I am working on a manuscript for publication in a chemistry journal. I find that my approach to the problem is very different to previous work on the matter. I find very hard to fit its argumentative line into the common scheme: Abstract, Introduction, Methods, Results, Discussion, Conclusion, Acknowledgements, References.

I am scared that my paper will be automatically rejected if it doesn't follow conventional sectioning. Is this an issue?

What should I do to increase chances of acceptance of my paper?

I can only think of the following alternatives, but I am not experienced and I am unsure of which is the best way to go:

  • Force the standard sectioning at any cost of the argumentative line?

  • Use many paragraph in the introduction to describe the sectioning?

  • Just use my criteria and structure the paper in the way I think it is better?

My thesis is that previous work did not consider some peculiarities and because of that their results aren't reliable. I need to prove the existence of these peculiarities and argue that my approach is better in these cases. Only experts in method X itself know that method X fails in some cases, and most researchers just apply method X assuming that its results are very good, so I feel myself in the need to show believers in method X that the method fails in certain cases (I'm not an experienced published author, so maybe my ideas are wrong about the need to convince others).

Finally, as there is not a method Y that works for all cases, I used results of many methods for each case and made a deep analysis considering the strengths and weakness of each method. Due to this, and the particularities of each case, I used a different method for each case. Even if I list methods for each case, I would have trouble fitting this into a Methods section early in the paper.

Due to the peculiarities I mentioned, I analyzed secondary properties as well. This gave rise to a lot of methods and situations that are hard to describe. I performed many calculations with a variety of objectives (first showing the existence of the peculiarities, and then calculating a useful value), and as many of the readers are not experts in the method itself, I think I need to provide a lot of background information before describing the methods themselves.

Note: The reason I chose a journal in which I know that most readers aren't aware of these kinds of situations, is that the properties calculated are of their main/exclusive interest. In other fields these kinds of things are well known and this work does not provide any insight on them. I'm just trying to correct previous values reported in one field, by applying well known facts from another field.

  • 1
    Hi, and welcome to Academia.SE! Unfortunately, I'm having some difficulty following your question. Can you please try to shorten and focus it a bit, and also say what field you are working in (publication conventions vary radically from field to field)?
    – jakebeal
    Nov 19, 2015 at 16:53
  • @jakebeal I hope that the changes done have made the question clearer and more useful. Nov 19, 2015 at 17:18
  • Most people use this structure because it works. Are you sure it isn't suitable for this? Nov 19, 2015 at 17:19
  • @FábioDias Thanks for reply. I am pretty sure. I can expand on why I think so. I can edit the answer including all the details that I removed (with the first edition) about why I think so, if you consider that this can be helpful. Nov 19, 2015 at 17:30

4 Answers 4


There are two entirely separate issues here:

  1. Does the journal require a particular format?
  2. What's the appropriate narrative with which to present your work?

Some journals require a strict organization of sections, while others merely suggest it. If it just suggests, you can do what you think best. If a journal requires a certain organization, though, you need to follow that organization. If it doesn't make sense for your narrative, then you might pervert the organization to better serve your narrative: for example, this paper of mine is about a methodology, so the method effectively is the result, and a rather large discussion of methods lives in the results section.

Another option is to make heavy use of supplementary information. For example, many high-impact journals require strict adherence to a format and also have tight length limitations. Supplementary information for such papers is often far longer than the papers themselves, and you can readily tuck in a pointer to a thorough explanation of methods by a reference to one of your supplementary documents.


I struggled with this very issue for my first couple of papers. I'll bet it's the middle part (Methods, Results, Discussion) that you feel doesn't fit the paper you're trying to write. That may because your view of the type of thing that goes into those sections is too narrow.

First, it may help to approach the writing a bit differently. Forget about the sectioning for the moment, and just write, in any order. Make the points you want to make. Add in the arguments or experimental results that you need to support those points. As you're writing, you realise that there are things your reader needs to know in order to understand your paper, so write that as well.

Now you can start to think about the order that the information should be presented in. So rearrange what you've written into logical paragraphs, and put them in a sensible order.

As you're doing this, a natural structure for your paper should emerge. There's stuff that the reader needs to know in order to understand your paper. That goes at the beginning, in the introduction. You'll also want a short summary of your main points. This will go in the conclusion. You might want to suggest areas for future research.

