Homeschooling in the United States is often imagined as an all-or-nothing decision: Families who choose it do so full-time, for all their children, across their entire K–12 education. Recent findings from Johns Hopkins University’s Homeschool Research Lab revealed a more complex picture of students moving between educational settings (homeschool, public school, co-ops, etc.) at different points in their lives. But explanations for and descriptions of these switches were absent. A new report from the lab’s researchers remedies that.
Albert Cheng, Angela Watson, and Matthew Lee use data from three administrations of the National Household Education Surveys (NHES) to a nationally representative sample of U.S. households in 2012, 2016, and 2019.[1] Each household included provides information about schooling experiences, parent involvement in their child’s education, family life, and demographic characteristics for one randomly selected school-aged child in the household. While NHES defines a household as “homeschool” or “conventional school” based on the response for that single sampled child, the survey does actually ask about the educational mode of other school-age children in the home. The analysts exploit this additional data for their study. Households in which the sampled child had no siblings or only had siblings who were too young/too old to be enrolled in primary and secondary schooling were excluded. The final sample comprised 25,261 households.
First, the team provides summary statistics to illustrate the prevalence of what the they call “sector mixing”—that is, a family simultaneously having at least one child in homeschooling and at least one other child in a district or private school. Over all three waves of NHES data, an average of 2.5 percent of U.S. households with more than one school-aged child practiced sector mixing. This includes a majority (53 percent) of the multi-child households that NHES deemed “homeschoolers” via their single-sampled-child methodology. The remaining 47 percent of multi-child homeschoolers were indeed exclusively in that setting. Nearly all (97 percent) of the multi-child households deemed “conventional schoolers” were exclusively so—although that could include a public and a private school within one family, a different type of “mixing” than the analysts are looking at here. The percentage of conventional schooling families embracing homeschooling for one or more children rose between the 2012 and 2019 administrations of the NHES survey, from 1.6 percent to 2.8 percent.
Family size, parental education, and parental employment status had the largest impact on sector mixing. Keeping in mind that the overall rate of sector mixing is low (2.5 percent), even small changes in likelihood are significant. Every additional child increased the likelihood of a family practicing sector mixing by about 1 percentage point. White families were more likely to practice sector mixing than Black, Hispanic, and Asian/Pacific Islanders by roughly 1 percentage point; households with college-educated parents were more likely to practice sector mixing than their high-school-diploma-only peers. Households in which both parents are working for pay were 1.6 percentage points less likely to practice sector mixing than households in which one or neither parent is working for pay. Sector mixing declined, however, at the highest income levels, with the wealthiest families in the survey nearly 2 percentage points less likely to do so than the lowest-income families. Sector mixing was more common in the South and West, but there were no significant differences in mixing rates based on urban, suburban, or rural locations.
NHES homeschool families were asked to consider eight specific reasons[2] for homeschooling their child and to choose all that applied to them. The analysts found that reasons related to “household needs” (such as providing religious or moral instruction and wanting a non-traditional approach to education) led to far less likelihood of sector mixing (by 10 to 15 percentage points, compared to the full NHES sample), while reasons related to “child-level needs” (like illness or mental health concerns) led to a much higher likelihood of sector mixing in the household (16 to 19 percentage points).
The researchers believe that their estimates of sector mixing among U.S. families likely underreport the true numbers due to the fact that the NHES surveys are not designed to provide the full data required for proper calculation. Nevertheless, they conclude that the picture of “the variegated nature” of homeschooling their estimates paint—prevalence, motivation, demographics, etc.—is valuable. This is especially true of sector mixing between homeschooling and traditional public/private schools. More data are required, but the questions needing answered by those data are becoming more and more clear.
SOURCE: Albert Cheng, Angela Watson, and Matthew Lee, “An analysis of educational sector mixing among U.S. homeschooling families from three waves of the national household education survey,” Cambridge Journal of Education (August 2026).
[1] 2023 data were captured, but are not publicly available as yet.
[2] (1) concerns about school environment such as safety, drugs or negative peer pressure; (2) dissatisfaction with academic instruction in school; (3) desire to provide religious instruction; (4) desire to provide moral instruction; (5) the child has a long-term mental or physical health problem; (6) the child has a temporary illness that prevents him or her from going to school; (7) the child has a special need that the school cannot or will not meet; and (8) interest in a nontraditional approach to the child’s education.
