[agentbellnorm]

Other than some interesting themes and anecdotes, I’d be very careful to draw any conclusions based on the survey.

There are many intuitive reasons why unsuccessfully unschooled people would not answer this survey.

A huge concern with unschooling is lack of social development, and the respondents are found through social networks.

In the worst case, suicide would prevent someone from answering the survey.

The title is misleading because the survey can’t possibly answer the question with a survey like this, and without addressing the issue of equal opportunity, which might be the biggest factor when designing education.

[anelson]

Granted this survey isn't robust enough to be taken as evidence of the superiority of unschooling or even homeschooling generally. It's anecdotal, and if I had to guess I'd say the target audience is parents who have already decided to unschool and want some social validation, or at least parents leaning that way.

That said, I think your huge concern is that unschooled kids are at a higher risk of insufficient or delayed social development, and a survey that recruits on social networks will biased against such socially underdeveloped children. And furthermore that those socially underdeveloped children are at a higher risk of suicide than their conventionally-educated peers such that a survey which doesn't count the dead ones is further biased.

If that's a correct assessment of your concern then I'd be really interested to know upon what facts or even anecdotal experience it's based. If it's just that the idea of unschooling seems suspect, or that you can't imagine how such a system could produce better outcomes than conventional education, then you're certainly not alone in that position but I would submit that you are misinformed.

In my experience the social (to say nothing of intellectual) development of the average American high school graduate was laughably primitive when I was that age 20 years ago, and nothing I've seen since makes me think it's gotten any better. As for suicide risk I find it hard to believe that an educational style that presumes a higher than average degree of parental involvement and an almost certain lack of bullying and abuse in school leads to a higher propensity to suicide. In fact I would confidently wager that, were such data available, it would show the exact opposite.

[agentbellnorm]

I do not have have any data, and maybe you are correct.

But I am not claiming to answer a research question. To answer the research question, you need a better method.

I am just pointing out the bias.

[iateanapple]

> without addressing the issue of equal opportunity, which might be the biggest factor when designing education.

Why would that be a factor at all?

[agentbellnorm]

Because all children deserve a decent education where they are not separated by their parents income.

I definitely think desireability needs to be part of the conversation.