In order to write a lit review from papers which I already have, should I read all those papers deeply and summarize them, or can I just skim and scan and just pick the information that I need to connect it with other information available in another paper?

The main point that I want to understand, do I have to understand the paper fully that I took some things to write from? For example, If I found a paragraph in a paper and it is good to include in one of the themes, then do I have to understand that paper fully?

4 Answers 4


Short answer Probably not.

In order to write an effective literature review, you should definitely understand the material. So...

[D]o I have to understand the paper fully that I took some things to write from? For example, If I found a paragraph in a paper and it is good in include in one of the themes, then do I have to understand that paper fully?

I would say that you may not need to understand the entire paper completely, but you should be sure that you understand the primary points well enough to be absolutely sure that you are not taking the quote/paraphrase out of context. Again, as other replies point out, whether you can determine this from skimming will depend on your current level of understanding of the subject. I assume you have already read these papers, and have a fair knowledge of the content; however, I would still advise you to read carefully anything you will be quoting--this will improve the quality of the lit review you are writing (and anyway it is never fun to be called out for having misrepresented someone's paper!). Bottom line, unless you know the subject--and the relevant papers--very well, you will probably be better off with reading thoroughly.

  • For example, if I took points of some papers will the supervisor when they assess my lit review ask me about the whole paper?
    – Kaser
    Aug 30, 2013 at 18:33
  • @Kaser It depends...your supervisor may ask about the whole paper if 1) you made an interesting point, 2) it is something s/he has a deep interest in, or 3) it appears that you don't know what you are talking about. In any of these scenarios, I stand by the main point of my answer-- you should know the material you are quoting! Sep 1, 2013 at 23:27

It really depends on your level of understanding in the specific topic you are researching. Having said that, it does not hurt (and more likely benefit) you to read each article carefully - taking notes of the important points, terminology, equations and any other relevant aspect of the paper.

One effective means that I have found (has worked well for me), is to paraphrase these key points as you go, this resource about paraphrasing and summarising may be of help.


If the paper you are reading is familiar material, it might be easy to assimilate the data from scanning. If it is news to you, you may have to go slower. The answer is, read as fast or slow as you have to in order to actually understand the material.

  • Please look at the update of my question. Thank you
    – Kaser
    Aug 29, 2013 at 20:26
  • 2
    After your update, I stand by my answer. If you only find a paragraph that works for your project, you stand a good chance of taking it out of context. It could be that the author did not imply by that paragraph what you inferred from it. Aug 29, 2013 at 20:29

I would add to the above answers that it also depends on what exactly you want to cite. Is it the description of methods, results, conclusions or general discussion or something else? I can think of a couple of scenarios that I came across:

  1. I was writing about some algorithm on networks and cited a physics paper that used similar approach. I only skimmed the paper as just wanted to highlight that the method is widely used in various fields in various contexts.

  2. I cited two papers that conducted similar studies but obtained opposing results. Here I read the papers thoroughly, especially methods section, to pick out the differences and try to understand what was the cause of such discrepancies.

  3. I cited some general discussion and future hypotheses to test from one review paper. I liked the reasoning of the guy and I knew that he is one of the leading scientists in the field. I did not hesitate to cite his opinion because I had read a couple of his previous papers and knew his contribution to the field. I would probably be careful with citing opinions/conclusions from some obscure sources. If you don't know what brought the author to conclude something, you are risking that you cherry-pick nice sentences out of context (as already mentioned in previous answers).


You must log in to answer this question.

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