The Scientific Method is often taught as a linear process that proceeds from hypothesis to theory. In practice, science -- like any other human endeavor -- can be much more haphazard. Science isn't perfect, but it has some built in processes for error correction to help weed out mistakes and outright fraud. There are some well-known issues with the traditional peer review
mechanism, but science is adapting and coming up with new ways to improve its results and conclusions. Here are just a few links on judging the quality of science.
If you'd like to read more awesome and interesting stuff, check out this unrelated (but not entirely random!) Techdirt post