The scientific approach to knowledge blends observation (empiricism), inductive logic, and deductive logic.
Step 1: You begin with some phenomenon that you would like to explain (i.e. you have some interesting data obtained through observation).
Step 2: You derive a theory that would explain the phenomenon (i.e. you use inductive logic).
Step 3: You then devise a way to test the theory. First, begin by assuming the theory is true (i.e. treat it as the premise in a logical argument). Second, assuming the theory is true, determine what else must also be true (using deductive logic). In other words, if your theory is true, what testable predictions can you make about the nature of the world, predictions that go into new territory, beyond the data that led to the theory?
Step 4: You then return to gathering data to test the prediction. If the prediction made by the theory is wrong, then the theory is discarded. If the prediction is correct then the theory is kept, and any other theories that failed to make that prediction are discarded. As long as a theory continues to make correct predictions concerning further experimentation it is retained. As soon as it fails to make a correct prediction it is discarded, and a new theory must be devised. Thus, while a theory can be conclusively proven to be incorrect, all theories that have worked so far are accepted only tentatively, they may have to be discarded some day in the face of new evidence.
Concluding thought
By the end of the scientific revolution the unique methodology of science set it apart from other approaches, not necessarily 'better', but certainly distinct. From this it seems reasonable to state that while science and religion are both philosophies, science is not a type of religion, nor is religion a type of science; just as turtles and birds are both animals, but a turtle is not a type of bird, nor is a bird a type of turtle.