As the post-Covid homeschooling boom continues, the practice is receiving renewed media scrutiny. In the past year homeschooling has been critiqued by, among others, TV host John Oliver (who acknowledged that “for some kids, getting to be homeschooled can be genuinely transformative” but also called for its regulation), a Washington Post series titled “Home-School Nation” (including pieces on the movement’s increasing diversity and a critique of homeschool researcher Brian Ray’s work), and an article in The Nation about homeschooling and mental health. These critiques, like the one Elizabeth Bartholet published some years before Covid, focus on the conservative Christian ideology within the homeschooling movement and charge homeschooling families with a lack of academic rigour.

Public critiques of homeschooling often ignore the fact that the decision to homeschool is not a choice made in a vacuum. Many parents homeschool because of concerns about the quality of the public school system, concerns shared by the broader public. This April, the Pew Research Center published a poll reporting that about half of Americans think the public K–12 system is headed in the wrong direction, with an additional third being uncertain about its direction. In a separate Pew poll regarding school impact, 53 percent of respondents thought that K–12 schools were still having a positive influence on the country, down from 61 percent who said the same thing in 2021. Nor is it merely parents who are worried: more than half of K–12 teachers surveyed by Pew this year said that they would not recommend teaching as a career, most citing the stressful and overwhelming nature of the job and an understaffed work environment. As for the students themselves, they seem to be voting with their feet, in the most morbid way possible: A 2022 working paper from the National Bureau of Economic Research analyzed data from the last thirty years and found that being out of school (e.g., for summer vacation) resulted in lower teen suicide rates. Lyman Stone summed up the data provocatively on X: “School makes kids commit suicide.” (While research also suggests that teen mental health deteriorated during the pandemic, it is possible that this had more to do with the pandemic itself than with being unable to attend school.)

As more parents question the value of schools post Covid, they are seeking out alternatives, including charter schools. The National Center for Education Statistics (NCES) found that while public school enrolment fell by 4 percent during the 2012–2022 decade (from 47.3 to 45.1 million students), public charter school enrolment increased by 65 percent (from 2.3 to 3.7 million students). Notably, even as charter schools capture a slightly larger share of the student population, the total share of ordinary public and charter school students still dropped from 49.6 to 48.8 million.

Some of those missing students are in private schools. In 2021 (the most recent year with available private school enrolment data) K–12 enrolment had risen 5 percent since 2021, from 4.5 to 4.7 million. If one combines private school enrolment from 2011 and 2021 with public and charter school enrolment from 2012 and 2022 (a statistically dodgy move, but enough for a rough estimate), it appears that there were 54.1 million students enrolled in some traditional form of schooling in 2011–2012 and 53.5 million in 2021–2022. During that same decade the US population of five- to seventeen-year-olds grew from 53.7 to 53.9 million. If a country with 53.7 million school-aged children yielded 54.1 million traditional students in 2011–2012 (given, e.g., some four-year-olds in kindergarten and some high school seniors eighteen and older), one would expect to see about 54.3 million traditional students in 2021–2022 (53.7 / 54.1 = 53.9 / x, where x = expected enrolment in 2021–2022). Instead, there are only 53.5 million.

In other words—bearing in mind that this is only an estimate—there were possibly 800,000 more homeschoolers in 2021–2022 than there were ten years before, bringing the total number of homeschoolers up from about 1.8 million in 2012 to about 2.6 million in 2022 (not counting population growth among the original 1.8 million homeschoolers). This would suggest that about 4.6 percent of the children being educated in 2022 were being homeschooled, an estimate consistent with the NCES study finding that in 2019, pre-pandemic, 3.7 percent of students aged five to seventeen “received instruction at home (either homeschooled or in full-time virtual education).” (See at NHERI an alternative approach to estimating from the pro-homeschooling side, and at CRHE yet another approach from the homeschooling-cautious side.) Homeschoolers seemingly are both increasing in number and capturing more of the K–12 education share; in fact, at roughly 2.6 million students they are operating at the same order of magnitude as charter schools and private schools (3.7 and 4.7 million, respectively).

