Jon Regino

One of the most well thought out conversations about AI and education by @mrbartonmaths and @adamboxer1. You should actually check out @mrbartonmaths whole AI series if you work in education. pca.st/podcast/9fe42a…

From our latest substack article open.substack.com/pub/scienceofl…

Yeah, call me a Luddite if you want, but this is crazy: Some Lower Merion parents want to ‘opt out’ of Chromebooks in classrooms. District says they can’t. inquirer.com/education/lowe… via the great @maddiehanna

‘Differentiation’ does not work. It does not work as a term to describe something because it means so many different things, some of which are opposite to each other. For example, a writing scaffold can be described as ‘differentiation’ and allowing a child to record audio instead of writing can be called ‘differentiation’. The scaffold addresses a need but the audio recording accommodates it. These are opposites. We can go to a department meeting, talk about ‘differentiation’ and mean completely different things. Unlike inquiry learning or explicit teaching, there’s no superior version of differentiation that advocates stress that is competing with a weakened version caricatured by critics. Passionate advocates call for forms of differentiation that simply lack evidence. Yes, we all differentiate to some extent. But the word isn’t helpful. And it causes us to focus on difference, whereas the most powerful strategies often focus on commonality.

Sitting down to speak with @rastokke recently, I think I found the simplest way to explain the distinction between a small steps approach, and an atomisation approach. Small steps is a great idea, I’m all in favour it. It helps students master mathematical processes. Atomisation both helps students master mathematical processes and also mathematical concepts. It builds the connections between mathematical ideas, reduces what students must learn by linking everything together, and ultimately develops more flexible knowledge that can be applied in more novel, non-standard situations. So in this post we look at two concrete examples of how a first set of exercises for students would look different if they were governed by small steps versus atomisation. The atomisation approach helps students to generalise much faster while still building that same procedural automaticity. Importantly, one does not preclude the other. It’s not so much that atomisation is ‘better than’ small steps, it’s that it reveals additional options. Small steps is essentially focusing on a narrow subset of what atomisation makes available to us. link below 👇

<cough cough> iReady "isn't personalized by question type. That is the funny thing. It is personalized by score. If you get a score of say 458 on numbers and operations or vocabulary, you are placed at the same lesson as all the other kids with 458."

Paging educators 🚩🚩🚩 This is the second time in 2 weeks that @JonHaidt has called to get Ed Tech out of classrooms by September: "I hope that many elementary schools will remove devices by Sept., or create an analog path for parents who want that. Let's see if kids do better with books and paper" Haidt drove the cell phone ban legislation. What are the odds his words carry in a similar fashion? Are we doing enough to signal this shift to districts? 1/2

A snippet from a recent conversations with @learnwithmrlee about timed math fact practice and how it looks when done in line with the empirical literature. @rscodding @rastokke

Today, we're introducing Spectre I, the first smart device to stop unwanted audio recordings. We live in a world of always-on listening devices. Smart devices and AI dominate our world in business and private conversations. With Deveillance, you will @be_inaudible.

@VirginiaJim @Doug_Lemov For secondary math Sadlier has a similar lesson style to Saxon.

@Doug_Lemov Does anyone have a recommendation for a math curriculum that follows those principles? When I have had full control, I have chosen Saxon, but the fully spiraled nature and the integrated geometry doesn't work well in a large department.

The number of math curricula that are not aligned to cognitive science—that presume discovery learning as a design principle—is a quiet scandal in the US. Talking to a great district last week. They take the science seriously. They can barely find a math program that doesn’t design daily lessons around novice learners guessing inefficiently at solutions.

The anti-Ed Tech and pro-book movements are gaining momentum, especially for elementary grades. Let's get the memo to the nation's curriculum selection committees, pronto. They're making curriculum choices right now that schools will live with for the next 5-7 years. My latest for @CurriculumIP: Feat. @JonHaidt @natwexler @mark_bauerlein @KelseyTuoc

🎯 Why teaching multiple strategies during acquisition does not work by @amandavande1 "Logically...presenting a task using totally different problem formats or set-ups & then teaching the student 2-4 ways to solve a problem at the same time..is likely to produce confusion" springmath.org/202602-multipl…

🚨For math ed folks, the latest @rastokke podcast w @sarahpowellphd is a can't miss. Major math ed org (@MathEdLeaders)'s statement takes aim at evidence-based math instructional practices - the episode breaks down how misleading it is. Summary below 🧵 chalkandtalkpodcast.podbean.com/e/science-of-m…

