The May-June issue of Harvard Magazine carries an article, “The Risks of Homeschooling,” promoting the argument of Harvard Law School Professor Elizabeth Bartholet that the U.S. should enact “a presumptive ban” on homeschooling. Homeschooling is essentially unregulated, Bartholet argues, and many parents adopt this method of educating their children for nefarious reasons including indoctrinating the parents’ values into their children, isolating the children from society, and abusing them. Parents should be assumed to be incompetent and dangerous educators of their children. Therefore, specific parents may homeschool their children only if government officials determine that allowing them to educate their children at home is worth the risk.
The article prompted a tsunami of critical responses, in Education Next (see “Harvard Professor’s ‘Absurd’ Claim that Homeschooling is Child Abuse”) as well as here, here, here, here, here, here, here, and here. We seek here to move the discussion beyond the 1,000-word Harvard Magazine article that sparked such opprobrium by carefully considering Bartholet’s 80-page Arizona Law Review article that inspired the story. We expected it to be rigorous and fact-based but were sadly disappointed.
We are empirical social scientists who have studied and written about homeschooling along with other education policy topics. One of us, Wolf, homeschooled one of his now-adult children years before the Covid-19 pandemic turned tens of millions of Americans into some version of home-schoolers. Another, Watson, is homeschooling by necessity now. Upon reviewing Professor Bartholet’s article, we conclude that it suffers from contradictions, factual errors, statements of stereotyping, and a failure seriously to consider that the alternative to homeschooling–public schooling–shares the problems that she attributes to home education. One need not have personal experience with homeschooling to identify the article’s many flaws.
Professor Bartholet often seems to be arguing with herself. Early in the article, she states: “We have no way of identifying, based on existing information, the total group of homeschoolers.” For the remainder of the article, however, she confidently describes what is true of “many” (used 90 times), “most” (53 times), and “a majority of” (6 times) homeschoolers. If no one knows the denominator, how can anyone say what is true for many, most, or, especially, a majority of the homeschool population? Similarly, in the section where Bartholet argues that “many” homeschooling parents are incapable of educating their children, she asserts that homeschoolers are disadvantaged relative to non-homeschoolers. In the later section where she dismisses the empirical research on homeschooling outcomes, she argues that the superior results for homeschooled students are simply because their families are advantaged relative to non-homeschoolers. Throughout the article, homeschoolers are characterized as almost monolithically (“up to 90%”) conservative Christians seeking to indoctrinate their children. On page 10, however, they are described more accurately as highly diverse in their religious affiliations and motivations. On page 5, Bartholet states, accusingly, “Most all miss out on extracurricular activities.” Five pages later she admits that many parents who homeschool “make efforts to enable their children to participate in certain school programs such as sports.” If families homeschool primarily to indoctrinate their children, shield them from any outside influence, and abuse them, as she argues throughout the essay, why do most families homeschool a child for only part of their K-12 education, as is implied by statistics on page 9 of the law review article?
One contradiction stands out. Professor Bartholet says “homeschooling in its current unregulated form poses serious risks of abuse and neglect.” That is the central claim of her paper. Later in the article, she lists a dizzying array of government regulations of homeschooling judged by courts to be “legal.” These include her preferred regulations that parents be preapproved by the state before they can educate their children, be certified instructors, use a specified curriculum, provide annual reports, administer the state test to their child, and allow government officials to inspect their home. So homeschooling is unregulated but her prescribed regulations of homeschooling are reasonable because they operate legally today?
