I'm writing a journal article that involves many comparisons within and across groups. Visually, these comparisons are quite easily understood. However, describing them in text requires the use of long sentences with several acronyms and numerous references back to sub-parts of the figures.

In general is it acceptable in academic writing to simply say a couple of high-level sentences (and maybe point out some nuance that a reader might miss) about a complex figure and allow the reader to absorb the bulk of the information by viewing the figure?

  • 7
    ... what about blind and color-blind researchers?
    – Bakuriu
    Commented Jul 26, 2016 at 8:53
  • 2
    My university required alt-text descriptions in the final PDFs for all figures that would allow someone using a screen reader to understand the content as well as a sighted person.
    – Kathy
    Commented Jul 26, 2016 at 16:26
  • It depends, if the figures are clear interpreted and if the paper draw the conclusions clear from the figures. To be sure, you can ask a researcher who has published papers in this field before on the clarity of the manuscript.
    – Nikey Mike
    Commented Jul 26, 2016 at 16:38

5 Answers 5


It is evidently true that figures/images can convey things difficult to convey in words (e.g., see Edward Tufte's The Visual Display of Quantitative Information).

On the other hand, the exercise of verbally approximating what may be visually apparent is in itself highly informative in various ways. Among others, your verbal approximation of the visual will partly explain to your readers your own interpretation of it, to which they can add or contrast their own. Or, they might otherwise fail to see in it what you see.

That is, with regard to the latter point, it is often good to "state the obvious", because the notion of "obvious" turns out to be subjective.

  • 2
    Another way to look at it is that if you do not give the 'verbal' (in writing) interpretations leading the reader through the figures, then it is equivalent to expecting the random reader to 'read your mind'.
    – Carol
    Commented Jul 26, 2016 at 14:33

Depends a bit on your field.

But, in general, I'd wager that it is a good idea to guide the reader through the image, towards the information you want to convey with the image. Not only by explaining, but also by annotating the image, exactly to ease such explanations. It is easier to reference/identify "the region denoted by (a)" than "the third peak from the left", for instance.

update (to clear editing confusion): What I said applies not only to the text that cites the image, but to the caption, and the image itself. Whenever you can make the image "tell a story", with a clear sequence of points of interest (guiding the reader through the image), it should improve the understanding of what you are trying to show.

An example of this idea:



The point is to be as clear to the reader as possible, yet as succinct as possible, especially when there are limits to the number of pages or words or figures you are allowed to use in the paper. In general, however, it is not a bad idea to repeat the same idea both in figure and text. Bear in mind that many of your readers may not be familiar with the specialised subject of your paper. Also the journal might have some guidelines on the matter.

  • 1
    A good diagram is like a well-made comic. It's a balance: The picture explains those things which it does best, the caption those which it does best and (here the comparison ends) the text explains the details too long for the caption. Redundancy is ok where it supports understanding. Keep in mind: nothing is more frustrating than a picture with unexplained features, especially if the paper is interesting. Commented Jul 26, 2016 at 8:55

I was told that article shall be written in the way that all readers, listed below, won't lose the take-home message I want to give. There are ones that:

  • read abstract only (they who are behind paywall);
  • read abstract, conclusions and introduction only;
  • read text only;
  • read figures and their captions only.

On the other hand, the figures are supporting the claims you are writing in the body; you don't need to describe the figures fully but you should write what the reader is supposed to see in it - even if it may seem dumb or overkill to you.


It is possible to define a diagrammatic terminology and then use it, as long as the translation is deterministic and unambiguous. See for example how the proofs on pages 14, 19, etc. are done in this reference.

You must log in to answer this question.

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