The pilot study opens once its plan is locked

Earlier today we said families could join our pilot study. We have paused that for a short while, and we want to explain why.

A fair study writes down, before anyone joins, exactly what it will measure and how the results will be analysed. If you decide those things afterwards, it becomes too easy to pick whichever measure happens to look good. Our own build plan requires this step, and we had skipped it.

So the study now needs a locked plan before it opens. The plan names the main result we will compare: the 12-question check-up, two weeks after the 8 weeks. It also lists the other things we will track, how we will count families who leave, and the exact version of the analysis. And it says plainly that 60 children is enough to test the process, not to prove the product works. Once saved, it can't be edited. A different plan would mean starting a new study.

We have also double-checked the check-up questions. We hold back one question in each of our 22 topics. Any held-back question a child could already have seen in practice has been swapped for one nobody has seen.

The study invitation will reappear on your child's page as soon as the plan is locked. Nothing about your child's practice has changed in the meantime.