Treecritter- your "Then it slowly builds up" comment concerns me. How slowly? Is there a way to test out of material or do you have to cover each lesson no matter what?
This seems like the key issue. And like the 408 example, it's hard for a computer to know when a kid has picked up a weird idiosyncratic idea, and my kid has picked up a few along the way. In fact I think being precosious makes one more vulnerable to the having weird mistakes because of the asynchrony. Of course it isn't easy for a human to spot these odd ideas either.
I remember complaints about my handwriting during 1st grade, leading to me sitting down with one of my mom's friends to check me. She observed for a while and finally asked me why this 8 here was so much taller than the 4 over there.
I replied: "Everyone knows that - it's because 8 is twice as large as 4."
She explained the convention of all numbers being equally tall. I think she even asked me how big 497 would be written in my system so I could see how impractical that would be!
Hence my maxim: Logical isn't the same as workable!
Interestingly - AR Math, according to the website, doesn't claim to actually teach kids either - only to diagnose, provide practice and accredit. There is lots written about teachers indroducing concepts to the whole class, and working in small group and individually as needed. It is said to 'work with any textbook' which should tell us something about how similar every approach to math is in elementary school. Apparently it's more something to be practiced than understood.
For better or for worse, there isn't much in elementary school Math that can't be figured out on one's own. So another question becomes - how high does the AR Math go? What is planned if your son 'finishes?'
Love and More Love,
Grinity
