I've been reading conflicting advice on SE about the placement of an illustration (figure, table, algorithm). My questions are thus (refer to Section 3.3.1 in the attached figure):

  1. I understand that, ideally, an illustration should always be placed inside the section where it is first referenced. Should an illustration always be placed in a section after it is referenced? Could it be placed in the same section before it is referenced?
  2. Can an illustration be the last entity in a section (as shown in Section 3.3.1)? Or does there need to be text after it, such as a short line on what is to be discussed in the next section?


enter image description here

3 Answers 3

  1. Figures always go after the first reference to that figure. Figures always have a caption. In almost all styles captions go after the figure, but before tables.

  2. Having a figure as the last thing in a section looks a bit odd and you might want to stick in an extra sentence just so it doesn't look odd. Having said that- you can avoid this by not dividing the text very finely (that is, no sub-sub-sections).


If you use LaTeX for your thesis, then figure placement is handled by LaTeX automatically with some degree of freedom. I.e. you have limited control on where exactly LaTeX puts the figure. As LaTeX is widely used in academia, and it was created to produce well structured scientific texts, the outcome of LaTeX having its way, should be ok with your university.

I am not aware of any guidelines on the very issue you are asking about, so sorry for not answering your question. Official guidelines usally cover the style and placement of the figure caption, e.g. Table 3.3. has to be bold and that caption has to be above the actual table. Generally, I would say the figure should be near the text passage which discusses the contents of the figure. As this is not always possible/feasible/sensible, I expect no hard rule to exist on this matter.

It is the standard behaviour of LaTeX with some document classes to flush all pending floating objects (figures, tables, etc.) before the next section/chapter. So, what you describe with Table 3.3. being the last item of Section 3.3.1 might be the standard set by the document class, it might also very well be by chance.

As for your first point, you can tell LaTeX to place a floating object not before the first reference to it.

Sorry for being very LaTeX centered, I haven't used Word much for scientific writing. When I used Word, I had non-floating figures, i.e. the placement was specified be me. Then my first rule for placing figures and tables was to make the paper "look nice", something that LaTeX does for you.

  • Downvoted. "Use LaTeX" is not an answer to the question. LaTeX offers essentially infinite freedom in placing figures, especially if you use packages like float, floatflt, or placeins. On the other hand, "Use LaTeX's default's behavior" is an answer but not a good one — Why should Leslie Lamport get to decide what my document looks like?
    – JeffE
    Commented Jan 3, 2017 at 23:35
  • @JeffE: While I agree with everything you wrote, at least some of the time there's a reasonable case to be made for: "Use LaTeX's default behavior for the layout and spend your own time focusing on content." (Not that I often follow that approach myself.) Commented Jan 4, 2017 at 1:31
  • @JeffE: In fact it is programs such as Word, that give you infinite freedom of placement. LaTeX comes a certain set of rules and some accompanying leeway. So in a sense, in absence of guidelines from my university regarding thesis formatting, I am way better off to hand in the output of LaTeX, than whatever I managed to produce on Word.
    – Dohn Joe
    Commented Jan 4, 2017 at 8:22

You have to check your department guidelines for thesis, so the answer is, it depends! I would never place it before, and generally any figure has a short text that goes with it.

  • Not all universities or departments have guidelines specifying such details: it would be better to provide a more complete answer. Commented Dec 4, 2016 at 19:14
  • @MassimoOrtolano please feel free to edit my answer. To me it seems one should do what's more readable in the absence of guidelines. Commented Dec 4, 2016 at 19:21
  • If there are no official guidelines, check the recent theses from your department. This gives some guiding on the accepted style.
    – Dohn Joe
    Commented Jan 4, 2017 at 8:24

You must log in to answer this question.

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