What do we know about how homeschooled children fare once they reach adulthood and enter the world on their own? The answer is a lot—and a little. We have a good amount of data on how well homeschool graduates perform in college, but we have comparatively little other data on homeschool graduates’ experiences as adults. We also do not know for sure what percentage of homeschooled students attend college. Anecdotal feedback from the first generation of homeschooled students, now in their 20′s and 30′s, indicates that those who are homeschooled responsibly frequently do well in college and professional life while those who were neglected or subjected to an abusive homeschooling environment often face low-wage job prospects, poor integration and connection with their communities, and struggles with poverty and dependency that could have easily been prevented. In this brief we will examine just what actual data we have on homeschool outcomes and homeschool graduates’ college performance.
It is important to remember that many children are homeschooled short-term, or are homeschooled part-time, which may complicate attempts to separate students into groups based on whether they were homeschooled or attended conventional schools. Further, finding representative samples of young adults who were homeschooled can be challenging. Research on college students with homeschool backgrounds is simpler and therefore much more common, but the population of homeschoolers who attend college is by its very nature not representative of homeschoolers as a whole.
Outside of studies of college students with homeschool backgrounds, there are very few studies of homeschool outcomes. It is our hope that comparing two of the studies we do have—one conducted by Brian Ray and the other by Cardus—will help shed light on the complexity of conducting research on this topic.
The first study, “Homeschooling Grows Up,” was published in 2003 by researcher Brian Ray. In this study, Ray analyzed surveys filled out by over 5,000 adult homeschool graduates, all of whom were educated at home for at least seven years. While this number sounds impressive, Ray did not use a random sample but instead gathered participants through various email lists and homeschool advocacy groups. Further, the vast majority of the participants in his study were in the 18 to 24 age bracket (only roughly 15% were over 24, and 2% over 29), which means that it was likely a bit early to study outcomes for many of these participants. Finally, Ray did not correct for background factors like parental income, education, or marital status. Ray frequently conducts studies of homeschooling in which huge majorities of participants are from white, middle and upper-middle class, well-educated, two-parent households, and it is likely that the demographics of this study’s participants were similar. This makes Ray’s constant comparison to the national average throughout the study, as well as his suggestion that his survey participants’ success was due to being homeschooled rather than other factors, both unscientific and incredibly misleading. At best, Ray’s study tells us how some homeschooled students fare in the years immediately following their graduation.
The second study, “The Cardus Education Survey,” was published in 2011 by Cardus, a Christian think tank. This study was designed to examine Christian schooling and compared public schools with Catholic schools, Protestant schools, nonreligious private schools, and homeschooling. Participants were asked either what type of high school they had attended or what type of school they had “primarily” attended. The Cardus used a random sample, corrected for background factors, and looked only at young adults aged 24 to 39. In other words, the Cardus succeeds where Ray fails. There are nevertheless a few limitations: The Cardus only included homeschooled students whose mothers attended church regularly, and it had a sample size of less than 90. Still, scholar Milton Gaither has called the Cardus “by far the best study we’ve ever had of adults who were homeschooled.”
Ray found that 74% of homeschool graduates aged 18 to 24 had taken at least some college classes while only 46% of the general population in that same age bracket had done the same. However, given that Ray did not even address background factors—such as the educational background of his homeschool graduates’ parents—this comparison is completely disingenuous. In contrast, the Cardus, which did correct for background factors, found that homeschool graduates were more likely to attend an open admission university, less likely to attend a prestigious university, and less likely to attend both college and graduate school than conventional school graduates. In a 2004 study, Belfield found that less than one third as many SAT takers self-reported as homeschooled students as should have been expected given the number of children homeschooled at the time. He also found that while homeschooled students outperformed public schooled students, when background factors like parental education, income, and marital status were corrected for, the difference between homeschooled students’ scores and those of their public schooled peers decreased markedly.
Ray reported that homeschooled students are significantly more likely than average to report being “very happy,” finding life “exciting,” and being “very satisfied” in their work. Yet without correcting for background factors—the income level of the families of these homeschool graduates, for example—this tells us very little. Roughly half of Ray’s sample were full-time college students, significantly higher than the average at the time his data was collected, which may itself explain some of the discrepancy in reported levels of happiness and satisfaction. In contrast, the Cardus, which did control for background factors, found that homeschool graduates were more likely to report “lack of clarity of goals and sense of direction” and “feelings of helplessness in dealing with life’s problems” than any other group. They were also more likely to marry early, to have fewer children, and to divorce.
Ray found that homeschool graduates had higher than average levels of civic involvement, including writing letters to the editor, voting, and engaging in political campaigning. Yet because Ray used a volunteer sample and did not correct for background factors, this tells us little to nothing. In contrast, the Cardus, which both used a random sample and corrected for background factors, found that homeschool graduates were actually less likely to be politically or civically involved in almost every area. Of course, the Cardus included only homeschool graduates whose mothers attended church regularly, thus excluding secular homeschoolers or those whose religious attendance was less regular. Still, the Cardus results suggest that the rosy picture painted by Ray’s study may lack needed nuance.