The article contains a number of assertions about homeschooling that are clearly undermined by the facts and unsupported by her sources. Bartholet’s central claim is: “Many families choose homeschooling precisely because it enables them to escape the attention” of Child Protective Services. Her source that supposedly proves this explosive charge merely states that there “is anecdotal evidence” of such ill intent. Similarly, Bartholet asserts that the “majority” of families that homeschool are “descendants of the original conservative Christian wing.” She writes that estimates range “from a majority up to 90%.” The sources for this claim include a federal government survey that indicates only 16% of parents homeschool primarily to provide “religious instruction,” an article from The Atlantic on homeschoolers suggesting that “[R]oughly two-thirds are Christian,” and speculation by an author that 90% of families that homeschool “are religious.” If only 67% of homeschoolers are even Christian, how could “up to 90%” be specifically conservative Christians? The numbers don’t add up.
Bartholet states that many families that homeschool “are at the low end of the socioeconomic ladder, with 19% below the poverty line and 36% between poverty and 200% of poverty, significantly more than that of those in public and private schools.” She is claiming that 55% is a comparatively high proportion of low-income students in a population. It is not. Poverty and near-poverty rates are either missing or unreliable for most private school students, but over 52% of the public school student population has incomes at or below 185% of poverty, meaning that the proportions of students at or below 200% of poverty are similar for homeschool and public school students, in spite of the fact that only 28% of homeschool families contain two adult working outside the home. She asserts, without evidence, that children who are homeschooled lose “out on opportunities to learn things that are essential for employment and for exercising meaningful choices in their future lives.” A recent study by the research center CARDUS, which Bartholet cites at other points in her article, reports, to the contrary, that employment rates and earnings are similar for homeschooled and non-homeschooled adults, and the two groups are “not much different in their pursuit of new experiences in life.” Bartholet claims, “Homeschooling presents…democratic concerns.” The only empirical study of that question that we know of concludes, to the contrary, that students who were homeschooled display significantly higher levels of political tolerance than otherwise similar students who attended public schools.
Speaking of tolerance, some of the statements in the Bartholet article concern us. She states: “Members of a variety of religious groups are included today in this conservative Christian wing, including many Mormons, Jehovah’s Witnesses, and Seventh-day Adventists. These homeschooling groups hold similar ideas about the importance of keeping their children isolated from conflicting cultural values.” Professor Bartholet presumes to define which religious groups are and are not “conservative Christians,” then engages in stereotyping by ascribing the same “isolationist” values to individuals in those groups. She criticizes parents who homeschool for having values that run counter to those of the larger society, thus undermining the values of pluralism and toleration of people with views of the good life that differ from one’s own. She demonstrates prejudice against immigrants by asserting that non-English-speaking parents are incapable of homeschooling their children. She argues that some parents who homeschool “are mentally ill or disabled, or caught up in substance abuse,” as if having a disability necessarily disqualifies a parent from educating their child. These are some of the statements that leapt out as offensive to us.
Child abuse is a serious problem. It concerns us greatly. It occurs in the homes of some parents who homeschool their children but also in public schools around the country. Bartholet’s article pays scant attention to the usual alternative to homeschooling–public schools–and the problems of child abuse and disturbingly low levels of civic knowledge reported there. A rich debate could be had about when unconventional approaches to parenting cross the line into abuse and neglect. We did not find such a balanced or nuanced discussion in the 80 pages of this article.
Professor Elizabeth Bartholet’s claims that homeschooling contributes significantly to the scourge of child abuse fail to survive scrutiny. Ironically, Bartholet pleads with us to trust government officials to decide which select parents have the capability and correct values to be safely granted the privilege of educating their children at home. We now know what that looks like. Last month, government officials ordered almost all K-12 children to be educated at home in order to promote the safety and wellness of themselves and our society. We are all homeschoolers now, by government decree. That development appears to have left Professor Bartholet, well, schooled.
Patrick J. Wolf is a distinguished professor of education in the Department of Education Reform at the University of Arkansas and co-editor of Educating Citizens: International Perspectives on School Choice and Civic Values.
Matthew H. Lee is a distinguished doctoral fellow in the Department of Education Reform at the University of Arkansas.
Angela R. Watson is a senior research fellow at the Johns Hopkins Institute for Education Policy.