BigKids
The next exercise is the whole job.
A maths and English tutor for young children that reasons before it drills. In simulation, its lesson planner now gives a child a lesson they are not ready for about three quarters less often.
The idea
A good teacher watches a child for a minute and knows what to hand them next: not the page they have mastered, not the one that will make them cry. BigKids tries to make that choice from the child's answers alone.
Read the full story →Why it matters
A tutor that drills teaches a child to wait for the answer. One that picks the wrong exercise teaches them they are bad at maths. Both are failures of what comes next, so that is the part that gets tested.
Adaptive tutoringMathsWritingSimulated learners
How it works
- In
- What a child has answered so far.
- Out
- The next exercise: one they will mostly get right, with effort.
Closed source · results onlyHow a learner is modelled and how the next exercise is chosen are not published. The measured figures are in the full results paper, shared on request.
Explore deeper Each path is a different view of the same work.
Why it exists
A child learns from the exercise that is just hard enough. Choosing it is what a good teacher does by watching, and what most apps do not do at all.
BigKids is a maths and English tutor for children of about five to ten. Its tutor asks before it tells and never writes a child's sentence for them.
What it is judged by
The planner is tested on simulated learners whose knowledge can be seen. Its mistakes are counted: lessons out of order, exercises repeated too soon, runs of failure.
Against those sit time near the right difficulty and the share of exercises that ask for reasoning. One number combines them.
What it found
Lessons given before the learner was ready fell by about three quarters, and repeats by about a quarter. The combined score moved from below zero to above it.
Reasoning exercises are still only about a tenth of what a learner meets, and frustration is not yet low.
What it does not claim
Nothing here says a child learns more. The learners are simulated, and the planner was improved against the simulation's assumptions.
One school year of maths is covered in depth, and there is no comparison with another programme.
What was measured
Each number is read from the file named under it. Nothing here is rounded up.
private research log: status notesprivate research log: status notesprivate research log: status notesprivate research log: status notesCommon questions
Where are the actual figures?
In the full results paper: the measured counts before and after, and where the planner stands now. It is shared on request.
Has it been tested with children?
No child's learning has been measured. The figures are about a planner teaching simulated learners.
Does the tutor give answers?
It asks first, then hints, and tells only after the child has really tried. It never writes the child's sentence.
Can I see it?
Ask through the request form.
Ask a question
Ask about what it does and what it measured. Answers come from what is written on this site.
Carry thisOne exercise does not teach a child. The choice of the next one does.
proprietary · the result is public in relative terms · the figures on request