Social Science Statistics Blog

I think a better control "group" for your cereal demo isn't a different cereal, but a non-magnet that doesn't pick up the possibly "sticky" Total.

You could even show an interaction effect by comparing Total to a sticky cereal with no iron (something corn-syrupy, I bet, like Golden Grahams?) which would stick to both a magnet and a non-magnet.

To reiterate a point made by Gary to me before I attempted this in my daughter's class: You must squish the cereal before the you do the demo. That is, you can squish some as an example for the class, but then you need to pull out a pre-squished bag for the magnet. The reason for this is that it takes a very long time with a moderate amount of cereal to get the pieces small enough that they will be picked up by the magnet.

But, even with this problem, the kids had a fine time. I look forward to Gary's future examples. Surely he has at least one for each year of elementary school . . . ;-)

well, you don't have to squish it, but with kindergarteners, its always helpful to be as prepared as conceivably possible. squishing on the fly is more convincing of course, so its preferable if you have the equipment, such as a really good rolling pin.

using a metal non-magnet would make a good control group too, tho you have to figure out if its as impressive to 5-year-olds.

(a 4th grade experiment will appear here in a few days.)