Copyright © by National Home Education Research Institute
[Also, see the new population size estimate for 2021-2022]
Abstract
The purpose of this study is to estimate the number of homeschool students in the United States during the 2020-2021 academic school year. The estimate is derived by establishing the size of the nationwide school-age population, ascertaining the percentage of all students that were homeschooled, and assuming a level of underrepresentation of homeschooling in the data that represent the portions of all students who were homeschooled. Homeschool registration and enrollment data from 16 states and data from the U.S. Census Bureau’s experimental Pulse Survey were used. The conclusion is that there were an estimated 3.721 million school-age (K-12) homeschool students in the United States during the 2020-2021 school year.
Keywords: homeschooling, home education, population size, numbers, United States
Introduction and Purpose
Parent-directed home-based education (homeschooling) was the norm before 1900 in the United States but had nearly gone extinct by the 1970s (Ray, 2021a). During the 1980s and 1990s, the homeschool population grew very rapidly. Many want to know now, how many kids are homeschooled today?
Lines (1991) estimated that there were some 13,000 homeschool students in the late 1970s in the United States. The United States Department of Education (USDE) (2021c) reported that about 3.3% of school-age children (“5- to 17-year-olds,” p. 58), or 1.690 million, were homeschooled by the spring of 2016, and found that this was not significantly different from their estimate that 3.4% of school-age children were homeschooled four years earlier. [note 1] However, using different methods, Ray (2021b) estimated that there were over two million K-12 homeschool students during the spring of 2016, and then estimated 2.6 million during March of 2020 (Ray, 2021b).
The United States Census Bureau (USCB) (2021b) reported that the number of households with school-age homeschool children doubled from March of 2020 to March of 2021 (from 5.4% to 11.1%). The USCB was careful to explain the many limitations of their sampling and some nuances built into their estimates. For example, the unit of analysis was adults, not households (but then the USCB weighted data by household). The response rates to their surveys were very low and the wording of questions was changed during different weeks of the survey. Third, the USCB found (in their Table 1) that there was no significant difference in “homeschooling rates of households by states” between spring of 2020 and fall of 2020 in 23 of the 50 states and the District of Columbia. In addition, the USCB report was built on the number of adults living with a homeschool student and not the number of homeschool students in each household. To this author’s knowledge, the Census Bureau has not published an estimate of the number of homeschool students during 2020-2021 as of the publication of this paper.
Various authors have given estimates of the U.S. homeschool population size during the 2020-2021 school year (e.g., McDonald, 2021). Some estimates have put the number at 5.0 million or more. This author reported an estimate of 4.5 to 5.0 million (Ray, 2021b, July 1, 2021); however, more relevant data have emerged since, thus motivating the present analysis.
The purpose of this study is to arrive at an estimate of the number of K-12 homeschool students in the United States during the 2020-2021 academic school year. In sum, the estimate will be derived by establishing the size of the nationwide school-age population, ascertaining the percentage of all students that were homeschooled, and assuming a level of underrepresentation of homeschooling in the data that represent the portions of all students who were homeschooled. Relevant data for the size of the total student population and for portions home educated come from multiple sources.
It is challenging to find consistent data on how many total school-age children (ages 5 to under 18), in all schooling sectors (public school, private school, and homeschool), there are in the United States. The USCB (2021d) reported that there were 53,503,042 on July 1, 2019. Assuming that the school-age population grew by 0.5% from fall 2019 to fall 2020 (USCB, 2019, 2021d), there were 53.771 million school-age children during the 2020-2021 school year. On the other hand, the USCB (2021b) estimated from its Census Pulse Survey Week 34, asking adults to look back at the 2020-2021 conventional school year, that there were 51,552,327 students during March of 2021.
The USDE (2021a) reported that in the fall of 2019 there were about 49.2 million students attending public schools in kindergarten to grade 12, and then about 48.1 million in the fall of 2020. The USDE also reported that there were, in the fall of 2017, about 5.7 million students attending private schools (and this estimate includes pre-kindergarten enrollment in schools that offer kindergarten or a higher grade). The USDE’s most recent estimate for homeschooling was that there were 1.69 million homeschool students in the spring of 2016. Overall, then, assuming homeschooling grew by 2% per annum from spring 2016 to fall 2019 (four school-year cycles; 1.83 million students) and private school enrollment plateaued during that period, then the USDE numbers would lead to an estimate of about 56.53million school-age students (not including pre-K) in public schools, private schools, and homeschooling during the 2019-2020 school year. Using the USDE’s number and assuming an annual growth of 0.5% (USCB, 2019, 2021d), there were 56.813 million in 2020-2021.
