Science education needs to do a better job explaining some things. As another poster points out, there is no such thing as a fact in science, that is, nothing you can be sure of. You can think of anything in science as a proposition and a probability that it's true or false.

For example, is George Bush the president of the US? (Notwithstanding stolen elections, etc). The probability of this being true is very close to 100%, and the probability of it being false is close to 0%. But, it is never zero. There is a non-zero probability that any arbitrary crazy-ass thing you can think of is true.

The problem, though, is that the above is no way to run a thinking mind in the real world from day to day. You have to make real binary decisions about whether to do something or not do something, and you can't be constantly confused by doubting reality, so most of these probabilities from an epistemological standpoint can just be rounded off to a 0 or a 1 so you can get some actual work done.

Evolution is both a fact and a theory. From a fact perspective, evolution as an explanation for life has a probability so close to 100% that the remainder can just be discarded. From a theory perspective, evolution has tremendous predictive power.

In any case, I think the epistemological underpinnings of science are something that most people don't understand, and the idea that nothing is certain in science can be frightening and confusing (and misleading if used for devious purposes as the IDers do).