medical research commonly reports "per protocol" effects in RCTs that are heavily contaminated by selection bias.

They don't use IV or estimate a LATE

"Quant research in < FIELD > is bad" is a take that is both (A) lazy and (B) wrong. Certainly not something that would appear in an outlet that punishes "sharp, well-argued opinion pieces."

The reality is that there is a distribution of quality across and within disciplines.

Not to say it isn't a problem, but how much is about how ed researchers interact with broader society vs. the larger way that society and science communicate.

Regardless, I think ed researchers on the whole are very interested and invested in addressing this issue!!

I've been thinking about this a bit and am wondering if this is specific to ed or really is just a function of scale.

There is plenty of bad med research ("reading an hour a day makes you live longer).

If there are ~4.5x as many teachers and doctors, shouldn't we expect 4.5x the amount of BS?

Strongly support this point. Conflating prestige with rigor is anti-scientific and all too common!

Also love this from the article:

"We should not waste instructional time or squander teachers’ goodwill by spending time and money on programs that don’t help anyone but the companies selling them."

What?

I've never seen a hotter take.

Alternatively: "More than a third of smokers will develop lung cancer. Some researchers say trying different brands could help individuals choose a better tasting cigarette."

A split bar chart of YouGov polling data with the headline: "Americans are most likely to say T. Rex is their favorite dinosaur, but many don't have a favorite." The chart has the sub-headline: "Which of the following is your favorite dinosaur? (%)." The chart has the note: "Note: "Other" includes responses of archaeopteryx, spinosaurus, plesiosaur, ankylosaurus, allosaurus, parasaurolophus, dilophosaurus, diplodocus, iguanadon, and pachycephalosaurus, as well as responses of "other." We know pterodactyls and plesiosaurs aren't dinosaurs. Opinion about dinosaurs comes from the question, "How much do you like or dislike dinosaurs?""

A shocking new poll result: Many Americans somehow don't have a favorite dinosaur.

And only 6% give the correct answer (triceratops).

Check out YouGov's new polling on Americans and dinosaurs: yougovamerica.substack.com/p/whats-your...

The only use of 'framework' I allow is pejorative

CALDER Webinar: The Four-Day School Week: New Evidence for District Leaders and Policymakers Featuring Emily Morton and Andrew Camp

📢 Just one week away!

Join Emily Morton and @andrewmcamp.com next Wednesday, April 22 @ 1 PM ET for a webinar exploring the latest research on four‑day school weeks—what the evidence shows and how it can inform district decision‑making.

Register: bit.ly/4rNbwP3

The community we study here has protective policies and a welcoming institutional posture towards immigrants. ICE/CBP has been less active in this region (northeastern U.S.) than other areas.

If effects are this large here, they may be substantially larger in less supportive contexts.

Our results are informative for researchers studying this issue in other contexts. Studies using MLL status etc. find smaller effects than we do using student birthplace.

We argue this is due to measurement error. Birthplace is a tighter proxy for who is vulnerable; other est. are lower bounds.

Event study plot showing weekly estimates of the attendance gap between foreign-born and U.S.-born students. The x-axis spans 20 weeks before and after inauguration. Pre-inauguration estimates cluster around zero with no clear trend. Post-inauguration estimates shift upward to approximately 2-3 percentage points and remain persistently elevated through the end of the school year, with no sign of attenuation.

Effects increase with grade level.

While there is near-zero effect in Pre-K through 3rd grade, 5th grade+ students are 3-6pp more likely to be absent following the inauguration.

Older students have more agency over their own attendance and possibly face more direct personal risk.

Event study plot showing weekly estimates of the attendance gap between foreign-born and U.S.-born students. The x-axis spans 20 weeks before and after inauguration. Pre-inauguration estimates cluster around zero with no clear trend. Post-inauguration estimates shift upward to approximately 2-3 percentage points and remain persistently elevated through the end of the school year, with no sign of attenuation.

Importantly, the effect isn't driven by spikes *around* specific raids or arrests in the local community.

It's a *persistent* elevation across the entire post-inauguration period.

Line graph showing monthly absence rates for the 2024-25 school year. Two lines track students born inside vs. outside the U.S. from September through May. Both groups follow similar patterns through the fall, but diverge sharply at a vertical dashed line marking the January 20th inauguration. After that point, foreign-born students' absence rates remain elevated while U.S.-born rates decline as expected, eliminating the pre-inauguration gap.

NEW PAPER: We study the effects of immigration policy on student attendance in 2025. Students born outside the U.S. were 2.2pp more likely to be absent after Jan 2025 (a 37% increase).

That's ~2 extra missed days per semester.

LINK: edworkingpapers.com/ai26-1453
@annenberginstitute.bsky.social

This is a clear area where public policy + public health can do some meaningful preventative work and it's frustrating that after 25+ years nothing has been done.

(I lived, played, and went to school inside of the Omaha Superfund site; it's massive)

Senior professors - if you want to help your junior colleagues in these times, I am begging you, review our papers. I have done 25 manuscript reviews in the last 6 years. But my own manuscript is stalled out waiting for reviewers. Relatedly, I won't be doing any more reviews until tenure. 🧪

Research Notes: Higher Salaries Increase Teacher Retention, Study Finds In 2023, Arkansas legislators passed the LEARNS Act, a comprehensive education reform package that, among other goals, aimed to strengthen teacher recruitment and retention. The law raised the…

Pay teachers more ➡️ increase retention

But we don't need to stick with a rigid salary schedule to do it

www.future-ed.org/higher-salar...

Rocks - The Geology of Virginia Rock (n.) – an inorganic solid composed of a variety of minerals that comprises the Earth’s lithosphere. There are three types of rocks, each with their own origins, mineralogy, and uses: *scroll over...

Related: geology.blogs.wm.edu/minerals-roc...

CALDER Webinar: The Four-Day School Week: New Evidence for District Leaders and Policymakers Featuring Emily Morton and Andrew Camp

📢 Webinar alert: What does the evidence really say about four‑day school weeks?

Join Emily Morton & @andrewmcamp.com next Wednesday, April 22, for a deep dive into the realities of four‑day school week implementation and its effects on teacher recruitment and retention.

Register: bit.ly/4rNbwP3