Jon Wartes has presented another piece of well-executed and reported research. The methodology is clearly explained in the report and he carefully presented considerations of sampling error in a special section entitled “For other researchers — some information about sampling error in homeschool research” (p. 19-22).
Wartes reported that the home schoolers scored as well or better than their national peers on the Stanford Achievement Test series. Then, he quickly added that “This data should not be used to make homeschool – conventional school comparisons” (p. 13) and gave four explanatory reasons for not doing so. One reason was that Washington State norms (which were somewhat above average) should be used for comparative purposes. Unfortunately, he did not report the state’s norms. Another reason he gave for not using the data to make home school-conventional school comparisons is that it is difficult to determine whether the higher scores are a result of the educational delivery system (home schooling) or some other variable such as parental support. I would mention that it would be very difficult to separate out these two variables in educational research. In addition, many consider extensive parental support to be an integral facet of the home education “treatment.” The distinction may be worth considering in future research.
Wartes has provided an appropriate presentation of the data, along with a thoughtful analysis of the project’s findings. The research was appropriately executed by a group of people who are apparently dedicated to gathering and presenting factual and valid information about home schooling in the State of Washington.