What Hard Evidence Is There for the Efficacy of Homeschooling?

Anyone reading this blog will quickly surmise that I am in favor of homeschooling as a viable educational alternative. While I wouldn't presume to advocate homeschooling for all—it's options I favor, not absolutes—I do believe it can be an excellent choice for those who seek an educational experience that is flexible, customizable, and highly individualistic. I'm encouraged by what I've seen work in practice: I've successfully homeschooled my own three children, and I've assisted and known many other homeschooling families who likewise have done quite well.

For the skeptics, there are also books filled with personal stories, such as Homeschooling for Excellence (already mentioned in a previous post), Homeschoolers' Success Stories by Linda Dobson, A Sense of Self by Susannah Sheffer, and Real Lives by Grace Llewelyn—to name just a few.

So, why should I bother asking whether there is "hard evidence" that homeschooling "works"? Isn't the answer self-evident? Well, no, not really.

The Difference Between Anecdotes and Evidence

In their book Think Like a Freak, Leavitt and Dubner state, "An anecdote is a snapshot, a one-dimensional shard of the big picture. It is lacking in scale, perspective, and data." A homeschooler talking about how her kids learned algebra is sharing an anecdote. She cannot tell you to what extent other homeschooled children mastered algebra as a direct result of using an identical approach, nor can she say whether an alternative approach might have worked better. While her anecdotes might be helpful to other homeschoolers who are looking for methods to try with their own kids, it can't be construed as proof that this particular method will work just as well for anyone else who tries it.

So, how do we determine whether homeschooling in general, or a specific homeschooling method in particular, is effective? Is it really true that children do well because they are homeschooled? How do we know the same people wouldn't have done equally well or even better in school? (Of course, a similar question can be asked about schooled children: how might those same people have done if they had been homeschooled?)

Everyone knows there are many successful people who were homeschooled as children, but correlation doesn't prove causation, as the following cartoon humorously points out:

An Impractical Research Proposal

What we need is a controlled study.

For example, let's start with a set of identical triplets (extremely rare) and treat them exactly the same in every way except for the type of schooling they receive (a level of consistency which, any parent can tell you, is practically impossible to achieve).

We might then be able to separate the impact of homeschooling from everything else.

As far as I know, this type of study has never been done—and it's unlikely it ever will be—for an obvious reason: parents don't like to gamble with their kids' futures. When parents choose one educational alternative over another, they generally have very good reasons for doing so, which is why it's hard to imagine how someone could be convinced to choose a different (theoretically less desirable) option for only one child out of two or three, especially when the children are genetically identical with very similar educational needs.

Faced with this predicament, how might a researcher assess the efficacy of homeschooling? Such an assessment would need to begin with a clear definition of "homeschooling." It would be necessary to identify and control for confounding variables, such as family dynamics, educational philosophy, methods of instruction, materials used, quality and quantity of parent-child interaction, and all the other details that define a homeschooling lifestyle. Perhaps some forms of homeschooling are highly effective while others are not.

Unfortunately, there's not yet enough information to come to any conclusions.

My Anecdotes Are Not Evidence

All I know for sure is this: my kids and I enjoyed homeschooling and learned a lot. Although I can't clearly identify exactly which aspects of homeschooling were effective, I can look at what they've been able to accomplish, both during and after our years of homeschooling, and honestly declare that they've done "well": they've earned high marks on nationally standardized achievement tests and, subsequently, at competitive colleges; they've formed healthy, longterm relationships with their peers; they've never been arrested or even accused of a crime; and (so far) they've been able to find and keep decent jobs. They've also received positive feedback and praise from their teachers, classmates, employers, and coworkers.

What's more, they seem to be happy about having been homeschooled.

I'm very grateful for my kids' success and relieved that the dire predictions of how I might be ruining their chances at a good life by homeschooling have not come true. I'm reluctant to draw broad conclusions from my experiences, though, because I realize things could have easily gone differently. What if my kids hadn't done well on tests, didn't choose to go on to college, or couldn't find jobs or mutually satisfying relationships? I already know homeschooling parents who are second-guessing themselves now because their grown kids are struggling in one way or another. Was it a mistake for them to homeschool? Should they have sent their kids to school instead?

To what extent did homeschooling determine the outcomes? Who knows?

A Call for More Research

Without a controlled study, it's anyone's guess as to why some students do well and others appear to flounder. Any attempt to claim that formal schooling would have produced better results than homeschooling is nothing more than conjecture. And, to be fair, any attempt to claim that homeschooling produces better results than schooling—without solid evidence of causation—is also conjecture.

When all we've got are collections of anecdotes as "evidence," how can we be sure it was the decision to homeschool that resulted in an individual's success (or hardship) and not something else? If we are to judge whether homeschooling is efficacious or disastrous or something in between, we need real data from research that takes into consideration all the confounding variables, including genetic predispositions, socioeconomic conditions, geographic location, family circumstances, and even plain old luck.

With all the advice freely given on whether and how to homeschool, it would be helpful if there were some way to distinguish the good advice from the bad. Research on homeschooling might also help address the questions schools have been asking for years: what type of teaching works best? What makes some schools seemingly more effective than others at graduating students who go on to become successful in college, work or life in general?

If we can identify the most effective practices, perhaps we can share them and improve the educational outcomes for children in all settings.