As you scale in the academia field you get ask more and more often to review papers. At the beginning I felt it was my duty to contribute to the process of scientific communication, so I took this task seriously, accepting only to review the manuscripts I think I can help to improve, reading some extra papers, ensuring that the methodology is appropriate and expending time to write complete reports with positive suggestions.
However, some times I run into a manuscript with several methodological and statistical mistakes and poorly reported. Most of those works they lack, in my opinion, the proper rigour to be published, independently of the journal prestige. I have no problem when my opinions do not match other reviewers; that is why we need more than one. But when three reviewers agree that a manuscript shouldn't be published in that state and the editor also agree, the paper gets rejected and few months later gets published in another journal with virtually the same problems... What is the point?
If nowadays anything can get publish, why should we keep reviewing papers?