Director with @TNTP & father of 2

Joined November 2011
Follow Marcy Stein for DI math and reading
Replying to @MrZachG
Incredible work digging through the archives. This is the story of Follow Through that should have been told decades ago.
5
7
64
6,442
Mickey retweeted
We’re opening a research preview of Claude Duet, a new app where Claude works as a proactive agent. It comes to you with what needs your attention, completes tasks you ask of it, keeps working when you step away, and checks with you before it acts on your behalf.
2,351
4,220
4,719
55,083
10,683,854
You can hear from my co-author @ColleenDriggs and one of my all time favorite reading teachers, Jaime Johnson, in this article on "re-centering" reading--making it an in-class, communal activity. edweek.org/teaching-learning…
2
6
35
3,578
Mickey retweeted
“Critical thinking” without knowledge is just confidently not knowing what you’re talking about.
10
29
7
116
3,771
Something I hear over and over when I’m visiting schools: “They were really engaged.” Fine. What did they learn? Can they explain it, use it on a new problem, and remember it next week? Keeping kids busy is a low bar. We’re supposed to be teaching them.
36
Read a discussion on this recently where the author agreed but made the point that we moved to skills because it was easier than teaching political consensus on what knowledge to teach.
"...a pedagogy focused on supposedly generalizable skills like 'critical thinking,' rather than on the acquisition of knowledge, is the chief culprit for the stagnant American performance in the National Assessment of Educational Progress as well as PISA." Yup. jamestraub.substack.com/p/th…
1
75
Does strategy instruction matter? I’m not convinced. We know that, ahem, knowledge matters. We know that foundational skills matter. There is evidence that teaching strategies can be beneficial but is it the best use of our time?
Strategy instruction matters. But strategies can’t replace what a student doesn’t know. Building vocabulary + background knowledge strengthens reading, oral language AND learning across content areas. 🔗: harriettjanetos.substack.com…
1
1
128
Direct Instruction I know, definitions are *boring* but it's worth just being clear about this. 'Direct Instruction' with a capital 'D' and a capital 'I' generally refers to the programs developed by Zig Engelmann, Doug Carnine and others and available from McGraw Hill. These are modern versions of the programs tested in Project Follow Through. Teachers follow a script that includes the students' opportunities to respond. The programs can be used in small groups or whole classes. For example, at my school, we teach Spelling Mastery to all students in regular English classrooms, but we use the other programs for small group sessions with our Progress Support Team. The materials are quite costly. So, there's a business idea for someone who can devise something of the same quality. Direct Instruction is therefore a highly specific kind of explicit teaching. Confusingly, explicit teaching is often called 'direct instruction' with a lower case 'd' and 'i' and that's why I avoid this term.
14
31
5
159
11,901
Apparently, I was early. On Feb 5, 2024, I wrote: “Science of Reading had its moment. It’s time for direct instruction.” More than two years later, that call feels timely. Cards 1–4. Read the blog + subscribe: thriveedk12.com/ideas
1
26
@Doug_Lemov 🤝 @ehanford Excited to see these two leading our field toward the evidence. I can’t wait to see what comes next as we grapple with what effective teaching looks like in practice. features.apmreports.org/sold… xitter.cf/Doug_Lemov/status/2105…
Excited to reveal given all the conversation about it today that I will have a new book out next year on Explicit Direct Instruction. Full of great videos of real teachers.
23
Yep. This.
"Math needs a phonics-style reckoning: Teachers should teach students how to do stuff correctly." - @mattyglesias slowboring.com/p/math-needs-…
1
24
The evidence for teacher-created curriculum is weak, and the evidence for the systemic use of quality curriculum is strong. A national conversation about this would be welcome.
What are the most effective methods of improving student achievement? An NYT investigation. For me, the data point in the bottom left quadrant is absolutely critical to understanding why student achievement goes wrong so often: teachers making up their own curriculum. nytimes.com/2026/09/28/upsho…
45
25
9
257
17,371
Preach
People have heard about the phonics vs triple-queuing fiasco but there is a similar issue in math where ed school ideologues have stigmatized the normal and necessary work of drilling novice students in the fundamentals. slowboring.com/p/math-needs-…
35
Can’t wait to read it!
Excited to reveal given all the conversation about it today that I will have a new book out next year on Explicit Direct Instruction. Full of great videos of real teachers.
3
75
Lots of discussion out there about our next model(!), so I wanted to give an early look as soon as possible. Introducing Gemini 4 Argon! It shows frontier performance in complex workflows, cyber defense and software engineering. Teams are using it extensively at Google, from coding to quantum computing, great feedback. Here’s a look at the benchmarks:
1,385
2,306
1,225
25,879
3,919,382
Initially I was a little skeptical of this, but I do think we need models for thinking through this so I’m tentatively open minded.
How much AI should students be allowed to use on an assignment? Zero AI. Consultative AI. Full AI use. Giving students a clear level for each task makes expectations around AI in education much easier to understand.
41
Can’t tell you how many PD sessions I sat through where we were told to essentially avoid recall/knowledge questions.
It's difficult to overestimate how damaging this view of teaching and learning has been for so many kids. Virtually every teacher trained over the last 50 years was taught that knowing stuff is the least important part of thinking. This image taken from Utica University's assessment resources for faculty: utica.edu/academic/Assessmen…
1
63
Haven’t tried this yet, but if it works well it’ll be a huge QoL upgrade and just plain nice.
Claude Code will now try to find a graceful stopping point when you hit your 5-hour limit mid-task, instead of cutting off mid-edit. It gets a small, fixed allowance pulled from your weekly limit to wrap up what it can.
28