Thomas S. Dee

Know any aspiring PhD students interested in quant studies in education? Please direct them to our programs in ed policy, econ of ed, or ed data science! I hope to welcome an energetic & ambitious student to join me in our new building 🌴☀️
Applications due November 17
ed.stanford.edu/admissions/a...

🚨Do you know a school, system, or network doing amazing work in support of student success in high school, particularly through a continuous-improvement approach?

Let's recognize them! Encourage them to apply for the Carnegie Award for Impact (cfdn.at/4o4gp58).
Applications due on November 10.

I have watched states delay releasing test scores for months as they work to message results and hope no one is paying attention.

Teachers sue over Trump's immigration crackdown, saying students are staying home Labor unions representing educators and school employees are suing President Donald Trump’s administration over its immigration crackdown

It's gratifying to see my recent immigration-raid research actually being used.
wapo.st/42mWdTn

High Quality Research Rarely Informs Classroom Practice. Why? (Opinion) The connection between education research, policy, and practice is broken. Here’s what it would take to fix it.

"Educators should also understand the limitations of research evidence. Rigorous impact evaluations...can only tell us what has worked someplace at some time—not what will work universally for educators facing specific challenges." - @tomdee.bsky.social
www.edweek.org/leadership/o...

High Quality Research Rarely Informs Classroom Practice. Why? (Opinion) The connection between education research, policy, and practice is broken. Here’s what it would take to fix it.

Why aren't practices based on research implemented in schools? @tomdee.bsky.social has insights, and ideas for change. www.edweek.org/leadership/o...

High Quality Research Rarely Informs Classroom Practice. Why? (Opinion) The connection between education research, policy, and practice is broken. Here’s what it would take to fix it.

"...embrace opportunities for quick and meaningful wins. These are the common-sense education initiatives that share three critical features: evidentiary support, low financial costs, and scalability." - @tomdee.bsky.social www.edweek.org/leadership/o...

High Quality Research Rarely Informs Classroom Practice. Why? (Opinion) The connection between education research, policy, and practice is broken. Here’s what it would take to fix it.

Just out in @edweek.org's #K12BigIdeas, I criticize the quality & connectedness of education research, policy, & practice and discuss what it would take to do better.

www.edweek.org/leadership/o...

Thanks for flagging that. There seems to be some agreement on the role of post Great Recession spending declines?

I'm curious how people view hypotheses about these achievement declines. I incline towards (1) the failure to refine & sustain school accountability & (2) the rise of smart phones & social media. But that's conjectural without better evidence.

Have we given too much "credit" to the pandemic for learning losses? Almost 50% of the post-pandemic decline in achievement on 8th grade NAEP was to be expected (without the pandemic!), simply using the same rate of declining achievement that began unfolding starting in 2013!

The @nytimes.com spotlighted a new study from #EdWorkingPapers that examines the profound impact of immigration raids on attendance in California's Central Valley.

Read the NYT article below.

‘There was a lot of fear’: Central Valley immigration raids drive up absences in schools, study finds Immigration raids in January and February increased absences by 22% in five school districts, according to a newly released Stanford study.

Immigration raids in January and February increased absences by 22% in five school districts, according to a newly released Stanford study.

🚨New research from SIEPR Senior Fellow @tomdee.bsky.social finds that student absences rose amid intensified immigration enforcement, indicating lost learning opportunities. 👇
bit.ly/3I1vGDJ

Immigration Raids Add to Absence Crisis for Schools

I appreciate the work of @wtgrantfdn.bsky.social grantee @tomdee.bsky.social showing student attendance, already in crisis after COVID, is further disrupted by immigration raids www.nytimes.com/2025/06/16/u...

Immigration Raids Add to Absence Crisis for Schools

The current wave of federal immigration raids began in January with a “rogue” operation in California’s Central Valley.

My new study finds these raids increased school absences by 22%—a leading indicator of the resulting family stress & lost learning opportunities👇

www.nytimes.com/2025/06/16/u...

The gift of time? School starting age and mental health Using linked Danish survey and register data, we estimate the causal effect of age at kindergarten entry on mental health. Danish children are supposed to enter kindergarten in the calendar year in w...

My research w/ @hhsievertsen.bsky.social indicates redshirting benefits kids lacking the self-regulation needed to thrive in more structured classrooms.

That argues for kindergarten-readiness assessments that inform targeted—not universal—responses. onlinelibrary.wiley.com/doi/10.1002/...

Should Boys Start Kindergarten a Year Later Than Girls?

I spoke with the NYT for this article. I'm unconvinced that the universal kindergarten redshirting of boys is a good idea...

www.nytimes.com/2025/06/14/u...

Seen on the streets of Paris today. #EdPolicy is everywhere.

This provocative, data-infused essay by James Wyckoff — edworkingpapers.com/ai25-1197 — serves as a reminder that fully understanding the declines in student achievement requires us to look well before the COVID-19 pandemic.

Citizen Stewart Show - The Branch

Wow the latest episode of the @citizenstewart.bsky.social show nails it when it comes to the evidence on merit pay. It's hard to get it right (implementation details, supports, etc.), but it can move the needle on student achievement.

But 👇
www.thebranchmedia.org/show/citizen...

Key to Improving America’s Schools: Rethinking School Staffing & Teacher Quality The status quo for in-service teachers — low-stakes assessments & rigid single-salary schedules — needs to change if student outcomes are to improve.

It's frustrating that The Discourse often frames teacher evaluation/pay reforms as failing when a more accurate take is that they are politically & logistically difficult to implement well but effective when this is done. I try to describe this literature here:

www.the74million.org/article/40-y...

I'm surprised that that training teachers to use "three-cueing" in their early-literacy instruction is not yet on the list. Too obvious? Or still controversial? #ScienceOfReading

That's so kind of you!

I'm glad to know others share my concerns about our current state of practice and recognize that we can do better without throttling our opportunities for discovery and insight.

I was so pleased to cross the Bay today and speak at the Berkeley Initiative for Transparency in the Social Sciences (www.bitss.org) about the need for (and feasibility of) adopting preregistration of studies making causal claims with observational data. #OpenScience

Something rotten in AI research Why so many shady studies?

"With Silicon Valley in full AI hype mode, we need independent studies of AI in education that we can trust to fairly evaluate this technology’s impact, positive or negative. We’re not getting that right now and it’s a problem."

open.substack.com/pub/buildcog...

Thanks! IMHO, this is a fundamentally important issue. But we also can't simply "cut and paste" RCT preregistration practices.

For example, we need new standards (e.g., preregistered decision trees? outcome-blind exploration?) relevant to the often messy, uncertain analysis of observational data.

Causal inference friends, check out The Case for Preregistering Quasi-Experimental Program and Policy Evaluations by @tomdee.bsky.social

tom-dee.github.io/files/Dee_Ev...

YouTube video by Cognitive Resonance AI Will Not Revolutionize Education

In April 2025 I delivered a speech at the ASU+GSV ed-tech conference titled "AI Will Not Revolutionize Education." It touches on human cognition, gen AI, and the nature of scientific and social revolutions. I worked hard at this, I hope you'll watch and share.

www.youtube.com/watch?v=u0_t...

A study by Professor @tomdee.bsky.social and GSE doctoral student Daniela Ganelin suggests that access to CS Principles could account for roughly two-thirds of the growth in AP computer science exams taken nationally over the past decade: ed.stanford.edu/news/stanfor...

#EduSky #AcademicSky