It depends on:

• Rank of the journal
• Quality of the paper
• The final decision (Acceptance or Rejection)
• Simplicity and complexity of research
• Closeness to research interests of the reviewer 
• The degree of concentration during review time
• Having experience in reviewing papers

Shorter if prepared well

Dear Mahfuz

I think your question needs a little more details to give correct answer.

If you are referring to a journal paper, it may take anywhere from one to three weeks depending upon the content of the paper in terms of its subject matter. On the other hand if it is a research report depends on its content. One hour will be requires to quickly browse the contents and scan the abstract and conclusions.

My record was ca. five minutes. The text was a couple of pages and I suggested to publish it.

Certainly, the process varies between reviewing a conference paper and an article for publication.

As a reviewer of a 20 page paper - I typically spend 2-3 hours.  As an associate editor it runs longer - perhaps 4 hours.

Dear @Mahfuz, I have had a research projects to review, not articles and conference  papers only, and it takes a week or more to make a serious review (hundreds of pages!).

Suppose this is a journal paper of technical nature. The average time taken from the acceptance of the invitation to review to submission of my comments is 3 - 4 weeks. The average net time taken to review it is perhaps 4 - 5 hours.

Reviewing papers and handling review of papers takes much of my time.

From 2 hours to 12 hours, depending on the quality of the paper and journal.

Indeed, the time I spend may be 2-5 hours - but the "elapsed time" from when I receive a request to the time I've finished typically runs 4 weeks.

Usually writing a review to a journal paper takes a few hours, but if it is extremely bad (e.g. lots of typos or plagiarism or the topic is not suitable for the journal) and has no chance of publication, it can take less than an hour. Journals give 4-6 weeks for me to write a review, but I usually try to send mine in a few days, sometimes on the same day if I have time.

It depends on:

• Rank of the journal
• Quality of the paper
• The final decision (Acceptance or Rejection)
• Simplicity and complexity of research
• Closeness to research interests of the reviewer 
• The degree of concentration during review time
• Having experience in reviewing papers

Shorter if prepared well

Reviewing and article for conference/ journal or a project is a volunteer work. Usually it takes a lot of time and if you accept some reviews others will come soon. Anyway some people must do this work and maintain a high level for the conference/journal or project competition. The time for review depends on number of pages, quality, etc.. I start always by checking with an antiplagiarism system. Sometimes I need more time since I have too many articles to review, but when I start I need a couple of hours.

Of course, it depends on the length of the research paper. However, I did that on several occasions & it took 3 days to 1 week " I have other things to do for my job". I believe that this is an immense responsibility & that a hasty review will not correct the science mistakes + the language mistakes completely.

I review both article manuscripts and book-length manuscripts which can run to hundreds of pages.  With Tiia, I can recognize a poorly written manuscript in a split second.  One deserving to be rejected outright requires very little time to outline the reasons, no matter the length.  One deserving to be accepted with corrections takes the most time. For a 20 pp. article I need about a half a week. For a book-length manuscript, one or two months suffice.  And for a decision on tenure and promotion, for which I must read all the articles and manuscripts that a candidate has published, I need three months.

It depend upon many factors

 But a well written paper for a high impact factor journal has taken sometimes more than a week with breaks in between. But In case the paper is not rigorous enough, I have finished the work in 10-20  minutes in some cases.

Dear Nelson,

I agree with you that a paper deserving to be rejected outright requires very little time to outline the reasons, no matter the length, whereas a paper deserving to be accepted with corrections takes the most time.

It depends on the paper itself. However, if the paper includes all the appropriate contents, it takes at least one week to be reviewed well. 

I agree with most of you.

A poorly written manuscript can be recognized very soon.

However, if that is a good paper, it could take from two days to two weeks, depending on the subject, the rank of the journal, and your previous experience as a reviewer, among other things.

Dear Jesus, I agree with you.

Previous experience plays a significant role in the length of time to review a paper.

