Wednesday, 28 January 2015

Using Bayesien Logic to Decide if You Should Homeschool

"I can't homeschool"

"It's too difficult"

"I have to work"

"I don't have the patience, smarts, fill-in-the-blank"

So, you've thought about homeschooling . . . back and forth, pulling hair out until there are little bald patches appearing that can't be easily covered with a comb-over. You've read too many books and articles, and you still can't decide.

Have you thought about applying a bit of Bayesian thinking to the problem? What? Huh? It's OK, I'll walk you though it.

Bayesian thinking is a way of applying statistical probabilities and using that to help make decisions. I know, maths right? It's OK. It's actually simple.

First up, pick the chance of you homeschooling your kids. Let's say it's, 60% (yep, I made that up). That means there's a 40% chance that they'll go to school. Bayesian's cool with approximations and plucking numbers out of the proverbial.

Now pick a conditional, it could be, say, 'my kid will get bullied in school'. What are the chances of that happening? Well, according to this article, 1 in 4. So there's a 25% chance of getting bullied. Or 23.2- 27% according to this article, so 25% will do.

Then you can calculate the Bayesian:

Formula time!

x ~ old probability (my kid goes to school)
y ~ yep it's true (my kid is bullied)
z ~ no, it's not (my kid is not bullied)

Your new probability is: xy / ( xy + z(1-x))

So for our numbers, 0.4 * 0.25 / (0.4*0.25 + 0.75*(1-0.4))

(Arghh! It's numbers ... run!)

Which comes out at: 20% chance of your child going to school and then getting bullied. (I think, my numbers are pretty rusty here and I'm no statistician!)

The good thing, is that you can keep applying different probabilities - either looking up real stats, or putting numbers on your gut feelings. The numbers aren't really important. What is important is having a good look at each of the things that you fear could go wrong - or right - and quantifying that fear. Work through the many questions one at a time.

For instance, consider your child's level of giftedness / special need. How well would that be catered for in your child's selected school? If it's a select entry school and your child is moderately gifted, it might be great, say 80%. If your child is PG, or 2e and there's no IEP or gifted program, it might be quite low, say 10%.

You can then apply it to the running total: bullied in school + MG in good school: 55% chance of going to school; bullied in school + PG in inadequate school, 0.02% chance of going to school.

You can keep going, say assigning probabilities to your finances (such as the percentage chance of being able to find part-time work, or hire a babysitter, or do school work after work etc.) As long as the probability is not 100% or 0%, this method will work.

The first assumption about finances that needs to be dealt with depends on the current state of your family's finances.

  • If you have a two-income family, will one parent be giving up their income to home educate their children? Ditto for single parent families. Is this the most likely scenario?
  • Or are you able to do work part-time? Shift-work or weekend work? 
  • If you are in a two-parent family, are you able to work at different times? 
  • Do you own your own business? Are you able to work from home? Do you have an hours- or performance based job?
  • Are you able to homeschool in the non-work hours and employ a baby-sitter, relative or fellow homeschool parent for the remaining time? 

Also have a think about and assign a percentage to your competence, any tutor's competence, your child's need for direct instruction, i.e. learning disabilities requiring scaffolding, advanced/ accelerated learning requiring extra instruction, your chance of being able to find relevant material / tutor etc.

You can even make estimates on effective legislation: Is there a Special Needs policy? Is it enforced? Is there a gifted policy? Is it enforced? Is it funded? Are students who are twice exceptional granted extra services based on their ability compared to age appropriate norms, or to their potential output minus their disability?

First up, is there a policy or mandate in your state that says that schools have to provide gifted education? For me, in Victoria, the answer is, yes
Secondly, is it funded? In Victoria the answer is yes, partially. Most of the funding is for development of outreach/extension programs or teacher training. This amounts to about $1.4 million dollars + funding for guideline documents, which is unspecified.

What if your child is twice exceptional? Is there extra funding for their special needs?
In Australia, yes. There is funding within the school system, that is not available when homeschooling. The 2012 Students with Disabilities Rates, state that a school with child with a Level 1 Disability is able to claim $6095 in extra funding. A school with a child with a Level 6 Disability can claim $46,519 in extra funding.

Which might seem like a lot until you put it in perspective. The median full-time wage in Australia was $57, 400 in 2011. So a school with a student with a Level 6 disability - which would be quite severe, would have access to an amount of funding that would be almost enough to employ a full-time aide if they were on a median wage (which is unlikely if they have qualifications enabling them to work with special needs kids. Of course that is an assumption, and possibly they don't need qualifications and can be employed at a lower-than-median wage rate). Though it is more likely they would be employed only part-time.

Of course, these funding models are predicated on the assumption that the disability will result in an impairment compared to age peers in classroom. These models impact heavily on twice exceptional students who are able to 'compensate' their disability with their giftedness - which can actually be detrimental to their mental health.

Other questions to consider might be, how does home education compare to school-based education? To my knowledge, there have been no quantitative studies done on home educators in Australia (I'm happy to be proven wrong, I just couldn't find any), so I have had to have a look at American studies.

There are a number of studies that have shown that homeschooled students perform, on average between the 83 and 90th percentile depending on income and educational level of their parents. But theses studies do have the weakness of being a study of volunteers. A Canadian study attempted to compensate for this by having both homeschooled and public schooled families volunteer, they also then controlled for the income and education level of the parents. It found structured homeschooled students performed +1 grade level better on average. Unschooled students performed  at parity with public school, though the second result I feel should come with the caveat that the testing was done with Woodcock Johnson Test of Achievement, which would favour structured learning outcomes and completely ignore non-curricula self-directed learning. This could lead to a bias within the test against unschoolers. NB. These surveys of grown-up unschoolers are also worth a read (Also check out here and here) for a different perspective.

Then there may be more drastic considerations:
  • Is downsizing, or moving house to reduce expenses a possibility?
  • Would you have to do these anyway in order to put your child in a 'good' school?
Each of these can be weighted with different percentages. I know, it's numbers, but the point is to think about each point, weigh the pros and cons, and break the problem down into more manageable chunks of worry.

Then you will have a much better idea about whether homeschooling is right for your family. 
If you want more information on whether you can homeschool, consider reading these books (Disclosure - they are affiliate links):

No comments:

Post a Comment

Connect With Gluten-Free Mum