I was wondering when someone would get around to pointing out that the education establishment is constantly trying to confuse the public about its systematic failures. For example, although it had to do with private schools rather than homeschooling, a study done by NCES purported to find that, notwithstanding the fact that private school students clearly do better on the NAEP, the public schools really do a better job. The NCES models were "inspired" by similar work done by the Lubienski's (an H & W researchers who teach at an Illinois university). Of course, anyone familiar with the methodology used by NCES for adjusting for various factors could see that the studies (NCES and Lubienski) were just politics in scholarly drag. Fortunately, a fairly good analysis for the NCES fraud was done by Peterson and Llaudet at Harvard's Kennedy School of Government (one may ponder why the obvious problems with NCES and Lubienski haven't been raised by ed school researchers). While Peterson and Llaudet fail to take into account curriculum alignment and some other considerations that would bias the assessment of government school performance in favor of the government schools, their analysis is quite thorough with respect to the variables they do consider. The deeper point that they make, however, is that the methodology of the NCES (and Lubienski) study is entirely inappropriate for the task at hand. In fact, they express regret that NCES has failed to keep up with contemporary research standards. homeschool
http://www.ksg.harvard.edu/pepg/PDF/Papers/PEPG06-02-PetersonLlaudet.pdf
I suspect that Peterson and Llaudet understand that the NCES study problem is not really one of failing to keep abreast of research standards, but rather one of complying with the imperative to provide phony data to help prop up the public relations efforts of the corrupt and failing government school system. Which brings me to homeschooling. There is no body of research that I am aware of that shows that homeschoolers on average don't do better academically than government school students. When the data is disaggregated by parental education level, even mothers without a high school diploma do a far better job with their own children than government schools do on average. By income, studies indicate that most homeschoolers are middle to lower income families because they are mainly single income. Homeschooled children do tend to come from intact families, but more single parents are finding ways to homeschool, and I suspect that the results when studied will prove to be very strong. While teaching in a government school is difficult, to say the least, there is no point in trying to put lipstick on the pig. Homeschoolers tend to smoke government school children, but not because they are brighter or more advantaged as measured by SES factors. They do better because, among other things, their parents can adjust their children's curriculum to their strengths and weaknesses, curricula based on foolish ed school pedagogical theories are not mandated (whole language, fuzzy math, etc.), an extremely low student to teacher ratio, an emphasis on developing self-teaching skills, an abundance of sound, easy to use curricular materials, coop classes and similar arrangements, video lectures, and online courses (here is just one of the many resources HSers are able to draw on that government schools typically can't match
http://www.gbt.org/).
The schools today are mainly a mechanism for allocating about $600 billion a year to various special interest groups and for providing privileged ideological access to children for various leftist special interest groups. This is why a failed government school system is so aggressively and dishonestly defended. But, even if homeschooling didn't yield better academic results, most would still homeschool because of the various ways in which government schools are destroying children morally and spiritually. I can fix an algebra deficiency easily - bad character, once it has taken hold, is almost impossible to fix.