Rodger Williams
October 1, 2024
A recurring criticism of the nationwide studies on homeschool academic achievement — conducted by Dr. Brian Ray and the National Home Education Research Institute — is that participants were self-selected. That is, the data was not randomly selected from the entire population of homeschoolers.
The idea is that parents of students with lower test scores would be likely to decline to report scores and thus bias the estimated average score higher.
This is, in fact, a theoretical possibility. The question is, did it actually happen. We now have evidence that it did not.
The data from these studies (1996, 2008 and 2023) shows a consistent pattern: Academic achievement of homeschoolers is approximated by a normal curve, just as public school achievement is.
There is no indication in the data that there were more high scores or fewer low scores among the reported homeschool test scores compared with the normal curve.
The data collected in these studies is not skewed by participation self-selection.
