[Eepsy]

This year, he stretched the lesson by a day and had students work in groups to try to draw the same type of graphic using the heights of the 15 boys in the class.

“Eventually, they figured out they couldn’t because the sample was too small,” Mr. Rios said. “They learned that the size of the sample matters, and I didn’t have to tell them.”

Wow, he managed to do in one whole class period something that used to take all of two minutes. He's certainly gotten his class deeper there- an object lesson in statistics and opportunity cost. That's two for the price of one....

[Othniel]

I think I see where this is going. The math requirements are modeled, in part, by the Japanese and Indonesian way of teaching math. They do go deeper into a concept, mining it as far as they can. A lot of the lessons I've seen deal with one or two problems. The problems are presented, the kids solve them, and the rest of the time is spent discussing why or why not different solutions work. Homework is often five problems, tops. Compare that to the 35 or so number-crunching problems that are presented in the US texts, with four deeper word problems at the end, and that's if you're lucky.

Which way of teaching will make it stick more in the kid's head? Which one will lend itself better to expansion of concepts? Not the way we do it now, that's for sure!

There are also fewer "standards" to cover, which means the ones that are there will be covered deeper and better. Also, the kids are tested after 25%, 50%, 75%, and 90% of the school year. This means a better picture of how the kids are learning than having one biga$$ test at the end of the year.

Just my two cents. (Or $40,000, if we're using Obama money...