Despite parents’ clear desire for alternatives to public schools, progressives remain concerned about homeschooling. Besides concerns rooted in individual children’s welfare, progressives will sometimes hint that too many homeschoolers might lead to disruption in the civil sphere. (See, for instance, the recent Amazon documentary Shiny Happy People.) These latter fears are overblown. Homeschoolers are nowhere near replacing the American liberal regime with a Christian commonwealth (and even if they were, the evangelicals and the Catholic integralists would never be able to agree on a constitution). And while some critics of homeschooling fear that homeschoolers are programmed to be good little soldiers for their subculture, it appears that—while homeschoolers largely retain their parents’ values—those whose parents were more paranoid and less rigorous are also more likely to be homeschooling-skeptical. In other words, the worse the homeschooling might appear to an unbiased outside observer, the less likely it is to be reproduced generationally.

Fearful Homeschooling Is a Bust

In 2023 I began taking non-random surveys of former homeschoolers, looking at correlations between experiences with and attitudes about homeschooling. The first survey, with 133 respondents in January and February of 2023, revealed drawbacks in overly protective homeschooling. While moderate parental protection was harmless, strong interest in protecting was correlated with former homeschoolers’ unhappiness about homeschooling and decreased willingness to homeschool their own children. The second survey, with 377 respondents between June and October of 2023, confirmed these results. For adult former homeschoolers, happiness from having been homeschooled was generally inversely proportional to how strongly protection motivated their parents. Among respondents whose parents definitely homeschooled to protect them, only 23 percent reported that homeschooling led mainly to their happiness, in contrast to 36 percent among all respondents who reported the same. By contrast, 54 percent of respondents whose parents had only “somewhat” sought to protect them reported that homeschooling led mainly to happiness. Likewise, the adult children of parents who definitely sought to protect them are less interested in second-generation homeschooling.

One respondent, K.M., expressing common sentiments about overprotection, wrote that homeschooling left them “feeling isolated, not simply socially, but the lack of intervention and resources for my mother to adapt to my learning needs with undiagnosed ADHD. [There was] zero access to mental health resources from other concerned adults.” Another former homeschooling student painted a similar picture: “I was isolated from the outside world specifically to prevent making friends, forming relationships with peers or any adults other than my parents, or encountering popular culture. . . . Often we ended up having to learn basic facts about the world on our own.”

Rachel O. recalled in post-survey correspondence that her parents “prioritized protecting me and my siblings from the influences of a world that they perceived as evil/corrupt. That was a key motivation for their decision to homeschool. They also thought they could provide a better education at home” than their rural public schools, which, she said, was probably true. Rachel described the pedagogy and the curriculum her family used (A Beka Books, since rebranded as Abeka) as traditional, rigid, and suited neither to her needs nor to college preparation. This upbringing backfired: Rachel’s younger sister and brother eventually left Christianity, while Rachel herself “‘rebelled’ by converting to Catholicism in college.” Yet despite her complex experience, Rachel (who has taught for several years at a Catholic girls’ school) would homeschool her own children “if it were in the best interest of the individual child, their needs, learning style, etc.”

If Rachel O.’s parents sound largely well-intentioned albeit mistaken in their methods, sometimes sheltering homeschool children is more about parental convenience than child well-being. That was the case for Renata E. Her parents did seek to protect her from the outside world, but she believes, she said, that they homeschooled mainly “because it enabled me to work more hours and do more chores than my not homeschooled siblings.” Her brothers, diagnosed with dyslexia and ADHD/ADD, were schooled traditionally; but because Renata was a girl, she was taught chiefly “to be a good wife and mother,” doing “the lion’s share of the chores” but not being allowed to learn to drive. In college Renata was diagnosed with learning disabilities similar to those of her siblings: dyslexia, ADHD, and an auditory processing disorder. She believes that traditional schooling would have uncovered these issues sooner. Given her own experience, Renata would be reluctant to homeschool her own children, although she notes that she “[does] not dislike homeschooling as a whole.”

Whether homeschool parents appear to have mixed motives, like Renata’s, or are well-intentioned but heavy-handed, like Rachel’s, homeschooling out of fear and a desire to be overly protective of one’s children can backfire. That said, parents worried about their children’s safety are not always wrong: violence and bullying in public schools are genuine issues. According to Pew, 23 percent of public K–12 teachers report experiencing “a lockdown in the 2022–23 school year because of a gun or suspicion of a gun at their school,” while more than half of teachers say there are “police officers or armed security stationed at their school.” Bullying, though less sensational than school shootings, is also a problem. Clearly, while many safety-conscious homeschoolers are motivated by cultural concerns, fear for their children’s physical well-being is also a component.

