Wrong journal – happens that the paper, no matter how good, isn't a appropriate.
The research question has been studied before and this study didn't add anything – a "me too" study
The research design was not good enough to be a valid test of the research question. Common reasons include sample issues – biased sample or just not large enough to capture important effect – poor quality data – for example failure to use a measurement instrument that captures the construct well, or failure to measure potential confounders making the results uninterpretable.
I rarely reject a paper on statistical grounds. If there is a better statistical method that the authors should have used, I ask them to try it. Rarely, you get a paper where the authors are just playing with the latest statistical fad and the paper doesn't add anything to what we know. Those get rejected; new methods should yield new insights.
Ethics – rarely, but I've encountered papers where proper consent was not obtained, or where the study was approved as a survey but it was actually a clinical trial (only once in my life, thankfully!)
As a reviewer of scientific journals, one of the main reasons for the rejection of an article is the absence of scientific novelty, it means, that it does not contribute anything new to knowledge, as well as lack of impact in any of the scientific, social or economic spheres.
Also, if there is suspicion of plagiarism, it is cause for rejection.
If it does not comply with the ethical norms established worldwide or if it does not follow the scientific method, it means, if it is methodologically inconsistent.
There are validated standards in medicine with the CARE WRITE, for the documentation of medical articles such as case reports, etc.
Good morning, Thank you for this question, because the answers here indicate a diversity of opinion about peer review. When I have recommended edits to a paper before publication, it has been on the grounds of foundation for the research. For example, I have looked to see if the research summary at the start of the paper is adequate and accurate. In my opinion, research should not be ahistorical.