Docs How it works
What goes in before the reply.
A model on its own replies generally: it gives the answer, explains too much, and answers everyone alike. The Harness works out what this moment needs from what a teacher already decided. Change the setting or the message below, and see what the model is told. This is the Harness’s real output, and no model is called.
A teacher set up one activity
Learners rewrite fractions with a common denominator before adding or subtracting, and can say why the denominators can’t just be added.
And the short list they chose from 9 pieces
- Believes that fractions are added by summing numerators and denominators as if they are separate whole numbers Eedi
- Believes that when multiplying fractions, the operation on the denominators is addition rather than multiplication Eedi
- Multiplies the numerators when adding fractions with a common denominator Eedi
- Say what is right or wrong and why, then one thing to do next. A bare "correct" or "not quite" teaches little. Wisniewski et al. 2020; Van der Kleij et al. 2015
- Comment on the work and the method, never the person: name the step or strategy, and leave out how clever or able they are. Kluger & DeNisi 1996; Hattie & Timperley 2007
- When you criticize work, say why: you hold it to a high bar because you believe they can reach it. Then say how to get closer. Say this only when the next step really is within reach. Yeager et al. 2014; Cohen et al. 1999
- Find out what they can already do on this step before you help, then fit the help to that rather than to the next item in your plan. van de Pol et al. 2010; Wood et al. 1976
- Give the least help that lets them take the next step themselves: a nudge to look again first, then a pointer to the part that matters, and a fuller hint only if that fails. Wood et al. 1976; Koedinger & Aleven 2007
- Draw their attention to the feature of the problem that matters, and let them work out what to do about it. Wood et al. 1976
A learner writes
How it reads the moment: They gave an answer or some working. By rules, with no model call.
What the model is told
How the teacher wants you to help in this activity:
Goal: Learners rewrite fractions with a common denominator before adding or subtracting, and can say why the denominators can’t just be added.
Standard: 5.NF.A.1 (Multi-State, grade 5) Add and subtract fractions with unlike denominators (including mixed numbers) by replacing given fractions with equivalent fractions in such a way as to produce an equivalent sum or difference of fractions with like denominators.
How far to go: Give hints and guiding questions, one step at a time. Don’t give the answer or do the work for them.
For this turn (how far to go still comes first):
Right now: The learner has given an answer or some working. Work it out yourself first, without showing it. If it is wrong, don't say it looks right: find out how they got it, or point to the one step to look at again. If it is right, say so and ask them to say why it works.
From research, where it fits: - Say what is right or wrong and why, then one thing to do next. A bare "correct" or "not quite" teaches little. (Wisniewski et al. 2020; Van der Kleij et al. 2015) - Comment on the work and the method, never the person: name the step or strategy, and leave out how clever or able they are. (Kluger & DeNisi 1996; Hattie & Timperley 2007)
The learner’s words aren’t in it, so what a learner types can change what is chosen, but can’t write instructions.
Why it needs no search
A person narrows everything down first. A teacher, or an assistant working with one, picks a short list for an activity, from every state’s standards and thousands of misconceptions. Each turn then only chooses from that list, by rules or a small model. A list of a dozen items needs no index, so the Harness has no vector store, no server and no model of its own.
Retrieval is for your own documents, such as a library’s guides. It belongs to whatever holds them: the Engine’s knowledge and the Developer Platform’s use keywords and meaning together.
How well it works so far: finding the right misconception for a message worked on only 4 of 19 turns. That is the weakest part, and the evaluation looks at it first. What was measured
| What | How much, and where | What reaches a turn |
|---|---|---|
| What the teacher decided | One activity’s settings | The goal and the limit, every turn |
| Teaching moves | 4, each from a study | One |
| Misconceptions | 8,117, carried in the package | The teacher ticks up to 8. At most 2 reach a turn |
| Guidance from research | 77 pieces on 8 topics, carried in the package | A person picks up to 20. At most 2 reach a turn |
| Standards | About 1 GB for every state, fetched from Learning Commons when asked | The teacher picks one, and at most 2 of its components reach a turn |
| Your own documents | Whatever you have, in your agent’s or the Engine’s search | What search finds |