You may not need any acknowledgements in your paper. Ask your supervisor. You obviously need references.

The abstract is a summary of what the paper is about. I tend to write the abstract last, because it's only then that I know exactly what the "story" the paper is trying to tell is!

But it's the middle part (Methods, Results, Discussion) where the structure can be a bit confusing, especially if your paper isn't about a typical chemistry experiment. This seems to be the area where there's the most difference in structure between papers. For example, the results and discussion of those results might be combined into a single section, typically called "Results" or "Results and discussion".

So let's think of those sections a little more generally. You need to explain how you approached the problem. If your paper is an argument for something, then you need to explain your underlying assumptions, and how you reached the conclusions you reached. This is the sort of thing that goes into the "Methods" section -- although you may want to give this section a different title.

For an argument paper, it might be difficult to separate the "Results" from the "Discussion". So perhaps those sections could be combined into one section that presents your argument. Look for some papers that present arguments, you'll get ideas on structure.

  • 1
    Thank you for reply! The issue is with sec. "Methods". My problem is that the paper is a deep discussion of each case of a series of compounds, so for each molecule I use different methods. Also, there are a lot of different methods for proving a common propriety and a lot more (tens) to try/show the influence of this properties in many others magnitudes. An additional problem is that most of readers of the field aren't familiar with the theoretical point, so I can't just say "For proving X I used the methods.." because 'X' and 'methods' aren't commonly known in the field. Nov 19, 2015 at 20:00
  • In summary, there are almost two hundred of methods used, many of them can be grouped and just cited but I still need to tell something about at least 15-20 (those that its results were analyzed and compared deeply), so I feel that would be a very long list of saying "For X1 I used Y1,Y2,.., For X2 I used Y3,Y16,...." But it will be lost the reasoning of why each method in each case (most time method Y5 was used due to result of methods Y7 and Y9 for example), so it only would be a large, confusing and non useful section. Should I still try to retain this structure? Nov 19, 2015 at 20:06
  • 1
    I feel much more natural listing only main used methods and include new methods as they have sense to get into scene. Is it a really bad idea? I already wrote the manuscript, but I am thinking that maybe it is better to rewrite it. Nov 19, 2015 at 20:08

I would suggest that you pick a small number of cases where Method X doesn't work. You can allude to there being many other cases (in other words, you may mention that there are many other cases, but don't list them in this first paper). Think hard about why the readers of Journal A will care that Method X fails on these particular examples.

Once you have selected the cases to include, and have identified the relevance to this audience, you will be able to pull out some of the material you have already written, to put into an article for Journal A.

If this is published and well received, you may be able to publish more examples in a series of future articles; an opportunity to present your complete compendium may also present itself at some point in the future; but my proposal is based on the motto, "Start somewhere."

For this (possibly first in a series) article for Journal A, Jake's suggestions are very good, i.e. find out if it's required to follow the common scheme (Abstract, Introduction, Methods, Results, Discussion, Conclusion, Acknowledgements, References); if so, create some sub-headings under "Methods" for the necessary background information.


TL, DR; Follow the norms of your discipline.

You have already considered the fact that if you do not follow the standard format used in your discipline, your paper will automatically gets rejected. Keep that thought and revise your manuscript accordingly.

I find that my approach to the problem is very different to previous work on the matter. I find very hard to fit its argumentative line into the common scheme

No, that sounds lazy. You are comparing apples to oranges. First, you can write your paper the way you want it and then you can rewrite it, following the standard format used in your discipline. Only then you can really compare which of the two formats suits your content better. Writing what comes natural (because it is easier) does not make your paper better. It makes it unstructured and hard to follow by the corresponding experts. And that usually leads to rejection. Also, the argument "I find very hard to..." is self-evident. Yes, writing papers is hard. Writing good or seminal papers is even harder.

In our first papers, some of us thought that we should write our paper exactly the way we wanted and came naturally to us. Guess what? Our initial manuscripts sucked. And then we improved and our papers got better. How? By reading good, seminal papers and following the example of prominent people and how they structured and organized their papers. You do not have to learn this by the rejection of your manuscript. This will be one of the many papers you will probably write in your life. Learning to write according to the norms of your discipline is inevitable (for your following papers) and is a skill that you should learn and practice sooner rather than later. And you should start with your current paper.

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .