Ahh yes. Good questions.

I will answer Billy's question, which will lead to MT's quesiton, which I will also answer.

The short answer is that there are currently no experiments that confirm or imply that string theory is correct, and (as it stands now, at least) there are few prospects for such experiments in the future.

There are several things that string theory needs to exist---it needs extra dimensions (7 more, to be exact), and it needs supersymmetry. The problem is that the extra dimensions may be impossibly small to see, and that supersymmetry may be at an impossibly high energy to test. To make matters worse, because string theory is such an ill-defined subject, it may never be able to be proven wrong! To check for sure, we'd have to build a particle physics experiment that is about the size of the galaxy. Considering it takes a photon about 20,000 years to travel around the outer rim of our galaxy, this is an impossible experiment to preform.

The good news is that supersymetry may be found soon. The Large Hadron Collider (LHC) is opening next year in Swizerland at the Center for European Nuclear Research (CERN), and many physicists expect to find evidence of supersymmetry. But there is another problem---if we DO find supersymmetry, it is no guarantee that string theory is right! We can have theories that predict supersymmetry, but that aren't string theory.

There are other proposals for testing string theory---for example, there are some space-based experiments (like WMAP) which may show us signs of string theory in the sky. Signatures of a phenomena called "cosmic strings" would be very good evidence for string theory. Basically, if the universe expanded very rapidly at a very early time, some strings could have expanded in the same manner, showing up in astronomical observations today. The problem is the same---if we don't find cosmic strings, we won't disprove string theory

So, there are ways to test string theory, but no ways to prove (or disprove) it. The question is, why do people study it?

A few answers are popular: the first one, which I don't love, is that string theory is so beautiful that it must be right. We are astounded by the mathematical consistency of the theory again and again. Einstein was asked about general relativity---what if it's wrong? He answered "Then I would feel sorry for God". The point is, GR was so simple and beautiful that Einstein felt that it HAD to be right. The same is true for Dirac's theory of the electron.

This argument shouldn't convince you

The reason that I believe in string theory is the way that it naturally predicts the types of symmetries that we see in nature. We have found that forces are described by certain mathematical structures called Lie Algebras. Indeed, at least two Nobel Prizes (and probably more than that) have been awarded over such discoveries. Well, these mathematical structures naturally appear in string theory. We can know how to make quantum mechanics consistent with such mathematical structures, but we have no reason to do so---that is, the only way that we know such extended symmetries exist is that we looked in nature and found them. Quantum theory is just fine without such mathematical structures, but they are required by string theory.

The other popular answer to "why do it" is that strings is the easiest way to unify quantum mechanics and general relativity. (Other programs, like Loop Quantum Gravity, have potential, and I don't want to short-change these alternative approaches.) Finally, string theory tells us what the dimension of space-time is---eleven. General relativity is perfectly happy in any number of space-time dimension---even Einstein noted this problem early on. String theory only exists in eleven dimensions.

Hopefully I have clarified some questions and raised a few more. If you would like background on any of these topics (supersymmetry, general relativity, cosmic strings...) we can talk about them some, and you may wish to check out wikipedia.