But it would be a mistake to conclude that concern for children’s cultural identity (as opposed to concern for their physical well-being) is the hallmark of a homeschool parent who will fail to pass their values on. The surveys did not reveal a negative impact when homeschooling parents wanted “to create a positive moral environment.” Families who set such goals were as likely as the average survey family to have adult children who associated homeschooling with happiness and were interested in homeschooling their own children. “My education was academically very good,” wrote Monica, “and was managed in such a way that it came (relatively) naturally. It also gave my parents many more opportunities to pass on their values, which were good. I was able to encounter moral challenges in a more controlled manner, at an age when I was more able to handle them maturely, having been equipped with a good moral foundation.” As Rachel O. put it, “[My husband and I] don’t want to make decisions for our family based on fear but with confidence and humility, understanding both our responsibilities and limitations as parents. Unlike my parents, we would welcome questions and doubts. We believe that we can help to guide our children’s hearts to God and create an environment that fosters their faith development, but we cannot force them to believe or behave in certain ways.”

Academic Enthusiasm Is Crucial

While cultural and safety concerns loom large in many media narratives about homeschooling, an equally critical facet of the homeschooling movement is academics. Certainly there are homeschooling parents who neglect their children’s education, but there are also parents who homeschool because they want to give their children a better education than public schools could provide. While American students largely measure up to their peers in science, their math scores (like those of several other countries) have been declining. Perhaps more jarring, their scores have been declining relative to previous years: the Nation’s Report Card shows largely stagnant or declining scores for subjects tested in the past five years (whereas for subjects tested ten to fifteen years ago there is a mix of increase and decline). Under those circumstances, it is unsurprising that parents who have the resources to avail their children of more academically rigorous education—charter, private, or homeschool—are inclined to do so.

Academics proves to be crucial for homeschool graduates’ happiness and their interest in homeschooling their own children. Respondents from the second survey whose parents definitely homeschooled to enable higher academic achievement were significantly more likely to say that homeschooling led “only or almost only to happiness” (35 percent versus 22 percent among all respondents). Respondents for whom this was mostly true were more likely to say homeschooling led mainly to happiness (43 percent versus 36 percent among all respondents). Similarly, parental interest in academics translated into a stronger interest in second-generation homeschooling: when it was definitely true that parents homeschooled to boost academic achievement, very strong interest in second-generation homeschooling increased by ten points (to 51 percent), and when it was mostly true, by eight points (to 49 percent).

Respondents who were homeschooled for academic reasons said their parents were involved in extracurriculars, allowed them to pursue their own interests, encouraged reading, “were always tailoring the education to [their] actual, present needs” (James B.), and gave their children control of their learning and so turned it into “a personal responsibility that I really wanted” (Michael G.). “At least where I live,” one respondent wrote in post-survey correspondence, “I have to report to the [local] school at the end of each year with a portfolio that has been looked over by a registered evaluator. However, if there have been problems, I might suggest having to report half-way through the year as well as at the end, that way mistakes or gaps can be caught quicker.” He also pointed out that in his state “there is already an (extensive) list of standards and curricula that meet the state standards,” which act as guidelines for homeschoolers. Another correspondent, Traci H., underscored the importance of parental involvement in homeschooling success, describing her mom as “extremely involved” and ready to answer questions: “If she didn’t know she would take the time to find out. She hired tutors on a few subjects for a short time and also had old college connections help with difficult concepts on the phone.” It seems unsurprising that for homeschool graduates like these, whose education focused on the intellectual development of the child and their growth as a person, positivity about their past experiences and about future homeschooling is high.

The Roots of Overprotection: Ideology or Psychology?

The importance of academics and the dangers of parental fear were findings in the first survey that the second survey confirmed; but the second survey was designed to shed further light on the role of psychology and religion in former homeschoolers’ reported happiness and interest in future homeschooling. Specifically, the second survey asked additional questions about mental health and neurodiversity, which clarified the origins of many negative homeschooling experiences.

