It's really due to the complexity of the field. No paper exists in a vacuum, and you have to study a lot to know enough about a specific field to know if a paper is good or not.
If you google "does X support Y" you can find at least one paper saying yes and one paper saying no. Ask anyone who works in the field, however, and they'll immediately be able to tell you why both papers are correct or why one of the two papers is iffy-science/didn't account for something.
People literally don't know good science. I can tell from experiments done and experiments not done whether the results support the authors conclusion. In almost every paper, there's at least one stretch conclusion that has to be asterisked.
Lay people can come in and read a paper and tell you what they authors say. But they wont be able to actually evaluate it critically without the authors influence.
Knowing the quality of the journal is important as well.
"A little knowledge is dangerous" is the absolute definition of retards these days linking papers that support their claims when really it doesn't, or its a sham paper that no author scienctists have been able to replicate.
t. cell biologist