Reply #135: See, the way to study this, and it's not that hard, is to

[Snow]

gather a group of people with cancer, another group without, and compare the use of lawn chemicals in both groups. If the cancer patients have used more, then you have proven your hypothesis. What your or anybody else's anecdotes lack is that bit about the rate to which non-cancer patients also use lawn chemicals.

I realize this is vastly oversimplifying the exposures, ie the chemicals can drift, the dog can bring them in, how many years elapses from exposure to cancer, so forth and so on, but the point I want to get across is that observation of what seems to be a cluster of cancer cases in an area of heavy lawn treatment is not enough to test the question of "Does lawn treatment raise the risk of cancer?" There's an implicit comparison, and until you collect exposure information on people without cancer, you can't address the question.

Then of course there's the question of other exposures - viz, do people who treat their lawns and sport Coleman yard signs also smoke cigarettes? You get the idea.