National Center for Education Statistics

Measuring the Homeschool Population

By Sarah Grady

How many children are educated at home instead of school? Although many of our data collections focus on what happens in public or private schools, the National Center for Education Statistics (NCES) tries to capture as many facets of education as possible, including the number of homeschooled youth and the characteristics of this population of learners. NCES was one of the first organizations to attempt to estimate the number of homeschoolers in the United States using a rigorous sample survey of households. The Current Population Survey included homeschooling questions in 1994, which helped NCES refine its approach toward measuring homeschooling.[i] As part of the National Household Education Surveys Program (NHES), NCES published homeschooling estimates starting in 1999. The homeschooling rate has grown from 1.7 percent of the school-aged student population in 1999 to 3.4 percent in 2012.[ii]

NCES recently released a Statistical Analysis Report called Homeschooling in the United States: 2012. Findings from the report, detailed in a recent blog, show that there is a diverse group of students who are homeschooled. Although NCES makes every attempt to report data on homeschooled students, this diversity can make it difficult to accurately measure all facets of the homeschool population.

One of the primary challenges in collecting relevant data on homeschool students is that no complete list of homeschoolers exists, so it can be difficult to locate these individuals. When lists of homeschoolers can be located, problems exist with the level of coverage that they provide. For example, lists of members of local and national homeschooling organizations do not include homeschooling families unaffiliated with the organizations. Customer lists from homeschool curriculum vendors exclude families who access curricula from other sources such as the Internet, public libraries, and general purpose bookstores. For these reasons, collecting data about homeschooling requires a nationally representative household survey, which begins by finding households in which at least one student is homeschooled.

Once located, families can vary in their interpretation of what homeschooling is. NCES asks households if anyone in the household is “currently in homeschool instead of attending a public or private school for some or all classes.” About 18 percent of homeschoolers are in a brick-and-mortar school part-time, and families may vary in the extent to which they consider children in school part-time to be homeschoolers. Additionally, with the growth of virtual education and cyber schools, some parents are choosing to have the child schooled at home but not to personally provide instruction. Whether or not parents of students in cyber schools define their child as homeschooled likely varies from family to family.

NHES data collection begins with a random sample of addresses distributed across the entire U.S. However, most addresses will not contain any homeschooled students. Because of the low incidence of homeschooling relative to the U.S. population, a large number of households must be screened to find homeschooling students.  This leaves us with a small number of completed surveys from homeschooling families relative to studies of students in brick-and-mortar schools. For example, in 2012, the NHES program contacted 159,994 addresses and ended with 397 completed homeschooling surveys.

Smaller analytic samples can often result in less precise estimates. Therefore, NCES can estimate only the size of the total homeschool population and some key characteristics of homeschoolers with confidence, but we are not able to accurately report data for very small subgroups. For example, NCES can report the distribution of homeschoolers by race and ethnicity,[iii] but more specific breakouts of the characteristics of homeschooled students within these racial/ethnic groups often cannot be reported due to the small sample sizes and large standard errors. For a more comprehensive explanation of this issue, please see our blog post on standard errors.  The reason why this matters is that local-level research on homeschooling families suggests that homeschooling communities across the country may be very diverse.[iv] For example, Black, urban homeschooling families in these studies are often very different from White, rural homeschooling families. Low incidence and high heterogeneity lead to estimates with lower precision.

Despite these constraints, the data from NHES continue to be the most comprehensive that we have on homeschoolers. NCES continues to collect data on this important population. The 2016 NHES recently completed collection on homeschooling students, and those data will be released in fall 2017.

[i] Henke, R., Kaufman, P. (2000). Issues Related to Estimating the Home-school Population in the United States with National Household Survey Data (NCES 2000-311). National Center for Education Statistics. Institute of Education Sciences. U.S. Department of Education. Washington, DC.

[ii] Redford, J., Battle, D., and Bielick, S. (2016). Homeschooling in the United States: 2012 (NCES 2016-096). National Center for Education Statistics, Institute of Education Sciences, U.S. Department of Education. Washington, DC.

[iv] Hanna, L.G. (2012). Homeschooling Education: Longitudinal Study of Methods, Materials, and Curricula. Education and Urban Society 44(5): 609–631.

Statistical Concepts in Brief: Embracing the Errors

By Lauren Musu-Gillette

EDITOR’S NOTE: This is part of a series of blog posts about statistical concepts that NCES uses as a part of its work.

Many of the important findings in NCES reports are based on data gathered from samples of the U.S. population. These sample surveys provide an estimate of what data would look like if the full population had participated in the survey, but at a great savings in both time and costs.  However, because the entire population is not included, there is always some degree of uncertainty associated with an estimate from a sample survey. For those using the data, knowing the size of this uncertainty is important both in terms of evaluating the reliability of an estimate as well as in statistical testing to determine whether two estimates are significantly different from one another.

NCES reports standard errors for all data from sample surveys. In addition to providing these values to the public, NCES uses them for statistical testing purposes. Within annual reports such as the Condition of Education, Indicators of School Crime and Safety, and Trends in High School Drop Out and Completion Rates in the United States, NCES uses statistical testing to determine whether estimates for certain groups are statistically significantly different from one another. Specific language is tied to the results of these tests. For example, in comparing male and female employment rates in the Condition of Education, the indicator states that the overall employment rate for young males 20 to 24 years old was higher than the rate for young females 20 to 24 years old (72 vs. 66 percent) in 2014. Use of the term “higher” indicates that statistical testing was performed to compare these two groups and the results were statistically significant.

If differences between groups are not statistically significant, NCES uses the phrases “no measurable differences” or “no statistically significant differences at the .05 level”. This is because we do not know for certain that differences do not exist at the population level, just that our statistical tests of the available data were unable to detect differences. This could be because there is in fact no difference, but it could also be due to other reasons, such as a small sample size or large standard errors for a particular group. Heterogeneity, or large amounts of variability, within a sample can also contribute to larger standard errors.

Some of the populations of interest to education stakeholders are quite small, for example, Pacific Islander or American Indian/Alaska Native students. As a consequence, these groups are typically represented by relatively small samples, and their estimates are often less precise than those of larger groups. These less precise estimates can often be reflected in larger standard errors for these groups. For example, in the table above the standard error for White students who reported having been in 0 physical fights anywhere is 0.70 whereas the standard error is 4.95 for Pacific Islander students and 7.39 for American Indian/Alaska Native students. This means that the uncertainty around the estimates for Pacific Islander and American Indian/Alaska Native students is much larger than it is for White students. Because of these larger standard errors, differences between these groups that may seem large may not be statistically significantly different. When this occurs, NCES analysts may state that large apparent differences are not statistically significant. NCES data users can use standard errors to help make valid comparisons using the data that we release to the public.

Another example of how standard errors can impact whether or not sample differences are statistically significant can be seen when comparing NAEP scores changes by state. Between 2013 and 2015, mathematics scores changed by 3 points between for fourth-grade public school students in Mississippi and Louisiana. However, this change was only significant for Mississippi. This is because the standard error for the change in scale scores for Mississippi was 1.2, whereas the standard error for Louisiana was 1.6. The larger standard error, and therefore larger degree of uncertainly around the estimate, factor into the statistical tests that determine whether a difference is statistically significant. This difference in standard errors could reflect the size of the samples in Mississippi and Louisiana, or other factors such as the degree to which the assessed students are representative of the population of their respective states. 

Researchers may also be interested in using standard errors to compute confidence intervals for an estimate. Stay tuned for a future blog where we’ll outline why researchers may want to do this and how it can be accomplished.