Where your child is stuck is the whole brief

If you are trying to help one child through one stubborn thing, you do not have to write a curriculum request or name the skill and the grade. The place where your child gets stuck, described exactly, is the whole brief. Add what they love, and the recipe, the open set of prompts, templates and tools behind Meta DAX that builds a package for one guide and one learner, has everything it needs to make you a page. This opens a series of five for parents and tutors, so it starts with what to tell us.

What it looks like

Open the homeschool page and the first sample is about a made-up pair, Megan and Liam. Liam is eleven, introduced in one line: he “knows his baseball stats cold and wrote 1/2 + 1/3 = 2/5.” That wrong answer is the diagnosis: two-fifths is what you get by adding the tops (1 + 1) and the bottoms (2 + 3). What comes back starts from baseball: “An inning has six outs, so half an inning is three outs, a third is two, and a half plus a third is five outs out of six.” By the end of session one Liam said, “oh, it’s just outs.” His page for session four is open to try: one tab for Megan, and one for Liam that is a game.

Two more stuck spots, in the words on the samples page. Eli, carrying tens: “Asked to add 27 and 15, Eli produced 312: the tens total (3) and the ones total (12) placed next to each other, with no carrying.” Dash, reading: “Sounding out s-u-n out loud, Dash would then say a completely different word that started with s.” After about two misses in a row, Dash “would cover the page and stop.” Everyone in these samples is made up.

A line like “he’s bad at fractions” gives a subject and no way in; “wrote 1/2 + 1/3 = 2/5” gives the page its first step. What they love matters just as much, and it is not decoration: it is where your child is a step ahead of you. Your child is the guide on box scores, or dinosaurs, or marble runs, and the page is built through that, while you stay the guide on your child. “You know your child,” the homeschool page says. “That is the part no system can do.”

What the record shows

The offer asks for five things and no more: your child’s age; the subject and the exact spot where it keeps going wrong; what they love; what you have tried, and what happened; and how they learn, such as “hates worksheets, thinks best standing up, goes quiet when wrong, twenty minutes and then done.” The samples page says of what it shows: “Every one of these started as one sentence from a guide and what the learner already loves.” That is a record of how the pages were made, not a measurement. Nothing about Meta DAX has been measured yet. The first measurement will be real families and the first ten scenarios people send, which Jason Jeyanandan, who set the Foundation up, builds in the open.

In the groups where parents ask each other for help, the stuck spot usually arrives in a line or two. That is our own observation, not research, and it is why five short lines is the right size.

Research can say something narrower. A review of 96 studies, all randomized controlled trials, of one-on-one and small-group tutoring found an overall pooled effect size of 0.37 SD (0.37 of a standard deviation, a standard measure of how far test scores spread), which the authors call “statistically significant and substantively large”. The abstract adds that effects are “stronger, on average, for teacher and paraprofessional tutoring programs than for nonprofessional and parent tutoring”, and the paper notes there were no large-scale impact evaluations of parent tutoring. So there is good evidence that close, responsive help works, thinner evidence for parents giving it, and none yet that a page built from five lines does the same. A tutor in the room watches the pencil. A page cannot, so your five lines are the watching.

What to do this week

Write your five lines today, in your own words, in the order above. For the stuck spot, write what they wrote, said or did, not what you concluded from it. What they love can be small. No names and nothing that identifies anyone; a made-up nickname is enough. Then keep the lines. After each session you add one about how it went, and the fourth piece in this series is about what that does.

Then either paste one prompt into the AI chat you already use, which asks about a dozen questions you can answer from your notes, or send the lines through the guided questionnaire on the homeschool page, by replying where you found this, or to Jason on LinkedIn. What comes back is a page like Liam’s, one side for you and one for your child. The chat is yours; your child’s side is a game with no AI in it.

This piece opens the series, so nothing comes before it. The next, Read your page before your child reads theirs, is about the page that comes back and why you are its last check. How a showcase is made shows how a page like Liam’s is built from lines like these, and if you tutor other people’s children, Obsidian Delta’s piece on a tutoring practice with a kit for every hard topic covers the same open work from the company’s side.

what is the exact sentence your child says right before they stop? Not what you think it means; the words. Send it, with no name, where you found this.

Sources

• Nickow, A., Oreopoulos, P., & Quan, V. (2020). “The Impressive Effects of Tutoring on PreK-12 Learning: A Systematic Review and Meta-Analysis of the Experimental Evidence.” NBER Working Paper 27476. Abstract (0.37 SD; stronger for teacher and paraprofessional tutoring than for nonprofessional and parent tutoring): https://www.nber.org/papers/w27476
• Same paper, full text (96 studies, randomized controlled trials only; “statistically significant and substantively large”; no large-scale impact evaluations of parent tutoring programs): https://www.nber.org/system/files/working_papers/w27476/w27476.pdf