Similarly, EducationData.org (2021), using data from the USDE’s National Center for Education Statistics and other sources, reported that there were 56.6 million students in public schools and private schools in 2019, but they do not mention a number for homeschooling.
To arrive at an estimate of the total school-age population during 2020-2021 for the purposes of this analysis, it was assumed that the USCB’s general census data are more accurate than their Pulse Survey data. The average of this USCB number and the USDE’s is 55.292 million and this number was used as the total number of school-age children during 2020-2021.
Methods
The primary sources of data for this analysis are the United States Department of Education (USDE) (2021a, 2021c), the United States Census Bureau (USCB) (2021c, Week 34 PUF), as well as data from all state-level departments with relevant publicly available data (see list below).
The Census Bureau began the experimental Pulse Survey in April 2020. For several weeks of the survey, adults were asked whether there were public school, private school, or homeschool students living in the household. The surveys’ findings reported on adults as the unit of analysis and did not ask how many individual students were engaged in the different school sectors (i.e., public school, private school, homeschool). The USCB began, in their Phase 3.2 questionnaire that was first administered during July of 2021, to ask adults how many public school, private school, and homeschool (“… that is not enrolled in public or private school”) students there were in each household during 2020-2021. The corresponding public use data files (e.g., Week 34 PUF) were accessed to analyze the data on these variables.
The survey response rate for the USCB’s Week 34 was 6.1% (USCB, 2021e, “Source of the Data and Accuracy of the Estimates for the Household Pulse Survey – Phase 3.2”). This is very low compared to response rates (around 90%) that the USCB obtains for surveys such as the American Community Survey (USCB, 2021a). Further, it should be noted that while the dataset for public school included up to 4 students per household, those for private school and homeschool topped out at 2, due to the USCB’s methods; therefore, the data did not reveal the number of households with 3 or 4 private school or homeschool students.
The USCB’s public use file data provided a person-weight variable and a household-weight variable. In the present study, data were weighted by household before calculating frequencies for public school, private school, and homeschool enrollment. The frequencies (by number of households with a given number of students per household) were used to calculate totals and percentages of the nationwide student population.
Various states’ gather and summarize data on the number of homeschool students in their states. Whether they do so depends partly on whether the state government has any motivation or incentive to report such data and on the state laws that do or do not address homeschooling. In some states, parents are free to privately homeschool without state controls and in other states the government exerts much control and regulation over private home-based education. Whether a state exerts much or little control over private homeschooling is not necessarily correlated with whether the state government collects and disseminates data or statistics on homeschooling students. For this project, the author was able to secure data from 16 states (i.e., Delaware, Florida, Georgia, Massachusetts, Minnesota, Montana, Nebraska, North Carolina, North Dakota, Oregon, Rhode Island, South Carolina, South Dakota, Virginia, Washington, and Wisconsin; see Table 1). Oregon data, however, represented the number of first-time registrations of a given child and not total enrollment. Various state laws define school age differently. These states represent all four major regions of the United States. However, the data cannot be fully integrated since various states’ laws define differently what is school age.
Findings
The USCB data showed that of all school-age children during March of 2021, 85.8% were public schooled, 8.2% were private schooled, and 6.04% were homeschooled (while the USDE estimated homeschooling at 3.3% in 2015-2016).
On the other hand, the 15-state (which excludes Oregon) weighted mean (i.e., weighted by the number of homeschool students per state) indicated that 4.74% of these states’ school-age children were homeschooled (Table 1). It should be noted that compulsory school ages in some states do not include ages 5, 6, and 17; across the states and District of Columbia, roughly 11% of 13 school-age years are not compulsory (United States Department of Education, 2021b). Therefore, the number of school-age (i.e., 5- to 17-year-olds) homeschool children is underrepresented in such states.
