Back in the 70's I took a graduate course in 'Quantitative Methods in Psychology'. It is telling that all the students were from applied math - not psychology - though that is a subject for another day.
We learned of a survey that had been taken of college faculty members. Here is the situation: you have done a study to test a hypothesis. The hypothesis fails. What do you do? The options are: drop the hypothesis, test the hypothesis again, publish preliminary results and publish as fact.
I don't remember what the exact results were, but something like 70% said to publish in some manner, with something like 20% saying publish as fact. Bear in mind that this is a survey of faculty. It is clear that publishing is considered to be so important that it is encouraged even when there is nothing factual to publish.
Even worse is that a significant number of our academics believe in publishing a hypothesis as fact even when the data does not support it. This survey was taken over 40 years ago, and I am sure the results would be worse now.
It seems that the search for scientific truth is not the driving force behind much of today's 'science'. Rather, studies are frequently done to support a predetermined outcome. In the case of global warming, the data is typically 'adjusted' in order to achieve the desired result. In the case of same sex marriage, the data may be completely faked as was the case in a recent published study.
Skepticism is always warranted when a new study receives major publicity, even when that study supports your own views (though for conservative Christians such studies are rare).