Good data on homeschool demographics—including race, educational background, and income—can be hard to come by. Neither state departments of education nor individual school districts collect demographic data, and studies conducted with volunteer samples cannot be assumed to be representative of the entire population. The best data we have is collected by the National Center for Education Statistics (NCES), but even it has its drawbacks.
In sum, while studies of homeschooling that use volunteer participants find that their samples are significantly whiter, wealthier, and more educated than the general public, the NCES data suggests these differences are far less significant and that these studies may be leaving out significant portions of the homeschool population, namely, those that are racial and ethnic minorities, economically disadvantaged, and less well educated. The NCES data suggests that homeschooling is far more diverse than it is typically depicted, and suggests new directions for research.
The Data We Have
Studies conducted by both homeschool advocacy groups and academic researchers almost invariably involve participants who are better educated, more wealthy, and whiter than the general public. For example, only 1.2% of homeschool fathers in a study conducted by Dr. Lawrence Rudner in 1998 had not finished high school compared to 18.1% of males nationwide. Similarly, only 0.8% of homeschool families in Rudner’s study had an income less than $10,000 while 12.6% of families with children nationwide have incomes below that threshold. Finally, 94% of survey participants were white as compared with 67.2% of the population as a whole. However, these studies were conducted using volunteer participants rather than randomly selected participants and therefore cannot be assumed to be nationally representative—and indeed, there are good reasons to think that they are not.
Every four years the National Center for Education Statistics (NCES) conducts its National Household Education Survey (NHES). The NHES uses a random sample rather than asking for volunteers and includes a huge sample size, focusing not on homeschooling specifically but education in general. It includes a slate of questions on homeschooling, covering everything from reasons for homeschooling to family size. Its findings suggest that homeschool demographics are more diverse than surveys of homeschooling that solicit volunteer participants may suggest. Of course, the NCES data comes with its own caveats. Some of the smaller numbers shift wildly from survey to survey, suggesting that caution is needed when approaching the results. Further, between 2007 and 2011 the NCES changed its methodology for conducting the survey, throwing comparisons between the 2011 data and previous years into question, and homeschooling parents uncomfortable with studies of homeschooling may choose not to participate, thus skewing the data. Still, it is the most reliable demographic data of homeschooling that we have available to us at this point in time.
Race and Ethnicity
Data released by the National Center for Education Statistics (NCES), like essentially every other study that has touched on homeschooling demographics, has consistently found that a higher than average percentage of homeschooled students are white.
There is some indication that while the homeschooled population is still more white than the student population as a whole, it is growing more diverse. The NCES’s finding that 68% of homeschooled students were white in 2011 represents a decline from previous years. In 2007, they found that 77% of homeschooled students were white and in 2003 they found that 74% were white.
For more, click here for a comparison of race and ethnicity between public school, private school, and homeschool populations using data from 2003.
Studies of homeschooling that recruit volunteer participants have consistently found that homeschool parents are significantly better educated than the average parent. NCES data has traditionally also indicated that homeschool parents are better educated, though not by margins quite as large as surveys that rely on recruiting volunteers. For an example, click here for a comparison of parental education between public school, private school, and homeschool populations using data from 2003.
More recent data has departed from this trend. In 2011, the NCES found that homeschool parents’ level of education was little different from that of the average parent.
It may be that this 2011 data is an anomaly and that future NCES data will once again find that homeschooling parents are slightly better educated than the average parent, but it may also be that homeschooling families are moving closer to the general population in terms of parental education.
For more, click here for a comparison of parental income in public school, private school, and homeschool populations using data from 2003.
Studies of homeschooling that recruit volunteer participants have consistently found that homeschool families are significantly better off financially than the average family with school-aged children. However, the NCES found in 2003 and 2007 that homeschooling families are less likely than other parents to be poor but more likely than other families to be near-poor.
Click here for a comparison of poverty rates between public school, private school, and homeschool populations using data from 2003. The chart shows that homeschooling families were slightly less likely to be poor than other families, but significantly more likely to be near-poor.
More recent NCES data, from 2011, suggests that homeschool families may now be no less likely to be poor than the average family. The data on how likely homeschool families are to be near-poor has not yet been released. In sum, homeschool families do not appear to be any more well to do than other families, and they may in fact be more likely to be poor or near-poor. This could be due in part to the fact that most homeschooling families have only one income.
Several findings in the data collected every four years by the NCES are fairly consistent. First, homeschooled children are significantly more likely than average to live in households headed by two parents.
The NCES data also suggests that homeschooled students are more likely than average to live in two-parent homes with only one parent in the labor force.
While the traditional picture of the homeschooling family—working father, stay at home mother—does represent the reality for the majority of homeschooled children, it should be remembered that there are also a significant number of both homeschooling families headed by single parents and homeschooling families with both parents in the labor force.
Another consistent finding is that homeschooled children are more likely to be in families of three or more children than are other children.
In some homeschool families (as many as half, according to some data), some of the children are homeschooled while others attend public or private schools. This may be especially prevalent for families who move from one educational option to another—from public school to homeschool to private school and back to homeschool, for instance—taking advantage of educational choice to seek out a variety of educational options.
Region and Locale
Data from the NCES has consistently found that homeschool families are more likely than average to live in rural areas and less likely to live in towns or in urban or suburban areas.
The NCES has also found that homeschool families are more likely to live in the South and less likely to live in the Northeast. This finding, too, has been fairly constant.
For the 1999, 2003, and 2007 NCES data, see this table or this report. For the 2011 NCES data, see tables 7 and 8 of this preliminary report. Additional statistics can be found in Trends in School Choice, 1993 to 2003.