Not all journals are created (or operated) equally, to paraphrase the colloquialism. For example, anything that publishes
"Improving Homeopathic Practice Using Bayes’ Theorem and Likelihood Ratio" is more than a little suspect; homeopathy ignores all known physics and has been demonstrated not to work repeatedly. The practice of this has hardly been harmless fun, either; there are cases where it has clearly led to the death of an individual. Perhaps the saddest, for me, was the little girl who had a simple skin condition (
eczema) that a cream would treat who refused proper medical treatment until she wasted away and died, screaming in agony and pain, of massive infections (including one that cause one of her eyes to dissolve). Her parents were charged in her death (and, in true hypocrite form, had been shown to use real medicine on themselves when needed) but deaths like hers, and those caused by the anti-vaccine crowd, are why people like myself speak up when this nonsense is put forth. It's not just a harmless belief; it is causing real harm and impacting the most vulnerable amongst us. (reference:
http://www.smh.com.au/national/parents-guilty-of-manslaughter-over-daughters-eczema-death-20090605-bxvx.html )
Even within more mainstream scientific journals (one's where respectable papers are published - not vanity presses that seek the esteem of scientific trappings without the actual work involved) one must learn to spot the differences between a well conducted study and one with problems. Did they fail to account for their variables properly? Are they relying on questionable data sets? A small sample size? Questionable mechanism for their conclusions? Confusion between correlation and causation? There are a number of things one must look for in order to evaluate the claims. It's a lot of work but if you care about knowledge it's worth it (in my opinion).