It is indeed very dependent on several factors, most covered already. Very seldom does it take less than an hour, and this is only in the case where the paper does not fit the journal at all, or - in very rare cases - that the paper is clearly way too bad to salvage. It can take a couple of weeks, full time, if the technicalities are many and deep, and if I need to search in previous papers and/or spend many hours checking proofs of theorems. It also takes time to check the literature for oversights in the referencing. The last one is becoming more and more important, as it appears that researchers themselves do not always take the time to check that their work is brand new, or is related enough for citations to be made. 

Dear Mahfuz

A research is well organized, simple and straightforward. Usually I read the article twice. The first superficially just to understand what it is, and the second well depth, questioning every sentence the author.  it can take from one to two weeks

Dear Mahfuz,

It depends on the quality of the paper. If the paper is poor then it takes more time to justify rejection. In general, I don’t like to postpone reviewing papers. My strategy is to scan the paper first then to dig into it later. So I don’t stay once on reviewing it.

It depends on many factors.

In my opinion,

• Type of the article(conference or journal)
• Length of the paper
• Closeness to my field of expertise 
• Quality of article

are the main factors affecting the review period.

For poorly written articles, the rejection decision is pretty quick. For a well written journal articles, I have spent between 8-20 hours spread on span on 2-3 days to reach a decision.

Review a research is a critical process and hence the time duration depends on various factors associated with it... type / kind / area / time available at disposal to the reviewer / voluminous ....etc.,

It depends on many factors: some related to the paper, others to my time; whatever, it usually takes between one week and 6 weeks.

It significantly depends on how the paper is written. The more coherent it is, the faster I can complete a review. Sometimes the review for a brilliant paper in terms of novelty and results   but with poor language and extremely complex structure takes a few weeks to complete. 

Dear Dr.Mahfuz, Thank you sharing your question with me. To be honest, I used to review a lot of papers till few years ago, but nowadays I don't accept reviewing a manuscript except it is from high impact factor journal. I prefer to write and publish my own results. 

About the time needed to complete a review: It depends on journal, editor, my interests, quality, ....

For manuscript with poor content and bad English: Just few minutes

For a good written manuscript from high impact factor journal: i spend few hours 

Reviewing process depends on individual schollars

 a first reading will give you overall direction in which the paper is written.

continue this process in second or third iteration of reading and simultaneously recording comments will help the author to address these comments systematically

Some times the theme of the paper may be good but the paper language and poor discussion requires the reviewer to comment for improvement in the quality.

usually in an academic environment of institutions it will at least take 10-15 days

As everybody mentioned it depends on many factors, but in case of having all properties, for me it takes two weeks to review it.

The review of a paper in my field is sustained by a accurate analysis of background, methodology, statistics, result relevance and Discussion. Additionally, a personal opinion about the originality and usefulness of the research is rquired. So, often it is necessary to read more other papers about the topic. And, finally, it is necessary to exclude aspects of plagiarysm. Most journals ask us to do all within two weeks. Experience may help.

It depends on all factors mentioned in the answers above, and take one week.

Usually it takes 1 week, but really it depends on research volume, its multifunctionality, and other factors.

As mentioned above it does depend on many factors but it normally takes one to two weeks.

It depends on my familiarity with the matters presented. No or low familiarity = refuse to review, appropriate familiarity = typically 2 days (time needed depends on my own workload). In contrast to some remarks above I do not think that I'm supposed to check for plagiarism or read a bunch of other papers to understand the one under review. After all, it should be self-contained.

In any case, I try to start as soon as possible since authors and editors are waiting and it is good habit to avoid unnecessary delays.

The time, in part, is governed by the "scope of work" assigned to you by the in-house editors.

Thank you for the question Dr. Judeh, the feedbacks are very useful.

Agree with Nageswara!

Most of the previous answers from the respected researchers/reviewers  bound the review process from minutes to couple of weeks, why most journals takes from 1 to 2 years to publish a paper?

Typically it takes more than an hour but less than a day, depending upon the extent of the research being described/proposed.

Dear @Ahmad, that is absolutely true! What a nonsense! Something should be done in order that such policy of Journals be changed!

Dear Colleagues, Had any of us apologized for not reviewing a paper due to presure of work.

