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.

Teaching What You Don't Know Yet (But Want to Learn)

Something I experienced often as a home educator was the desire to help my kids learn about a subject that wasn't exactly my area of expertise. In general, I had three choices:

1. Connect with an expert. Find someone who was proficient and ask if they would be willing to offer guidance and respond to questions as my child learned. This was an easy and obvious method, but it was also expensive, especially if the expert was a professional educator.
2. Encourage autodidactic exploration. Provide books, software, study groups—whatever was needed to enable my child to be self-taught, to figure things out independently through a process of trial and error. This method worked quite well when my kids felt passionate about a subject or were intrinsically motivated to develop a skill.
3. Become proficient enough to be the teacher.  

Admittedly, the third option requires a bit of ingenuity. How could I teach something that I had yet to learn myself? The most sensible thing would have been for me to teach subjects in which I had received formal instruction in the past. To some extent, I did exactly that when I helped my kids with math, science, and writing.

At other times, there were subjects or skills I was happy to learn along with my kids. My geeky self reveled in the whole process of sifting through books, researching subjects, exploring, asking questions and (eventually) finding answers. I loved discovering things together. I savored the moments when my kids surpassed me and became the ones doing the teaching.

How Does It Work, Exactly?

When I agreed to teach classes to groups of homeschooled kids who weren't my own, I took the job very seriously. Essentially, I did what I had been doing all along with my own kids, but I approached the task in a more focused, organized way. I assigned myself the following responsibilities:

1. Choosing resources carefully. I researched books and related materials to find gems that would be worth using. This was probably the most time-consuming step in the process but well worth it to avoid the mind-numbing drivel that's marketed as "educational" but can only be described as tedious. 
2. Preparing a tentative reading list (supplemented by a movie list, project list, field trip list, etc.). These proposed lists would always be significantly more extensive than what I would ever end up using during the year, but they weren't wasted effort. Having a readily available list of alternatives made it easier to improvise and come up with substitutions when something that looked good initially was disappointing. 
3. Planning and preparing well enough to stay one step ahead of my kids. Before I could teach, my understanding of a subject had to be at least a few grade levels ahead of what my children were learning. This was easy when they were very young but got progressively harder as they got older. By the time we were homeschooling high school, I had to be consulting college and graduate-level texts. 
4. Studying in advance, then studying again. Prior to the start of the school year or semester, I would typically read the books (and watch the movies) I intended to recommend to my children plus one or more advanced texts. Then, during the year, I would re-read all the books in parallel, as my kids read them. So, for example, as they read about the American Civil War, I read about the American Civil War—in their books and mine.  
5. Continuing to learn alongside my kids. By re-reading, or at least skimming, all the books as my kids read them, I was able to keep certain subjects in the forefront of my mind so I'd be prepared to respond to questions and recognize opportunities for related projects or activities. So, while I might not have been an expert on the American Civil War all the time, I was at the time they needed me to be. I also used the internet and connected with experts to get answers to questions that we couldn't answer on our own. 

Did I do this for every subject, every year? Of course not. I picked the subjects I would "teach" from one year to the next, and then I worked with my kids to cobble together a reasonably well-rounded course of study that included autodidactic, experiential and teacher-led learning (for the subjects that I wasn't as interested in studying along with them).

The Challenge

Were there any drawbacks to this approach? The short answer is yes. I recognized all along that my relative lack of expertise could be considered a disadvantage—for example, a more knowledgeable teacher could have provided with confidence more immediate, accurate answers to students' questions. When I was asked, "Is this right?" I always wanted to respond with a simple yes or no. There were times when it would have been so much easier if only I could have known the answers instead of having to grapple and search and say "I'm not sure." What made me think I could teach history or psychology or political science without a degree in those subjects? Was I overconfident, unable to see this approach as a case of the blind leading the blind?

I think all home educators—likely all teachers—know what it feels like to be asked questions we can't answer because we lack the expertise in a particular subject. I can't shrug the feeling off as inconsequential; it can be painfully frustrating and discouraging to feel useless in that moment. However, I'd argue the experience makes us even more conscientious about checking answers to be sure the ones we do give are either accurate or clearly identified as in need of verification.

The Optimistic Model of Error

In her fascinating book, Being Wrong, Kathryn Schulz describes an "optimistic model of error." According to this model, there is value in making mistakes. Not getting things right the first time or all the time leads to all sorts of benefits. As Shultz writes in Chapter 14 ("The Paradox of Error"): 

"Awareness of one's own qualms, attention to contradiction, acceptance of the possibility of error: these strike me as signs of sophisticated thinking, far preferable in many contexts to the confident bulldozer of unmodified assertions." 

"When we are aware that we could be wrong, we are far more inclined to hear other people out. [...] In love, as in medicine, as in life more generally, listening is an act of humility. It says that other people's ideas are interesting and important; that our own could be in error; that there is still plenty left for us to learn." 

Being an effective guide alongside means being willing to admit you have "plenty left to learn." It doesn't mean there's no such thing as a correct answer. It just means you haven't cornered the market on them.

The Most Useful Lesson

I believe "I don't know" is a beginning, not an admittance of defeat. Maybe that's why I've always felt there was something beneficial about my not having all the answers as a home educator, because it allowed me to show my kids by my example that we all start out by not knowing. Instead of being an encyclopedia filled with answers, I was a role model for a work in progress. I showed them that solving problems, conducting research, and developing new skills is a process of trial and error, not something that gets handed to you in a neat little curriculum package. Of all the things my kids learned, I think that lesson has been the most useful to them.