Over 95% of Ray’s sample reported being glad they were homeschooled while 74% reported that they planned to homeschool their own children. However, asking these questions of homeschooled students so close to their own graduations, before they have time to gain real life experience, may be misleading. Further, given that Ray recruited participants through homeschool channels, only those who still had connections with the homeschool world would have even heard of his study, meaning those who had cut all ties and not looked back would not have had the opportunity to participate. The Cardus did not ask this same question, but anecdotal reports suggest that some homeschool graduates are dissatisfied with the education they received at home or have chosen not to homeschool their children.
We cannot extrapolate from Ray’s study to homeschool graduates writ large. The Cardus is different, and to some extent we can extrapolate from it, although only to homeschool graduates whose mothers attend church regularly. However, because they were studying homeschool graduates rather than homeschooled children, both Ray and the Cardus were studying the results of homeschooling in previous decades rather than the results of homeschooling today. Further, the majority of children who are homeschooled are only homeschooled for a few years. There is a growing fluidity as parents move their children between public school, private school, and homeschooling, sometimes switching multiple times per child. This makes dividing children neatly into “public school” or “homeschool” groupings difficult and often simplistic.
Next we turn to homeschool graduates’ college performance. Research suggests that homeschool graduates who attend college do well academically and successfully integrate socially. Much of this research has relied on interviews, self reporting, and volunteer samples, which leaves some question regarding our ability to generalize from the samples studied. Further, homeschool graduates attending college are a self-selected group, given that homeschool graduates not academically prepared for college are not likely to apply or attend. Indeed, there are some indications that homeschooled students may be less likely to attend college than their peers, a finding that needs further research. Nonetheless, both quantitative and qualitative data suggest that homeschooled students who do attend college are indeed capable of succeeding both socially and academically.
Essentially every college or university now has policies in place for accepting homeschooled students, and many schools have admissions officers dedicated to homeschool admissions. In the past, most colleges required homeschooled graduates to take the GED before being admitted, but this practice has been discontinued and most colleges and universities are friendly to homeschooled applicants. Some admissions officers are very favorable to homeschooled graduates, even giving them preference, arguing that homeschooled graduates are especially well motivated and and have stronger academics than other students.
Both qualitative and quantitative data consistently find that homeschool graduates who attend college integrate well socially and often have higher GPAs than other students. The methodological limitations of these studies must be carefully understood. Many rely on self reporting and use volunteer samples, and many of those that avoid this problem by including a college’s entire homeschool population frequently fail to account for background factors. Even when these things are accounted for there may be questions about whether the results may be generalized to previously homeschooled undergraduates at other institutions. While research on homeschool graduates’ college performance tends to have positive findings, caution is needed to interpret its results.
Quantitative studies of homeschool graduates’ academic performance in college are often more nuanced than reported. For one thing, they are most frequently conducted on homeschool populations at private or Christian colleges, throwing into question whether we can generalize from their findings to all previously homeschooled undergraduates, especially those at public institutions. Further, at least one study has found that homeschool graduates tend to underperform in math and science and are are less likely than other students to pursue degrees in those fields. Taken together, however, quantitative studies of homeschool graduates’ college performance paint a positive portrait.
Qualitative studies—those that use a small volunteer sample and involve personal interviews—are instructive and often more nuanced. Some have focused on homeschool graduates who were especially motivated and able to access educational resources in high school, while others have found that homeschooled students were no different from other students, with a similar dropout rate. A number of studies point to early challenges to adjusting socially and culturally, but the general finding is that homeschool graduates do ultimately integrate socially into their new surroundings. Some studies have suggested that homeschool graduates generally retain the religious and political beliefs of their upbringing through college, are more likely to hold leadership positions, and struggle early on with writing essays and research papers. Overall, these studies paint a nuanced yet positive picture. Unfortunately, we cannot know whether the previously homeschooled undergraduates who serve as volunteer participants in such qualitative studies are representative of the larger population of homeschool graduates attending college.
The research we have so far suggests that homeschool graduates who attend college generally fare well both socially and academically. What is less clear is why this is so. One reason may be that homeschooled students who attend college are likely to have well educated parents and access to a greater array of resources and a larger amount of social capital. Another reason may be that homeschooling is especially conducive to some students’ educational development and achievement, offering them academic opportunities they might not otherwise have had and providing them with positive habits for lifelong learning. What is unclear is what the success of homeschool graduates who attend college tells us about homeschool graduates writ large. There are some indications that homeschool graduates may be less likely than other groups to attend college, and we know very little about the success of previously homeschooled students who do not attend college. In the end, the research tells us that homeschool graduates who attend college tend to fare quite well; it does not, however, tell us that homeschooling is a superior choice to other educational options.
As a final note, many students are homeschooled only for a short time, or are enrolled in public school part time. In an atmosphere with an increasing amount of educational choice, students may cycle through public school, private school, charter school, and homeschool over the course of their childhood and teenage years. Given this reality, studies on homeschool graduates’ college performance are often in a bit of a tight spot when it comes to determining who was homeschooled and who was not. Studies have sometimes limited themselves to those who were homeschooled for high school while others have included those who report being “primarily” homeschooled or, conversely, any student who was homeschooled at some point in time, no matter how briefly. In the future, researchers need to find a way to make sense of and interpret students’ increasingly diverse educational experiences.