And this is why education is stuck in such a rut, and constantly regresses back whenever we make an inch of improvement: because the progressive tropes are so intuitively attractive, easy, and simple, and the reality of learning is much more complex and difficult. You don't learn very well by watching an amusing or entertaining video; you retain the information in it like a river running through you. You're aware of it temporarily until it is replaced by the next packet of facts, but you don't retain it well or in a way you can retrieve easily later. You might remember novel, surprising, amusing, or shocking parts well enough, but miss the details that weren't obvious, or the links between them. Knowledge and skills are acquired through scaffolded explanations/ demonstrations followed by the student processing it, using it, thinking about it. The teacher then checks their understanding and offers high quality feedback, reinforces, corrects, redirects or reconstructs. Then the next lesson connects meaningfully with the last one in a similar way, infused with retrieval and revision of prior concepts to ensure deep learning. Videos can be part of that, but they are not that. We actually do know a lot now about how we learn, and how we teach, and collectively they tend to be called the Learning Sciences, or Evidence-Informed Education. It's urgently required across the world as an antidote to both the winner-takes-all grindhouse of poorly led-lecturing, or more commonly, the inane performative pantomime of progressive flim-flam. Mr B's references here demonstrate the old Keynesian adage of how so-called practical men who believe themselves free from bias, are usually in the grip of some long dead economist or philosopher. He trots out the most pedestrian and reactionary dogma about learning while believing himself to be a common-sense revolutionary. It's not his fault. I don't know anything about being a social media megastar. But tech, no matter how shiny, cannot replace the architecture of the human brain. There is a 1300g bag of neural porridge inside every one of us that isn't going anywhere fast, so we better get busy using it to understand how to replicate and build on how we already actually learn, rather than what we wish it was like- or what sells content.

Holy moley, this disconnect. We badly need a national discussion about math. Every educator I know believes schools need to focus more on structured math foundations (all kids knowing math facts, practice in early math to ensure fact fluency). Superintendents haven't received the memo. "The gap between what ed leaders think is wrong with math instruction and what families think is still…substantial. An EdWeek Instagram reel out this week has superintendents around the country talking about how math needs to be more “hands-on” and “ relevant” just as a new, 23,000-parent survey out from education reform org 50CAN reveals something much more surprising. Buried in the data of this wide-ranging parent survey that asks about everything from support for school choice to sports and the arts, is something every school leader should know about families’ relationship with math: Out of the 20% of students who are currently receiving tutoring… 80% are getting tutored in math. That’s nearly all of them." From @HKorbey's latest: HT @FiftyCAN @marcportermagee @NedStanley @lizcohen12

Study: A brain region specialized for recognizing text is smaller or absent in kids with dyslexia. Tutoring that improves their reading partly closes the gap. ‘“In kids with persistent reading struggles, we see that this key part of the brain involved in rapid word recognition is either missing or much smaller.” Among kids who had a detectable VWFA, the region was smaller on average in children with dyslexia than in typical readers. The size of each child’s VWFA was linked to his or her reading ability: Kids with a smaller VWFA had weaker reading skills. The neural response to seeing words — how strongly the region “lit up” during reading — was also weaker in kids with dyslexia than in typical readers. After the initial brain scans, some children with dyslexia received the reading intervention, an intensive program that enabled them to improve their reading levels by about one grade level over eight weeks. This group’s reading scores improve significantly, while over the same period, kids with and without dyslexia who did not receive tutoring showed no change in reading ability. After the reading intervention, the researchers could detect the VWFA in more children with dyslexia, while the chance of detecting this area in kids who had dyslexia but had not gone through the reading program was unchanged. “It’s as if evidence-based intervention builds this region in the dyslexic brain,” Yeatman said. The VWFA also grew larger in children who had completed the reading intervention; it did not grow larger in the control group.’

So disappointing. At exactly the moment we need ACT & SAT to embrace their role as objective arbiters of high standards they have abandoned the field. In England a national board sets standards for GCSEs (de facto college entrance exams). In the US individual states should do the same. Low standards & lack of meritocracy are issues of national security & long term economic viability.

What in the Sam Hill?! The ACT just quietly nuked their standards, and no one seems to have noticed Fewer questions, more time per question, fewer choices, shorter & fewer reading passages

Things that are medium effective are more effective than things that are ineffective. But medium effective things are not better than highly effective things. This concept needs to be understood when looking at popular programs that are not great, but simply better than nothing.

🚀Dropping this Friday: A look at what the NCSM statement gets wrong about the Science of Math with guest Sarah Powell. An episode about evidence, misconceptions, effective instruction, and why this matters for students.