Homeschooling and the Transition to College

It seems ludicrous now, but when I first began homeschooling, one question that arose was, "What about college?" My oldest child at the time was barely seven years old, and college seemed a long way off to me. Besides, I had other things to worry about, such as how I was going to manage to teach anything—never mind college-preparatory academics—to my precocious second grader when I also had an inquisitive four-year-old and a rambunctious one-year-old (not yet weaned) who were constantly vying for my attention.

I decided early on that we'd just have to take it one day at a time. We'd think about college "later." Still, the question of "What about college?" nagged me, especially when well-meaning friends or family members asked it.

So, when I had a spare moment to myself, I researched the topic. I started by reading Homeschooling for Excellence by David and Micki Colfax (1988). The book came to me highly recommended, and the authors wrote—without seeming to brag—about how three out of four of their children were accepted into Harvard after being homeschooled. (Not just accepted: I found out they went on to become doctors and lawyers. One of them is even a doctor and a lawyer.)

While the book reassured many who had doubted whether homeschooling could ever lead to college, personally I was overwhelmed and intimidated by it. It gave me an inferiority complex the size of the moon. The Colfaxes were incredibly industrious homesteaders: they built their own home from scratch, wired the power lines to it, grew their own food, and cleverly incorporated math and science lessons into practically everything they did.

Are you kidding me? No wonder their kids were accepted into Harvard!

It's time for me to confess: the closest our family ever came to building a house was banging a pile of boards together to create a modest tree fort (with heartfelt thanks to my husband for making that happen). On a good day, we were able to sprout beans between two paper towels, or keep a few "sea monkeys" alive, but we weren't exactly capable of self-sufficiency.

If our family had had to rely on our meager garden for food, we definitely would have starved to death.

Luckily for us, the ability to transition from homeschooling to college depends on more than "life skills." While my kids might be glad they learned how to knit, sew, cook and do their own laundry, it's unlikely those skills are what helped them get into college. As far as I can tell, the acceptance process seems to be all about extraordinary achievements and academic credentials, sprinkled with a little luck and storytelling ability.

In a recent article about "Highly Selective College Admissions for Homeschoolers" (from HomeschoolSuccess.com), homeschoolers are reminded that "competition is fierce" and "the fact that homeschooling is great and your kid is homeschooled isn’t enough." If your student is aiming for a top college, the article insists, there are a few things to keep in mind:

• College applicants, whether they are homeschooled or not, need a well-documented academic record with challenging courses. As the article points out, "A basic expectation for students who wish to be competitive in highly selective admissions is that they have a rigorous high school education with strong development in all of the core areas: math, English, social science, science, and foreign language."

• For homeschoolers who don't have an official transcript from an accredited high school to submit—as was the case for two of my three kids—the "documentation" can get complicated. Selective colleges typically want records that have been validated or provided by an outside source: letters of recommendation from teachers, mentors and coaches; community college or other formal course transcripts; evidence of awards won, especially at the national level; and various test scores. 

• And about those test scores . . . although some colleges are now "test optional," meaning they do not require SAT or ACT scores, "test optional policies may exclude homeschoolers." In other words, homeschooled students sometimes need to submit test scores that schooled students are not required to submit. For example, my son learned too late that Northeastern University (as of 2013–2014) requires applicants to the College of Engineering to submit scores for the SAT or ACT plus two SAT subject tests, one in mathematics (which he had taken) and one in physics (which he hadn't). These are in addition to: the required list of all textbooks used; the School Report portion of The Common Application (which includes a special set of questions required only for homeschooled students); and a transcript of high school courses, prepared by the parent or an outside agency.

• "Most [homeschoolers accepted to selective colleges] also have used the flexibility afforded by homeschooling to develop some special area of talent or extracurricular interest." For the Colfaxes, it was homesteading. Other homeschoolers have built robots, started a business, or excelled as artists, musicians or athletes. MIT, Yale, and Princeton all emphasize the importance of doing something extraordinary as a homeschooler, rather than simply assuming your homeschooling will serve as a Golden Ticket (Willy Wonka style) for admission. 

In spite of all these hurdles, it's certainly possible for a homeschooled student to be admitted and to excel in a selective college. Although the data is sparse on exactly how many and where homeschooled students attend college, there's this list of Colleges That Admit Homeschoolers as well as my own, anecdotal evidence: my kids were offered admission as well as merit scholarships to several colleges, including Simon's Rock College of Bard, UC Berkeley, Northeastern University, and Worcester Polytechnic Institute (in a year when only 10% of applicants were accepted). Their friends got into Vassar College, Bennington College, Berklee College of Music, and many other schools. Did these kids also receive rejection letters? Yes, just as their schooled peers did.

Homeschooled students can go to college. They can do well once they get there (for example, see Exploring Academic Outcomes of Homeschooled Students, Journal of College Admission, 2010). The real risk is assuming that the doors to selective colleges will be held open, or kept closed, simply because a student is educated in an unconventional way. It's up to students and their families to work hard, keep good records, be aware of any special admission requirements, and prepare the best application they can when the time comes.