I could understand that the question is about the time required by a reviewer to review a research paper. Reviewing time depends upon the familiarity of reviewer with the subject, length and content of paper, presentation quality, compliance of journal's guidelines, etc. A nicely composed and well explained research paper require less time than poorly written and disorganized manuscripts. The issue of plagiarism and correctness of references cited are dealt by the editorial office. I normally review a paper in 3-4 sittings within a time of 15 days, if not much occupied with added official assignments.

To my experience, a research paper takes 2 weeks (it's hereby understood and agreed): anthology- 1-2 days: expertise- strictly 5 days.To review it, firstly, I read and analyze the material,trying to find all the nuances and beauties, then I read as much as possible on the subject matter.

Dear @Mahfuz, The answer to your new question "Had any of us apologized for not reviewing a paper due to pressure of work."  would be YES. I do that due to tide schedule set by the journal. If I find a review inconvenient I press the "Decline to Review link"  inside the invitation letter by the editor. I often do that!

In my experience as reviewer, the time that I spent in reviewing a paper depends on the following:

1- The content of the paper, its simplicity and complexity of the research.

2- The clarity on how the paper was prepared by the author.

3- The knowledge of the author about the content of the paper.

4- The time that I have to carried out the revision of the paper and my experience and knowledge of the type of research included in the paper.

5- The type of recommendation that I am going to make as reviewer.

Normally, I only accept the revision of paper that I feel comfortable with its content, and make at least three revisions of the content of the paper and of my own comments and recommendations. For this reason, normally take me between three and fives days to present my recommendations to the editor.

During the revision of the paper I make clear recommendations about the content of paragraphs that I think are incorrect from my point of view explaining why I make these comments, indicating which paragraphs are not drafted in a clear manner and how to redrafted, any possible grammatical or typewriting error, important points related with the content of the paper that need to be included, among others.

I never said in my recommendations to the author that a paragraph is wrong, or the idea included in the paragraph is incorrect without indicating why I am making such comments, or that the paper needs to be revised without indicating where and why, or the the paper needs to be revised from the grammatical point of view without indications where these revision should be made, among others. 

I like to have two weeks to review an article or chapter, a month for a book or thesis, although if pressed I can produce a review in less time. (The experience of having to grade a pile of student reports over a weekend has taught me to be fast when necessary.) Of course, I am not talking about full-time dedication to the review, but a few hours here and there, perhaps an entire weekend. The time in between sessions is important, because it permits me to reflect on the text and on my comments. In practice this can vary a lot.

As for methodology, I like to print out the text and make notes on it by hand as I read it, then write the review based on these notes, consulting other sources during both parts of the process.

If research means scientific paper, 1-2 weeks, several sessions in this time. If research means student's work, from 2 hours to 2 days, if this is PhD - 2-3 weeks, with intervals inside this time.

Estimada Concha:

Just out of curiosity: ¿How do you say that in Spanish?

I'm sorry, Concha, for not being more precise. The Spanish term I was wondering about is the one you translated as "trapping my fingers." The "Estimada" part was just a polite formula to introduce my question, while indulging in a bit of code-switching, as bilingual people tend to do when they know they will be understood.

Querida Concha:

I do find "estimado" listed in the online version of the Diccionario de la lengua española, on the web site of the Real Academia Española, as the participle of the verb "estimar", on the "Conjugar" page for this verb. The RAE says this verb is derived from the Latin word aestimāre, meaning "Hacer aprecio y estimación de alguien o de algo" (to appreciate and esteem someone or something), among other uses.

In my correspondence, for the sake of variety, I like to alternate between "estimado/estimada," "apreciado/apreciada," and "querido/querida," sometimes adding the adverb "muy" to enhance the positive message.

I respect your views and preferences on this matter. Language use is a very personal thing and reflects our personalities, worldviews, and identities in many ways.

And of course the polite "adornments" in correspondence in Spanish tend to be more traditional and flowery than in English, at least here in Mexico.

Con un saludo muy afectuoso,

David

Dear Concha:

No need to apologize, this is how verbal communication works.

Thanks for the phrase "pillarse los dedos"; it is new for me (at least as far as my poor memory goes). Thanks too for your views on the protocol of correspondence.

