Past estimates of the number of homeschoolers vary by hundreds of thousands of children. Brian Ray, president of the National Home Education Research Institute, estimated the number of homeschoolers to be around 1.15 million during the 1996-97 school year, and predicted that the number would grow to at least 1.3 million during 1999-2000 (Ray 1997). Patricia M. Lines, through her research at the U.S. Department of Education, National Institute on Student Achievement, Curriculum and Assessment, estimated the number of homeschoolers to be around 700,000 during 1995-96, possibly growing to 1 million by 1997-98 (Lines 1999). Both Ray and Lines grant that their estimates probably anchor the range within which the actual number of homeschoolers could fall.

The methods used by Ray and Lines in the development of their estimates varied. Ray derived his most recent estimate of the number of homeschoolers using his own 1995 survey of homeschoolers and their use of curricular packages as his base and sales of homeschooling curricular packages to adjust for growth over time. Ray applied the ratio of users of curricular packages and nonusers identified in the 1995 survey to more recent sales of homeschool curricular packages to obtain his 1999-2000 estimate. Lines collected data from all states that obtained records on homeschooling children in both the 1990-91 and 1995-96 school years (32 states and the District of Columbia). Using the 12 states with high record-collection rates for homeschoolers, based on Ray's estimates of the percentage of homeschoolers who reported filing in their state, Lines estimated the percentage of school-aged children who were homeschooling in those 12 states. Lines estimated the number of children homeschooled nationally by applying the percentage distribution of homeschoolers from the state sample to national totals of school-aged children.

The National Center for Education Statistics (NCES) was the first organization to attempt to estimate the number of homeschoolers in the United States using a rigorous sample survey of households. A household sampling frame circumvents problems inherent in the use of incomplete sample frames, such as customers of curricular providers and administrative records. Attempts to develop estimates of homeschoolers through household surveys, however, can also be problematic. The first two efforts to estimate homeschoolers at NCES-through the October supplement to the 1994 Current Population Survey (CPS:Oct94) and through the Parent and Family Involvement in Education/Civic Involvement Survey of the National Household Education Surveys Program, 1996 (PFI/CI-NHES:1996)-produced very different estimates. One problem that may have contributed to the varying estimates was the difference in how the two surveys identified students who were both homeschooled and enrolled in school part-time. Neither survey collected precise data on these part-time homeschoolers. An NCES technical report, Issues Related to Estimating the Home-Schooled Population in the United States with National Household Survey Data, explores in detail the differences in survey design and execution that may have contributed to the disparity between the CPS:Oct94 and PFI-NHES:1996 estimates (Henke et al. 2000).

In this report, the Parent Survey of the National Household Education Surveys Program, 1999 (Parent-NHES:1999) is used to estimate the number of homeschoolers in the United States, to describe the characteristics of homeschoolers, to document parents' reasons for homeschooling, and parents' reports of public school support for homeschoolers. Students were considered to be homeschooled if their parents reported them being schooled at home instead of a public or private school, if their enrollment in public or private schools did not exceed 25 hours a week, and if they were not being homeschooled solely because of a temporary illness. The unweighted number of homeschooled students used in this analysis is 275 and the unweighted number of nonhomeschooled students is 16,833. Students are defined in this report as children ages 5 to 17 with a grade equivalent of kindergarten through grade 12.