As a frequent reviewer and journal editor, I'm seeing (and reading) quite a lot of articles. Unfortunately, I am also seeing a number of mistakes that are repeated over and over again - both by experienced authors and novice writers.
If you are an early career researcher, and need some guidance on how to write an article that has higher chances of publication, this post is for you. If you are mid career, but sometimes find that reviewers comment on your writing style, this post may be a good reminder of what you should do and may provide pointers on what to improve in your writing.
So let's see some very common mistakes authors make when writing a journal paper:
An abstract follows certain rules. An abstract is not a short summary of what you did. An abstract is not the background to your work. Each abstract should contain the following 5 elements: motivation, problem statement, approach/methods, results, and conclusions. Very often, I read abstracts that are missing the conclusions sentence. If your abstract doesn't answer the following: "What are the implications of your answer? Is it going to change the world (unlikely), be a significant "win", be a nice hack, or simply serve as a road sign indicating that this path is a waste of time (all of the previous results are useful)?", then it is missing conclusions and is stylistically not an abstract.
Don't assume that your readers know the importance of the topic that you are studying. When you are introducing your work (in the introduction section), you need to address which main challenge your work is focused on. What is the practical relevance of your work? Why should this topic be studied? Why do we even care about this topic? Tell the reader why your work matters in the introduction, and tie back to this contents in the discussion section. Talking about previous experiments related to yours, formulas, theories etc is not general enough for introducing the topic.
I'm not sure if it is one of the maladies of my field, but if I got a dime every single time I receive or review a paper that contains an annotated bibliography instead of a literature review, I'd be a millionaire by now. Seriously though, a literature review reviews the literature - as the names says. There should be an analysis of the literature. What happened chronologically in your field? Who agrees and who disagrees? Can you summarize information in an overview table? You need to analyze the literature to write a literature review. An annotated bibliography reads as follows: "Author X tested Y and found Z. Author A tested B and found C." etc etc. Do not make this mistake.
A research report reports experiments or other work done. A journal paper goes one step further: you need to frame your work in the current body of knowledge, discuss your results critically, and frame your work in a theory or explain the mechanics behind what you observed. Just reporting results is not enough.
A thesis contains more information than a journal paper, and will contain more subsections. Keeping the same titles as a thesis, and then just summarizing the contents will make for a journal paper that is missing flow and has a chopped-up structure. Change the structure, and make your journal paper a document that can stand alone; not just a summary of your thesis.
I've already repeated this comment a number of times throughout. The literature review introduces the current body of knowledge. Where you discuss your methods, you should motivate your choices based on the current body of knowledge. When you analyze your results, you should link to the theories and models available in the current body of knowledge. In your discussion, you should explain how your work extends, contradicts, or confirms the current body of knowledge. Your work should be embedded in the field as much as possible.
Are you tired of writing by the time you get to your discussion section? I sure am tired of the paper that I am working on sometimes when I get to the discussion. But the reader shouldn't know about this. If you feel tired of the topic, put the paper aside for a week or two, and then return with fresh eyes. Your discussion should provide a critical reflection, highlight the lacks of knowledge in your field, discuss which practical applications your work has, link back to the importance of the topic discussed in the introduction, and embed your work inside the current body of knowledge. The discussion is not an afterthought. Often, the discussion is what sets your work apart from a research report.
The "Summary and conclusions" section should not contain new information, ever. If you want to discuss your results, do so in the discussion section. Your "Summary and conclusions" section is for the reader who wants to know if your paper is worth reading in depth. Bring together the most important points from your entire paper in this section - do not leave any section uncovered.
Just don't forget formatting your references according to the style guide of the journal you are submitting to. You may avoid a desk rejection or a decision to return the manuscript to you when you properly adhere to the style guide. Even when you use formatting software (which you should use), you still need to check if everything is formatted properly. If your entries are not 100% correct in your database, your reference won't be correct either.
Again, even when you are tired of a manuscript and just want to submit it, make sure you have a bit more patience and have all details sorted out. Work that is full of punctuation, grammar, and spelling errors looks sloppy. As a reviewer, it sometimes takes me a lot of effort to read work that is not properly proofread. And when the writing is so sloppy that the contents is obscured, I may feel like throwing in the towel and just saying that the authors should rewrite before I rereview. Make sure you submit well-written work so that your writing does not distract from the contents and your science. If necessary, get the help from a professional.