In Nahuatl the idea of querido/estimado/venerado is often expressed with the morpheme -tzin attached to the end of a person's name.

I guess we should get back to the question at hand, that is, how long we take to review a piece of academic literature.

Más saludos,

David

Dear @David And @Concha 

it is amazing how scientists are curious and eager to learn things, they stick their nose where ever they may learn something, thanks to you both   

@ Concha, I agree with you. I try to check all references and go over the paper at least twice. But I should rather not get deeper into it as I have just received a paper to review.

Dear @Ahmad and @Ljubomir , Your objection of "why most journals takes from 1 to 2 years to publish a paper?" deserve more attention. I think this question can be a very good one to be asked in RG as a separate topic. I will ask it next week!

Many people have difficulty on "how we go about choosing a journal in our field to submit a paper we have written?" So we must define and explain to others "What are the criteria that we consider?" We must also answer "Does it make a difference if we submit to a journal that is published more frequently?"  We have made quite a few bad decisions over the years in selecting bad journals to publish our papers. Sharing our experiences and lessons we have learned can be valuable to new comers, I believe.  So please be patient and be prepared for my next question!

Dear Mahfuz

Yes, I have apologized for not being able to review a paper for the reason that there is too much in my plate, and I can not chew any more. I think, academics are normal humans and have limited capacity to work. When I have deadlines to meet, and I am scrambling to do that, then there is no choice but to say NO.

Also, there are offers from very low grade or even fake journals, where people just want to use your name for the sake of surving in the journal pulishing world, and their efforts are not otherwise appreciable, then also I had to pretend that I was too much involved and could not do that.

Dear Debi, I also apologized to review papers for some journals because of the tight schedule I had at that time.

Agreed with Nageswara.   it may take anywhere from one to three weeks depending upon the content of the paper in terms of its subject matter.

A research paper may take about two to three days for reviewing, of course, as rightly pointed out, it may depend on the type and subject of the paper!

As I said some time ago, I do not feel responsible to check for plagiarism and references. In my view, this is part of the editorial team's job who usually got paid for easy-to-manage quality of their publication. 

What are views here?

I believe that if the reviewer noticed any plagiarism in the paper, it is her/his responsibility to report it to the editor.

Plagiarism is so rampant these days, one of the first things I do with any paper is check for it. If I find any textual plagiarism, that is, identical strings of words, I can reject the paper immediately and save myself the work of doing an in-depth review. (A few times I have found plagiarized paragraphs in texts submitted for publication.)

I agree with you David. We can reject the paper immediately and save myself the work of doing an in-depth review.

It seem to me that the main responsibility of a reviewer is to review a manuscript from the scientific point of view and to recommend to the editor if the paper can be published or not explaining the reason why you are suggesting one thing or another. Perhaps I am wrong, but it look to me that it is not easy for a reviewer to confirm that a manuscript or relevant part of it is really a copy of another paper already published, unless the reviewer are very familiar with the text that have been copy. To do that the reviewer needs time not only to review the content of the manuscript from the scientific point of view, but to check whether the manuscript is a copy of another paper o papers already published. Of course, if during the revision of a manuscript the reviewer detect that the paper is identical to another manuscript known to him/her, then it is the responsibility of the reviewer to draw the attention to the editor on this delicate issue.

In my opinion, it is the responsibility of the editor to look into this matter and take the necessary action in consultation with the author.  

I agree with Jorge here, and, of course, if I see plagiarism, I report it. What I asked above was whether reviewers are responsible for actively check (e..g. by using software) for plagiarism. Still, I think, it's the responsibility of the editors. 

As far as I remember, the guidelines for reviewers I've got didn't mention search for plagiarism. Counter-examples welcome.

I agree with Jorge and Michael that if the reviewer sees plagiarism, s/he would report it. But the question here is, if the plagiarism is been detected after publication of the paper; does the reviewer held any responsibility for not detecting the plqagiarism before publication? 

@ Michael, I agree with your point that plagiarism is the responsibility of those people who are benefited in some way in the process!

Mahfuz, in my opinion if after the publication of a manuscript a plagiarism is detected, the editor is the only person responsible for this situation.