While it is popular to imply that acute Christian conservativism leads to bad homeschooling (see, for instance, the aforementioned Shiny Happy People as well as narratives like Tara Westover’s Educated), it seems clear that the real issue for the previous generation of homeschoolers—as for children in schools—was their parents’ own well-being. In families where a parental mental health issue, neurodivergence, or a learning disability was suspected by the adult children, strong interest in second-generation homeschooling decreased from 41 percent among all respondents to 32 percent. In families with actual diagnoses, strong interest in second-generation homeschooling also decreased to 32 percent, but moderate interest in second-generation homeschooling actually increased from 22 percent to 36 percent. Where happiness is concerned, the results are similar: a parental diagnosis of a mental health issue, neurodivergence, or a learning disability somewhat decreased their adult children’s enthusiasm for homeschooling, but when those problems were suspected in the absence of a diagnosis, the numbers were markedly less encouraging. Clearly, a person can homeschool well while being neurodivergent or facing mental health challenges, but their chances of having a positive homeschooling experience are significantly improved if they are aware of—and presumably therefore taking care of—the areas where they need support.

The second survey also sought to distinguish overprotection, with its significant effect on homeschooling dissatisfaction, from religious intensity by separately measuring homeschool families’ religiosity on three fronts: importance given by parents to religion, church attendance, and family prayer. Families who scored higher on these factors combined did not show significant differences from lower-scoring families. The most noticeable discrepancies were that very high religiosity slightly decreased homeschooling that led “mainly to happiness” (from 36 percent to 32 percent) in favour chiefly of homeschooling that had “mixed or neutral” effects on happiness (from 32 percent to 35 percent); the second highest level of religiosity increased homeschooling that led “mainly to happiness” (from 36 percent to 43 percent) at the expense chiefly of homeschooling that had “mixed or neutral” effects on happiness (from 32 percent to 28 percent). With religiosity and interest in second-generation homeschooling, there was equally little consistency of pattern. While religion is a factor, therefore, in some people’s dissatisfaction with homeschooling, it is not as simple as saying that more-religious homeschooling families have more-negative outcomes.

Since evangelical Christians are often considered stereotypical homeschoolers, it also seemed worthwhile to measure that group’s interest and happiness scores. As it happens, respondents with evangelical backgrounds (44 percent) were not markedly different from the respondents as a whole. Indeed, the marginal differences favoured evangelicals: they reported slightly more often that homeschooling led only or almost only to happiness (25 percent versus 22 percent overall) or mainly to happiness (37 percent versus 36 percent) and that they were interested in homeschooling their own children (45 percent saying “definitely” versus 41 percent, and 23 percent saying “yes” versus 22 percent). If there is a specific religious affiliation that tends to result in poor homeschooling outcomes, it is not simply identical to “evangelical Christian.”

The significance of mental health concerns in the homeschooling context is unsurprising given their increasing prominence in public schools and public discourse more generally. Mental health tops bullying as a parental concern for publicly schooled children, and while the pandemic seems to have made the problem worse, it has been around for some time. (Bullying and mental health may well be connected in some cases; the previously cited research connecting school sessions with suicide mentions bullying as a possible mechanism.) Moreover, if research implicating smartphone technology as a factor in teen mental health challenges is correct (see work by Jean Twenge, Jonathan Haidt, Zach Rausch, et al.), then parents wishing to avoid smartphone culture need to seek out other parents who will agree to delay smartphone usage for their children—which in some districts may mean pursuing private school or homeschooling. In fact, while overly sheltering one’s homeschooled children can have a deleterious effect on them (especially, perhaps, when parental mental health is implicated in the decision to shelter), some kinds of sheltering, such as protecting one’s children from smartphone addiction and cyber-bullying, may be profoundly important for protecting the mental health of children themselves.

Overall, the survey suggests that despite individual cases of barely literate homeschoolers on the one hand and National Merit Scholars on the other, homeschoolers are humans too, with the usual good and bad that being human entails. It is reasonable that people who experienced abuse or educational neglect (or both) while being homeschooled should seek answers. But the answer from the secular and progressive mainstream—that religion is the problem—belies homeschooling reality. In contrast to John Oliver, the Washington Post, Elizabeth Bartholet, and others, homeschooling’s real problems transcend religion and lie rather in the realm of psychology: in fears that lead some parents to overprotect (as opposed to age-appropriately protecting) their children, and (perhaps relatedly) in some parents’ lack of awareness of their own psychological needs and differences. The paranoid style in homeschooling is the product not of religion but of human nature, and the best way to strike at its root is not to regulate human failings into irrelevance but to find better ways to help homeschooling parents thrive.

The names of some survey respondents have been changed at their request.

This is an ongoing research project. If you want to be part of it, please take the survey here: https://www.surveymonkey.com/r/YKBRHNL.