Interesting chart.
They really did a bang up job putting it together.
I would prefer a different presentation of the percentile chart.The gloss over the average statistic when comparing homeschoolers to percentiles" of P.S. students.
Granted that there may be a shortage of data to break out the detail Im curious about for College-degreed and non-degreed parents and other things they control for, I dont get the impression that they are comparing homeschooling children of college-educated parents to PS-educated children of college-educated parents, for example. And that is actually what is significant to parents who of course either are or are not college educated, and know it, at the time they decide to or not to homeschool.
But I guess thats a problem with any statistic along those lines. The people who homeschool are a self-selected group. Some tried it, and didnt like it or found they couldnt do it. The only true statistic would come from parents who flipped a coin to decide whether to homeschool or not. And that aint happening, at least not very much.
But I guess that the thing that would be most interesting would be to graph out the entire distribution for homeschooler children of college-educated parents and superimpose it on the entire distribution for PS educated children of college-educated parents. The trouble with comparing to all PS children is that so many of them, and even of their parents, are indifferent to education. Thats a bad comparison, since none of homeschool parents are that way.