This is a tough question. Not all sources are created equal. Just because a journal article has been peer-reviewed doesn't mean it's accurate or true. Reviewers don't replicate the work they're reviewing and they don't recheck all the data. Plenty of papers pass peer review in top journals, only to be disproved or even retracted later on, for any number of reasons. And it's well documented that the scientific literature has a problem with, well, junk science. Peer review is far from perfect, but it does convey some added level of authority to what is published. On the other hand, it also tends to constrain "out of the box" thinking that may be truly innovative and ultimately revolutionary.
Ultimately it's up to the reader - that's you - to separate the good research from the bad. Experienced scientists use their accumulated judgment and knowledge, as well as an outlook of informed skepticism, to decide quickly if a particular paper is worthwhile or not. But what are students to do? Here are some questions you can ask yourself.
Like articles, not all journals are created equal. Major journals in every field are recognized for their overall quality and authority (even though bad articles can appear in them too). In analytical chemistry, these are some of the best known journals:
They come from reputable publishers such as the American Chemical Society, the Royal Society of Chemistry, Elsevier, Springer Nature, Wiley, AOAC, the Electrochemical Society, and the like.
"Reputable publishers" implies that there are also disreputable publishers. That's true. But there is a difference between "obscure" (that is, not widely known or read) and "disreputable" (known to publish fake or junk papers, or vanity presses that publish anything for a fee). This distinction may not matter much to your audience or instructor. When in doubt about a journal, it's OK to ask your librarian or professor for advice.
You can easily determine this by looking up the article in Web of Science or Google Scholar. Articles with lots of citations from later works are probably more reliable than those with few or no citations. But recently published papers may not have accumulated many citations yet, and sometimes citations are negative - a controversial or disproved paper might have many citations that refute its claims. It's even possible that the highly cited paper has since been retracted!
Some articles just seem "off" from the start. Perhaps the data don't make sense. Perhaps the conclusions don't seem to be supported by the data or evidence. Important data may be unclearly presented or missing altogether. The methods may not be clearly described. There isn't really any formal rubric for evaluating physical science articles in this way, but you can see some guidance here: A Rough Guide to Spotting Bad Science.
Bad science - Although scientific dishonesty gets a lot of media attention these days, outright fraud is actually rare in the literature. Bad articles are more often the result of sloppy experimental design, imprecise writing, or lack of adequate peer review or editorial oversight.
Fraud - One recent study estimates that over 2% of the scientific literature is fake: articles that are generated by digital "paper mills" and sold to unscrupulous authors looking to pad their CVs. The rapid rise of AI is making this problem much worse very quickly. Fraudulent research (which can include fabricated or doctored data, plagiarism, complete fabrication, etc) can be present even in the most prestigious journals, although it's more likely to be found in journals of lesser reputation, where it's easier to get accepted.
Bad writing - Poor grammar and language usage is often a giveaway to a paper that hasn't been properly vetted or edited. Really weird language is a clue that translation or AI software might have been used to write it, possibly to mask plagiarism. If you run across an article like this, it's a good idea to move on to another article.
Learn more about retractions, fraud, and other shenanigans at Retraction Watch.
There are no hard and fast rules, but in general, these types of information aren't appropriate for scholarly citations:
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 Generic License.