Journal · Scaffolding · September 8, 2026

Do-With, not Do-For

The most common mistake I see is families skipping the middle rung of the scaffolding gradient. Everything collapses back to Do-For or catapults up to Child-Does. Neither builds the skill.

The most common mistake I see is families skipping the middle rung of the scaffolding gradient. Everything collapses back to Do-For or catapults up to Child-Does. Neither builds the skill.

The gradient has four positions: Do-For (the adult does the whole thing). Do-With (adult and child, shoulder-to-shoulder, adult narrates). Watch (adult is present, silent, available). Child-Does (the destination).

Most families I coach have exactly two settings — Do-For (when the child is stuck) and Child-Does (when the parent has given up managing the moment). The transition between them is where the friction lives.

Do-With is the setting that actually builds the skill. You are sitting next to your kid. You are narrating the moves — ‘okay, first I'm reading the whole prompt out loud once. Now I'm underlining the verb. Now I'm writing my first sentence.’ The child is watching a real person do a real cognitive move in real time.

This is not enabling. This is not doing it for them. This is the only setting in which the underlying strategy transfers.

Try this: pick one friction domain. Sit down next to your child. Narrate one instance of the strategy out loud. Then release to Watch position for the next instance. Then Child-Does the third one.

Three iterations. Three positions on the gradient. In one sitting.

— Scott Adams

← Back to the Journal