Do Homeschool Kids Really Rate Better on Standardized Tests?
Homeschooled students score about 72 points higher than the national average on the Scholastic Aptitude Test (SAT). The average American College Test (ACT) score is 21. The average score for homeschoolers is 22.8 out of a possible 36 points. Homeschoolers are at the 77th percentile on the Iowa Test of Basic Skills.
Advocates of homeschooling point to these standardized test results to show that not only does homeschooling work, it is also superior to public schooling. Critics disagree and say that demographic and other explanations may account for the higher homeschooling scores.
Critics point out that these higher scores do not indicate that homeschooling is superior to other forms of schooling. Some arguments are:
- Students who take these tests are self-selecting, so we don't know if homeschoolers overall are doing better than other students. But all SAT and ACT takers are self-selecting regardless of how they were schooled. However, there is no way to know if homeschoolers are taking these tests at a lower or higher rate than public or private school students.
- If broken down by demographics, homeschoolers may not fare so well. Homeschoolers tend to come from higher earning and better educated families, which may account for the higher scores.
- Sampling is sometimes done to compare homeschoolers to public school students. Critics point out that successful homeschooling parents may be more likely to allow their children to be tested than less successful homeschoolers.
The Demographic Argument
Johnna Burns of Northeastern State University made the demographic case in a 1999 study called "The Correlational Relationship between Homeschooling Demographics and High Test Scores." 1
According to Burns, homeschoolers are more likely to come from homes with educated parents and higher incomes. Homeschooling parents are less likely to divorce (which is true of higher income couples in general). Homeschooled kids watch less television. All of this results in higher academic achievement. As a result, Burns says that there is "inconclusive evidence of the actual quality of homeschool instruction."
A U.S. Department of Education study found that homeschooling parents are about twice as likely to have advanced degrees. But the percentages with Bachelors degrees or some college is similar to the population overall. Blacks and Hispanics are underrepresented among homeschoolers. 2
The Home School Legal Defense Association (HSLDA) commissioned a study called the "Homeschool Progress Report 2009." 3 This report did look at demographics. This study found that:
"Homeschoolers are still achieving well beyond their public school counterparts—no matter what their family background, socioeconomic level, or style of homeschooling."
Homeschoolers in this study were actually not better off or better educated, when compared to other households headed by two parents.
"Homeschoolers’ median family income ($75,000–79,999) closely spanned the nationwide median (about $79,000) for families headed by a married couple and with one or more related children under 18."
Homeschool households where neither parent had a college degree did less well than households with one or more college graduate parents. But the differences in academic performance were not significant.
"Students whose parents both had a college degree performed better than those who had no parent with a college degree. However, this correlation is generally weaker for homeschool students than for public school students.The homeschooled students whose parents did not have college degrees still performed at the 83rd percentile."
According to this study, homeschoolers have significantly higher test scores than the national average.
"Homeschooled boys (87th percentile) and girls (88th percentile) scored equally well; the income level of parents did not appreciably affect the results (household income under $35,000: 85th percentile—household income over $70,000: 89th percentile); and while parent education level did have some impact, even children whose parents did not have college degrees scored in the 83rd percentile, which is well above the national average for public school students. Homeschooled children whose parents both had college degrees scored in the 90th percentile." 4
Since the HSLDA is a group that represents homeschoolers, it is possible that there is bias in the study, which consisted of 11,739 participants. Critics can easily make the self-selecting claim that only parents who knew their children would do well would participate. It is highly unlikely that the HSLDA study will end this debate.
Problems with Critics' Arguments
The only way to know if homeschoolers outperform public school students is to look at similar sample populations. Critics typically won't accept these kinds of studies due to the concern that only students who will do well on the tests will take them. This assumes that homeschooling parents would somehow know in advance how well their children would actually perform. Many homeschooling parents may not know how well their children would do on any particular test. Homeschoolers typically take tests like the Iowa Test of Basic Skills as a means to gauge how well they are doing. However, it doesn't make sense to compare homeschoolers to the national average because the demographics of both groups are different.
The argument that homeschoolers may be taking the SATs and ACT at lower rates is problematic too. Considering that homeschoolers are more likely to come from homes with college graduate parents, they are most likely taking college entrance exams at a similar or higher rate compared to other students.
Do you think that homeschooling is academically superior to public schooling?
While it's easy to explain away standardized test scores and comparisons based on sampling, it's hard to dismiss homeschoolers' successes in other areas. Homeschoolers tend to be significantly overrepresented in spelling bees, geography bees and other academic contests. All participants in these kinds of contests are self-selecting and would most likely come from similar demographic backgrounds. Consider that almost 88 percent of American students attend public schools, about 10 percent attend private schools and approximately 2 percent are homeschooled.
Scripps Spelling Bee - In the 2009 Scripps Spelling Bee, 6 of the finalists were public school students, 2 were private school students and 3 were homeschoolers. In 2010, there were 8 finalists. Four were public school students, 2 were private school students and 2 were homeschooled. Public school students made up only about 50 percent of the finalists in both 2009 and 2010, while private school students made up 18 percent and 25 percent respectively. Homeschoolers made up 27 percent of the 11 finalists in 2009 and 25 percent of the 8 finalists in 2010.5
National Geographic Bee - No homeschoolers made it to the top 10 in the 2010 National Geographic Bee but three out of the 54 participants were homeschooled. Homeschoolers made up 5 percent of the participants. Of the 10 finalists in 2009, one was a homeschooler. 6 In 2007, homeschooler Caitlin Snaring became the first girl to win the National Geographic Bee in 17 years. 7
Apangea Math Contest - The second annual Idaho Math Cup was won by a class from the Idaho Virtual Academy, a public online homeschooling program. 8 Brother and sister Garrisen and Cecilia Cizmich beat thousands of students nationwide to win the Apangea Summer Contest. Both are homeschooled through the Idaho Virtual Academy. 9
3M Young Scientist Challenge - Two of the 10 finalists in the 2010 and 2008 contests were homeschool students. Homeschoolers made up 20 percent of the finalists. One of the 2009 finalists was a homeschooler, making up 10 percent of the finalists. Only 60% of the 2010 finalists and 50 percent of the 2009 and 2008 finalists attended public schools. 10
USA Mathematical Olympiad - This contest has 12 winners each year. In both 2010 and 2009, a homeschooler was among the winners. So homeschoolers made up 8 percent of the winners both years. Only 7 of the 2009 winners were American public school students (approximately 58 percent), as were 8 of the 2010 winners (approximately 66 percent). 11
If homeschooling did not provide any advantages over public schooling, public school and homeschooled contestants should make up a similar proportion to their representation in the student population overall. However, public school students are obviously underrepresented in these contests, while private school students are overrepresented and homeschoolers are significantly overrepresented. The success of homeschoolers in academic contests indicates that homeschoolers are most likely doing much better educationally than their public school counterparts.
4. National Finalists: 2010. National Geographic Bee, National Geographic Bee finalists announced
7. Nampa kids win national math contest from The Idaho Statesman
8. 2008 Young Scientist Challenge, 2009 Young Scientist Challenge, 2010 Young Scientist Challenge