Scientific Theory For Dummies

I think subbie makes an excellent point here, and I just wanted to add on. The discovery of contradictory evidence does not immediately render a theory false. In the same way that a single datum or experiment cannot "prove" a theory true, one bit of contradictory evidence cannot tear it down. Good theories are ones in which scientists have a high amount of confidence. Contradictory evidence only serves to lessen that confidence. Once it reaches a point where the contradictions outweigh the positive evidence, then the theory is done away with.A great example of this is the way phylogenetic trees are constructed. Each tree is a hypothesis of the "true phylogeny" that is the relationships between the taxa of focus. Each hypothesis (tree) is tested with the collected data (morphology, molecular sequences, etc). If the same tree results with each data set tested then confidence in that tree grows until it becomes the accepted theory of how those taxa descended from a common ancestor. Once, that tree is well established, it's perfectly possible that new data could be discovered (let's say, a trait not used before is now used to test the tree) that gives a slightly different tree. Taxa A, B, C, and F are in a clade instead of A, B, C, and D. This one thing does not mean your original tree must be treated as complete bunk, it could very well be that there's something odd about that gene (perhaps it's a homoplasy not a homology).A good theory doesn't have to be a perfect theory, it just has to be the best possible explanation we have right now for the evidence we have on hand.

---

We have many intuitions in our life and the point is that many of these intuitions are wrong. The question is, are we going to test those intuitions?
-Dan Ariely

Well, we all know that's bollocks. Plenty of theories have been gotten rid of. I would say that depends on the nature of the anomalies and the reproducibility of them. The onus remains to demonstrate that one anomaly was not due to some error on the part of the people would found it. If the anomaly is confirmed then, of course it require further investigation, possibly some tweaking of the theory, and if it is deemed significant enough, dropping the theory all together.But again, saying any contradictory evidence means the theory is falsified is not true. When scientists started looking at the quantum level, they found phenomena that contradicted the Theory of Relativity. Plenty of solid evidence has been collected, and we know that Relativity does not completely explain the physical properties of everything we can observe, but it still has its own positive evidence, and it is still a useful theory. Both Quantum Theory and the Theory of Relativity are incomplete, but that doesn't mean that are false.

---

We have many intuitions in our life and the point is that many of these intuitions are wrong. The question is, are we going to test those intuitions?
-Dan Ariely

Yes, but it doesn't mean it's completely falsified. There are ways in which it can be wrong, but also ways in which it can be right, so you can't just dismiss the whole thing.Stepping away from relativity and quantum mechanics, and speaking more generally:
There will always be SOME contradictory evidence for any theory. Some study will come out that goes against a prediction of a theory. However, like I stated before, 1 bit of contradictory evidence is not enough to falsify a theory that is backed by loads of positive evidence. Contradictory evidence can arise for a number of reasons involving incomplete knowledge or human error, and it's not always obvious that these are the reasons for the evidence appearing contradictory. So, until more evidence against a theory amounts, it can't be dismissed right away.

---

We have many intuitions in our life and the point is that many of these intuitions are wrong. The question is, are we going to test those intuitions?
-Dan Ariely