I agree with you Jorge. The editor should take the responsibility, since s/he has the software or the techniques to check for plagiarism.

For me a review depends on the paper length , purpose,  quality and content. A good paper is easier to read and to estimate than a poorly written one.  First I check the theoretical background of the paper , previous works and the current status of the research subject. Then i check the presented results and their exactness and the future perspectives of the work for originality purpose. I take time with the references citations to catch any plagiarism. So it could takes me 1 week to 2 weeks if i'm not busy.

I agree with your standpoint Fairouz. A review depends on the paper length , purpose, quality and content. 

normally Editors check for plagiarism before sending to reviewers, so reviewers do not have to check for plagiarism, but they might do it for curiosity.

I personally do it as a first step, it is very easy step, let the software do the job for you, if the paper passes the test I go for the next step and complete the review process, if not there is no need to do any further processing, just reject the paper for a "good reason".

Dear Ahmad, Reviewers do not have to check for plagiarism, but they might do it for curiosity. That is the right statement.

I agree with Jorge. It is responsibility of the editor to look into this matter and take the necessary action in consultation with the author. However, if the reviewer is well with the aspect or he/she could detect any part of the manuscript plagiarized, he may report so. Rightly said by Jaya, responsibility of finding out plagiarism is the responsibility of those people who are benefited in some way in the process!

Who will take the legal responsibilty in case of plagiarism? I believe the author alone,  will take that responsibility.

In my opinion, the author of the manuscript should be responsible person in case of plagiarism.

Yes, the author is legally responsible in case of plagiarism . However, the publishers also need to be careful in this matter.

Dear all, I agree with your assignment of "responsibility". But the referee system should open the eyes and brain, to find out the most cases of plagiarism.

I agree dear @Hanno, everyone in referee system has obligation to search for plagiarism! Four eyes opened!

I agree with Ljubomir. Any kind of discrepancy/malpractice if detected by any body in the entire process of publication is always welcome in the interest of research. 

In case of legal responsibility, does the publisher or editor held any responsibility in case of plagiarism.

Dear Mahfuz,

I´ve no real idea. But if you could prove, that the editor or publisher didn´t take enough care when choosing the referees or just avoided to use a correct referee system, they will be responsible with all possible consequences.

I agree with you dear Hanno that the publisher should take enough care when choosing reviewers or using the right reviewing system.

Certainly, Prof Kamal. That is happening. If the authors work for university, then the university will investigate and take action, but if the authors are independent who will take the action against them?

Dear Mahfuz,

here we are again. The scientific society must be the judge.

I completely agree with you dear Hanno.

Dear Concha, I agree with you. The law in Spain which you have mentioned can be enacted in other countries. 

I just received a rejection letter and I am very happy because they rejected and replied fast, it took them only 2 weeks to take the decision. now I can improve my work and publish elsewhere. I sent them a thanks letter, this is the first time to do this, I normally send a hard letter.

but the problem is the comments are not so convincing:

"This article is rather short. It is more suitable for a journal specialized in publishing short papers, e.g. PR Letters. Pattern Recognition prefers to consider longer papers with more substance."

where my article is 7 pages long, about 3100 words, with math proofs and experimental results, the number of experiments are 4x19 repeated 10 times. and the results are significant compared to older methods!

what should I do to publish a paper with them invent a rocket!!!!

Dear Dr.Ahmad, You have done the right thing by sending them a "soft" thanks letter. I do not think that we are "the scolars participating in RG" the only ones who follow the RG topics. Surely, some profit-making businesses under the name of "scientific journals" have followed our discussions about dishonesty & fraud. So, their reaction will be to show that they (reject) even though papers ,such as yours, are worth publishing in better journals than theirs.

Dear Ahmad, I believe that the methodology and the true results are more important than the number of pages.

Dear All. As I promised here to ask a question on the subject raised by @Ahmad and @Ljubomir few days ago" why most journals takes from 1 to 2 years to publish a paper?"  I want to inform you that I did ask it today.  Interested readers can follow and help me manage that discussion by following the below link:

https://www.researchgate.net/post/What_is_your_journal_selection_strategy_in_publishing_papers_that_you_have_written?