The problem I noted is that 99% is nonsense when your sample size is 40. 39 out of 40 kids is 97.5%. The next gradation, all 40 kids, is 100%. The small sample

The problem I noted is that 99% is nonsense when your sample size is 40. 39 out of 40 kids is 97.5%. The next gradation, all 40 kids, is 100%. The small sample size is also a concern, of course.