Hello Atiqa, I would only review research that are within my expertise, to be fair to the author/s.  I have refused to review several papers.  There was an editor who asked me to review a research on the Koran; this I rejected.  I was asked to review another paper, but the abstract was not clearly written, with unnecessary stock phrases, and I rejected that as well.  If the abstract is so poorly written, and I have to GUESS WHAT THE AUTHOR REALLY WANTS TO CONVEY, I should reject reviewing the paper. Another reviewer may do a good job of it.

I accepted the offers to review papers that were related to my field.  I feel that I am able to be fair to the author/s.  When I review, I would recommend acceptance for papers that the authors clearly express the contribution in bringing the knowledge a bit forward in the field.  I would recommend papers where the findings are honest, as far as I can gauge, and the paper is interesting to me and likely to interest other readers.  Have a good day.

Hello Atiqa, I would only review research that are within my expertise, to be fair to the author/s.  I have refused to review several papers.  There was an editor who asked me to review a research on the Koran; this I rejected.  I was asked to review another paper, but the abstract was not clearly written, with unnecessary stock phrases, and I rejected that as well.  If the abstract is so poorly written, and I have to GUESS WHAT THE AUTHOR REALLY WANTS TO CONVEY, I should reject reviewing the paper. Another reviewer may do a good job of it.

I accepted the offers to review papers that were related to my field.  I feel that I am able to be fair to the author/s.  When I review, I would recommend acceptance for papers that the authors clearly express the contribution in bringing the knowledge a bit forward in the field.  I would recommend papers where the findings are honest, as far as I can gauge, and the paper is interesting to me and likely to interest other readers.  Have a good day.

I personally believe that the most important thing to consider before accepting any type of academic work for review is to make sure it falls within your working research domain. It includes the reviewer's prior research published in the domain or prospective research interests. This will help to ensure that a reviewer does not end up submitting a poor review. It is advisable to check whether the reviewer has sufficient knowledge regarding the extant work in the research area. This helps to quickly assess the quality of the paper. The reviewer should also spend enough time (wherever possible) to go through the full papers cited in the publication submitted for review. This will help in a balanced review by the reviewer and also helps to understand the appropriateness of the research methodology followed. In terms of the review, most journals have a structured format for a review. It is suggested that the reviews should be submitted in such a way that the authors do not feel offended. the authors should believe that the review has helped them to improve their paper and broadened their understanding about the research problem. The reviewer should ideally agree to review a paper only if he/she can  commit a minimum amount of time.

Dear Atiqa,

You may also see the following link:

http://twp.duke.edu/uploads/media_items/scientificarticlereview.original.pdf

In my opinion the language of the reviewer should be polite except in cases of plagiarism.

Best wishes,

Subir.

this depends upon criteria of the journal giving you a chance to review manuscript.  Every paper has standard structure which the reviewer has to check. Good literature review and citations should be part of introduction. All description of research problem must be cover in paper. Author way of adopting the way to present methodology, good quality graphics,standard reference format and concise  conclusion or results.

Mostly i write summery , critical comments , improvements and language flow of the paper after review

But all of these important thing is you should have strong grip on topic for which you are giving review.

Thank you Miranda Yeoh, for sharing your experience. It was really good to go for those regarding one's own field. It will be in a favour line to both authors and reviewers. 

I ll consider abstract a main point before reviewing it. 

Thankyou to all the others who share their opinion. 

Thanks alot Mr. Harshvardah for iThenticate pdf :)

How to review a publication?

Suggesting the following approach:

How to conduct a review (from ELSEVIER)

• Before you accept or decline an invitation to review, consider the following questions:
• Does the article match your area of expertise? Only accept if you feel you can provide a high quality review.
• Do you have a potential conflict of interest? Disclose this to the editor when you respond.
• Do you have time? Reviewing can be a lot of work – before you commit, make sure you can meet the deadline.
• Finally: Educate yourself on the peer review process through the free Elsevier Publishing Campus

Checklist for writing the review report

• Summarize the article in a short paragraph. This shows the editor you have read and understood the research.
• Give your main impressions of the article, including whether it is novel and interesting, whether it has a sufficient impact and adds to the knowledge base.
• Point out any journal-specific points – does it adhere to the journal’s standards?
• If you suspect plagiarism, fraud or have other ethical concerns, raise your suspicions with the editor, providing as much detail as possible. Visit Elsevier’s Ethics site or the COPE Guidelines for more information.
• Give specific comments and suggestions, including about layout and format, Title, Abstract, Introduction, Graphical Abstracts and/or Highlights, Method, statistical errors, Results, Conclusion/Discussion, language and References.

https://www.elsevier.com/reviewers/how-to-conduct-a-review

Its really helpful guidelines.

Thanks Atiqa. You may also like the following presentation from professor Flemming. Here is what he wrote about ethical principles to observe (page 2):

Integrity: Treat the contents of a manuscript confidentially! Refrain from unfavourable reviewing just because you are engaged in similar work not yet ready for publication (i.e. avoid partiality and reviewer bias). If there is a clash of interests, decline the review!

Fairness: Assess a manuscript by the merits of its content. Do not impose your personal opinion! Constructiveness: Always take a positive attitude! Recommend changes that will improve the Ms. This may also involve language corrections.

Expertise: Only accept a request to review if you feel comfortable with the subject matter. Don’t pretend to know everything about anything!

Article How to do a good Review -The Editor's view

These factors do matter in actual... No biasing should be done at all.

Thank you Behrouz 

Good information by Behrouz Ahmadi-Nedushan 

Dear Atiqua,

You wrote "Specially when the topics are little known to your own knowledge".

Normally Editors consider only those as reviewers who have expertise in the relevant field.

Personally I can tell you that in such cases I politely return the MS informing that it would not be possible for me to review the MS and if experts are known to me of that field, I suggest their names for consideration..