Of fact and theory
First a definition for evolution must be established. Contrary to what Kent Hovind and his sycophantic minions say, evolution is not a theory of everything. It doesn't explain the big bang, nor does it explain the formation of stars or planets. It's a theory that explains the diversity of life. And the mechanism for that diversity according to evolutionary theory is natural selection. Natural selection is not a random process, it's the opposite. From wikipedia:
In biology, evolution is the process of change in the inherited traits of a population of organisms from one generation to the next.It should be a simple concept to understand. It's nothing at all to do with the big bang, it's surprising that creationists take the concept as being so holistic. Evolution explains inheritable traits through mutation and natural selection that eventually leads to the diversity of life. In terms of explanation, it's very much the unifying theory of biology. Nothing in biology makes sense except in light of evolution. But this doesn't mean that evolution is assumed true for all experiments in biology.
A misconception about evolution is that it's not falsifiable, and therefore not science. But evolutionary theory makes many predictions, as all science does. It's predictive of what to find in the fossil record, it's predictive of the types of mutations an organism should experience, it's predictive of patterns that should be found in ancestry. Now the science behind each of these concepts stands on it's own. i.e. evolution is not required when finding fossils, nor is evolution needed to explain mutations or ancestry. Thus is the difference between theory and fact in science. A theory is an explanation for the data, it predicts what data should be found, but the facts exist independently. If the facts don't fit into the theory that encompasses it, then the theory has to either be modified to support the new data (again by making predictions), or it is thrown out and replaced by a new theory that can explain all the new data. This is the case with plate tectonics, it superseeded Continental drift. Continental drift could not adequately explain the data and thus was replaced with a theory that could.
New data is coming in all the time, there are millions of scientists worldwide doing tests on observed phenomena. No theory is static, nor is it complete. New data can fill holes in gaps of knowledge, likewise it can cast doubts on elements of the theory and call for modification. It's all part of the scientific process and not a problem for the practical applications of the scientific method. The uncertainty that comes with a theory is not a drawback of the scientific method, rather it allows for the possibility that new data can shatter a concept. And that is so important, and it's what separates scientific knowledge from dogmatism.
Lines of evidence
First a look at the fossil record. We can learn from the fossil record at what time certain species lived through Earth's history. Evolutionary theory predicts that many species that once lived but have since become extinct should be preserved in ancient rocks, and that some of these fossils should be "transitional" forms, in that they show features that transition from one species to another. So this is how we know that evolution is correct when it comes to the fossil record. We see invertebrates come before vertebrates, we see vertebrates before amphibians. We see amphibians before reptiles, and finally we see reptiles before both birds and mammals. Nowhere do we see any transitional species between mammals and birds or vice versa. All this evidence stands on it's own, it's just explained by evolutionary theory in the falsifiable predictions the theory makes. Even without evolution, there is still a progressive fossil record to explain; it doesn't simply go away if evolution does.
Next a look at mutations. When an organism reproduces, most species have male and female reproductive organs. So any offspring gets 50% of it's genetic code from it's father and 50% from it's mother. But children aren't carbon copies of their parents, while most DNA is transferred successfully, mutations can and do happen. Not all mutations happen in the reproductive cells and aren't passed down, but those that happen in reproductive cells are. The theory of evolution states that these mutations happen, and it's the basis for the theory. That advantageous mutations will mean an organism is more likely to reproduce than organisms with bad mutations. Now it seems that strict adaptation is not the only means of mutations getting passed on, a lot of mutations are neutral and get passed down through generations through a process called genetic drift. Again these mutations stand on their own, and still happen regardless of whether evolutionary theory is correct.
Now a look at common ancestry. Through looking at patterns in DNA, similarities and differences can be noticed between parents and offspring. Applying it further back should be able to show ancestry for an entire species, and even across species. Evolution argues for common descent, and it predicts that species have a common ancestor and thus would be reflected in the genome. Just as we can see that all humans have a common ancestor this way around 140,000 years ago, we can see that humans and chimpanzees had a common ancestor around 6 million years ago. Again, even without evolutionary theory, this does not stop the science of common descent at all.
Finally a look at speciation. When two members of a species become different enough genetically, the offspring as a result of mating will be infertile. And this is exactly what evolutionary theory predicts. It's important to remember that speciation doesn't happen on a single individual, rather on a population. Say on an island there was a volcanic eruption in the middle of the island, and as a result there is a species of tortoise that is separated by the lava. On the north side of the island, the environment favours tortoises that have longer necks because the food is on branches while on the other side it favours tortoises with sharper beaks for eating seeds. Advantageous mutations for each group of tortoises will change the base genetic code that is seen in each population, and eventually with enough isolation there will be no means for the two groups to produce fertile offspring any more. And when that happens, it's called speciation. They are now two separate species, not one original species. This has been observed numerous times, and it still exists even without an evolutionary theory to explain it.
Fundamentalists get called on logical fallacies all the time, because they make so many. It's with that they try and fight back and call out scientists on logical fallacies. Everybody makes them, it's not exclusive to creationists. But like everything else, learning what is a logical fallacy and why takes practice. Just because you think it's circular, it doesn't make it so. It takes training to identify just what is fallacious and what is not. Likewise, without an understanding of the scientific method, calling out fallacies in science is seldom going to be accurate. Without understanding the logic behind a method, how is someone meant to call out problems in the logic?