The test score myth and homeschooled students’ academic performance

A very common claim in conversations about homeschooling is that research proves homeschooled students perform better on standardized tests than their peers in school. This claim, however, is based on a few well-known studies that scholars have repeatedly debunked as unsound. CRHE has published in-depth reviews of these studies and their shortcomings (here, here, here, and here). Our key critiques are that: 

The studies don’t rely on random samples. Researchers recruited their participants by asking for volunteers within their networks — not by using random sampling. This means that their samples are not representative of homeschooled students as a whole, and that their conclusions cannot apply to all homeschooled students. In fact, the studies overrepresented white, wealthy, highly educated homeschooling families.

They confuse correlation and causation. Even when a study finds that homeschooled students have higher test scores, researchers cannot automatically conclude that homeschooling is causally responsible for these outcomes. Good parenting or high resources could potentially be responsible. The studies in question do not account for these variables.

They don’t control for other background factors. To truly compare homeschooled students’ performance to their peers in school, researchers would need to recruit participants from both groups that have similar backgrounds. Otherwise, they cannot isolate homeschooling as the only factor influencing performance.  

So, what does the research say about homeschooled students’ academic performance?

Credible research on homeschooling shows that homeschooled students can perform very well academically. Academic achievement, however, is closely linked with family background, influenced by factors like a parent’s level of education or the family’s socioeconomic status. This means that, while evidence shows homeschooling can lead to academic success, that success cannot necessary be attributed to homeschooling.  

Though more research is needed, current research points to some less optimistic trends in homeschooled children’s performance that often go unmentioned in homeschooling debates: 

The math gap. When family background is controlled for, studies have consistently shown that homeschooled children, on average, perform worse in math than their peers in school. While no one is sure what the cause of the so-called math gap is, some speculate that it is due to the fact that homeschooling is usually reading-intensive, and that homeschooled parents tend not to be as experienced in math (or the teaching of math) as other subjects. 

Low college attendance. While research suggests that homeschooled students generally perform well if they attend university, the best evidence we have from a randomized study suggests that homeschooled students may attend college at far lower rates than peers. More research on this subject is needed

Educational neglect. Little research has been conducted on educational neglect in homeschool settings, but a study of national data found that homeschooled children were two-to-three times more likely to report being behind grade level than peers. This does not necessarily mean that homeschooling itself led to these poor outcomes: students might have been behind before homeschooling occurred. State-level data do paint a worrying picture of homeschooling perpetuating ongoing educational neglect. A 2018 report from the Kentucky Office of Education Accountability found that nearly two-thirds of students pulled from school to be homeschooled were chronically truant prior to withdrawal. More research is needed, but this is a cause for concern.

Takeaways

While evidence undoubtedly shows that homeschooling can lead to high academic outcomes, claims that homeschooled students universally perform better than their school-educated peers are baseless. Evidence from peer-reviewed studies, and the testimonies of homeschool alumni, also demonstrate that homeschooling can lead to poor outcomes for homeschooled children. Both of these realities need to be acknowledged in debates about homeschooling’s effectiveness.