My PhD thesis consists of three stand-alone papers (aka PhD by publication) which are interrelated and aim to address an overarching research question (RQ). Each of the paper answers its own research question(s) (e.g.rq1, rq2, rq3) and has everything that a journal article would have. Now I am at this stage where I need to integrate everything into the final discussion chapter of the PhD thesis, for example, discussing findings of each paper, describing strengths and limitations and implications etc.

I have done this for each of the stand-alone paper and now I am just confused:

  1. how should I avoid repeating myself in this final chapter? I've compared my findings with previous research and everything within each stand-alone paper and now I just don't know how I could make it less repetitive;
  2. My current structure is that I have the chapter listed in the order of paper. For example, 1.1 findings from paper 1,2,3; 1.2 limitations of paper 1,2,3 ...my supervisor suggested that I reorganized the findings into different themes and discuss them using the themes (which I interpret as "overarching findings"). I've been struggling with grouping these findings from all three papers into "themes" ("overarching findings"?) because while all three papers aim to answer a broad RQ, each of the paper has its own research question(s) and it's just so difficult to bridge them up. I would appreciate any suggestions that you may have and please feel free to recommend any books/articles that you found helpful!
  • "I need to integrate everything into the final discussion chapter of the PhD thesis...."—That's not really a "thesis by publication" then, is it?
    – Buzz
    Commented Dec 4, 2020 at 3:10
  • 2
    @Buzz Thank you and yes it is still a thesis by publication because it is required by the department that each paper within the thesis can be understood separately and also should be in a coherent narrative which means that we need to bridge up everything in the end.
    – Josee Luis
    Commented Dec 4, 2020 at 7:59
  • 1
    Why not shorten the discussion and conclusions of each individual paper, and move those discussions to the final concluding chapter? That will stop you being repetitive. Commented Dec 4, 2020 at 8:38
  • @astronat I think the papers stand as published. I do not see how we can answer question of this type without saying something very general or vague. The writing skill of the OP would determine the final results whatever we say. I am already lost by reading the question so I suggest OP to sit down with the supervisor or senior colleagues.
    – Alchimista
    Commented Dec 4, 2020 at 9:44

3 Answers 3


For any future people with the same problem, this article provides some very useful guidance on the topic. Its written in a nursing context but the guidance is more broadly relevant:

Lewis, K.B., Graham, I.D., Boland, L., Stacey, D., 2021. Writing a compelling integrated discussion: a guide for integrated discussions in article-based theses and dissertations. International Journal of Nursing Education Scholarship 18. DOI: 10.1515/ijnes-2020-0057

  • Great answer. I deleted mine. Commented Feb 27 at 10:57

A linear integration of the results/discussion in the three papers is not what a thesis should be. Those papers likely focus on three aspects of your original main research goal. Also, you probably broke up your main objective into three more easy-to-deal objectives. Therefore, the scope of your thesis is higher and wider than the scope of the individual papers.

The thesis discussion section must reflect that wider scope. Try to answer the following questions:

  1. How can you generalize the individual results concerning the main research objective?
  2. Are the individual results applicable to all the scenarios the thesis considers?
  3. What are the conditions that could potentially threaten the applicability or the relevance of each individual result?
  4. Can the systems/methods/methodologies return better results when the three studies are combined in some sort?

These recommendations are a little related to more practical or numerical areas but I'm confident you can translate their meaning to your field. Of course, you will repeat yourself but just enough to bring the main results of each paper into the table and then provide a high-level discussion in the thesis document.


Look at the discussion chapter as THE place to showcase your academic and conceptual thinking skills. Zoom out - look at the other chapters as a whole. What overarching question are you answering (presumably you also have a larger or smaller introduction chapter with an outline of this thesis or something similar)? How do the different chapters contribute to this? Where do they align/overlap/point in the same direction? How are they different? How are they maybe giving opposing answers (even more room for discussion). How does all of this fits in the literature? In the current state-of-the art? What new roads have you uncovered that could be followed up on?

The possibilities are endless, try to find a way in which you can make this chapter your own. Note that I've also seen many students struggling with this chapter after having been bogged down by the details for so long. In that case it can be difficult not to stay stuck at the experiment level - but IMHO the discussion is where you get the chance to shine conceptually and theoretically!

Good luck.

You must log in to answer this question.

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