Table 1. Homeschooling as a Percent of All Students: State Reports and Census Bureau Pulse Survey
| Source | Number of homeschool students | Homeschool as Percent of all students |
| North Carolina | 179900 | 10.6% |
| US Census Bureau | not reported | 6.0% |
| Montana | 9868 | 6.0% |
| Virginia | 65571 | 4.8% |
| Florida | 143431 | 4.6% |
| Georgia | 85510 | 4.6% |
| Nebraska | 14780 | 4.3% |
| South Dakota | 6698 | 4.2% |
| North Dakota | 5308 | 4.2% |
| South Carolina | 28316 | 3.4% |
| Delaware | 4,950 | 3.3% |
| Washington | 39843 | 3.3% |
| Wisconsin | 31878 | 3.2% |
| Minnesota | 30955 | 3.2% |
| Rhode Island | 3335 | 2.2% |
| Massachusetts | 17127 | 1.7% |
Since the actual number of nationwide homeschool students has not been recorded and therefore cannot be calculated, the estimated number of homeschool students can be reached by estimating both the total number of school-age students and the percentage of those students who are homeschooled.
Further, nationwide estimates of percentages and actual student counts by the USDE and USCB have many limitations. At the state level, however, homeschool registration and enrollment numbers are more tangible and consistent and not based on the limited response rates and unknown representativeness of the federal national surveys.
Most of the calculations with data that preceded the presentation of numbers and decimals above were done at 3 or more decimals before rounding to the decimal places shown in this report.
Discussion
Based on this author’s decades of experience conducting research on the homeschool educational sector and other empirical evidence, not all students who are homeschooled are accounted for by state departments of education and homeschool parents are less likely than institutional school parents to respond to government survey questions about the nature of their children’s education and schooling. Further, as noted above, roughly 11% of 13 school-age years are not compulsory in the United States. Therefore, it was assumed that the number of homeschool students is disproportionately underrepresented by 20% in state-level data-collection systems and in survey research such as that done by the USCB and the USDE, collectively.
With the 20% underrepresentation adjustment, the 15-state weighted mean percent of school-age children (4.74%) for March of 2021 would be 5.69% and the USCB homeschool number (6.04%) would be 7.25% of the school-age population for March of 2021. Considering the limitations and challenges of both state and federal government agencies gathering data on homeschool students, it is difficult to know which of these two statistics regarding the homeschool school-age population is closer to reality. Since the USCB survey was nationwide and included all states, the statistic from it was given double weight to arrive at 6.73%. Based on the estimates of 6.73% and 55.292 million, there were an estimated 3.721 million school-age (K-12) homeschool students in the United States during the 2020-2021 school year. A reasonable range estimate would be, therefore, 3.349 million to 4.093 million during 2020-21.
Conclusion
The U.S. Census Bureau (2021b), using their “… experimental Household Pulse Survey,” reported that the number of households with (i.e., adults living with) homeschooled students doubled from March of 2020 to March of 2021, and some interpreted this as meaning that the number of homeschool students doubled but the latter is not what the USCB reported. Various news articles have described school-district specific and statewide increases in homeschooling of 50% to 100% from March of 2020 to March of 2021. Further, the 16 states listed in this study experienced growth in homeschooling by 9% to 270% during that same year, with 16 of them reporting 9% to 90% growth.
What is the true parameter, the most accurate number, to answer the question, “how many kids are homeschooled” or “how many homeschoolers are in the US”? Laws, reporting regulations and methods of reporting, and the proclivity of homeschooling parents to answer surveys about their children vary across the nation but the state-level data have a solid level of reliability and history. Many of the states have been collecting data on the number of registered homeschool students for more than a decade. The reliability of the data from the experimental USCB Pulse Survey, on the other hand, is less known and established so extrapolating from those data alone is unreliable. This inquiry has been a best effort to use a variety of the best sources of up-to-date and available data.
This analysis leads to the following answer: There were an estimated 3.721 million homeschool students (i.e., school-age, ages 5 to under 18, K-12) in the United States during the 2020-2021 school year.
The remarkably higher numbers in the homeschool population during the spring of 2021 compared to the spring of 2020 will likely subside (e.g., Nitcher, 2021), but not completely. Many observers expect the number of homeschool students during the 2021-2022 conventional school year to remain significantly higher than during 2019-2020 because more families, both parents and children, experienced the benefits of parent-directed home-based education during the government restrictions on schools during the past two school years. The growth in homeschooling will also continue due to the other stable reasons for homeschooling that have motivated parents over the past 40 years (Fields-Smith, 2020; Murphy, 2012; Ray, 2021b). Helpful data from sources such as the USCB, USDE, and state departments of education on which to base reliable estimates of the homeschool population during 2021-2022, however, will likely not be available until at least late October 2021.
Notes:
- The term significant typically means statistical significance in this paper.
Acknowledgements by Author
I thank William Lloyd of the U.S. Census Bureau and Angela Watson of Johns Hopkins University for giving their time and effort to help me improve this project and paper. This acknowledgement does not in any way imply that either of them agrees with the methods or content of this paper. All errors and omissions are those of the author.
References
Educationdata.org (2021). K-12 school enrollment & student population statistics. Retrieved September 2, 2021 from https://educationdata.org/k12-enrollment-statistics
Fields-Smith, Cheryl. (2020). Exploring single black mothers’ resistance through homeschooling. Switzerland: Palgrave Macmillan Cham.
Lines, Patricia M. (1991, October). Estimating the home schooled population (working paper OR 91-537). Washington DC: Office of Educational Research and Improvement, U.S. Department of Education.
McDonald, Kerry. (2021). New census data show homeschooling tripled during the pandemic—and one key group is driving the surge. Retrieved September 6, 2021 from https://fee.org/articles/us-census-homeschooling-triples-diversifies-during-pandemic-response/
Murphy, Joseph. (2012). Homeschooling in America: Capturing and assessing the movement. Thousand Oaks, CA: Corwin, a Sage Company.
Nitcher, Emily. (2021, September 3). Number of Nebraska students enrolled in home schooling down from last school year. Retrieved September 7, 2021 from Omaha World-Herald, https://omaha.com/news/local/education/number-of-nebraska-students-enrolled-in-home-schooling-down-from-last-school-year/article_2b9d8174-0a99-11ec-ac9c-97b0da23de51.html
Ray, Brian D. (2021a). An overview of the worldwide rise and expansion of home education homeschooling (chapter 1). In Rebecca English (Ed.), Global perspectives on home education in the 21st century. Hershey, PA: Information Science Reference (an imprint of IGI Global).
Ray, Brian D. (2021b). Research facts on homeschooling. Retrieved September 2, 2021 from https://www.nheri.org/research-facts-on-homeschooling/
United States Census Bureau. (2019). Drops in natural increase, net international migration resulted in 0.5% annual growth to 328.2M. Retrieved September 2, 2021 from https://www.census.gov/library/stories/2019/12/new-estimates-show-us-population-growth-continues-to-slow.html
United States Census Bureau. (2021a). American Community Survey, response rates. Retrieved September 9, 2021 from https://www.census.gov/acs/www/methodology/sample-size-and-data-quality/response-rates/
United States Census Bureau. (2021b). Census Bureau’s Household Pulse Survey shows significant increase in homeschooling rates in fall 2020. Retrieved July 8, 2021 from https://www.census.gov/library/stories/2021/03/homeschooling-on-the-rise-during-covid-19-pandemic.html
United States Census Bureau. (2021c). Measuring household experiences during the Coronavirus pandemic. Retrieved September 2, 2021 from https://www.census.gov/data/experimental-data-products/household-pulse-survey.html
United States Census Bureau. (2021d). Quick facts: United States. Retrieved September 2, 2021 from https://www.census.gov/quickfacts/fact/table/US/PST045219
United States Census Bureau. (2021e). Source of the data and accuracy of the estimates for the Household Pulse Survey – Phase 3.2. Retrieved 9/2/2021 from https://www2.census.gov/programs-surveys/demo/technical-documentation/hhp/Phase3-2_Source_and_Accuracy_Week%2034.pdf
United States Department of Education. (2021a). Back-to-school statistics. Retrieved September 2, 2021 from https://nces.ed.gov/fastfacts/display.asp?id=372
United States Department of Education. (2021b). Compulsory school attendance laws, minimum and maximum age limits for required free education, by state: 2017. retrieved 9/6/21 from
United States Department of Education. (2021c).Digest of Education Statistics 2019, 55th Edition, February 2021. Retrieved September 2, 2021 from https://nces.ed.gov/pubs2021/2021009.pdf