Rajesh - there is no such thing as 'self-plagiarism'. You can't plagiarise your own work. I think what you are referring to is 'self-citation'? There is no strict rule for self-citation - but reviewers and editors are likely to notice if an author/s over-relies and stresses their own work at the expense of other authors and viewpoints. There is a potential for it to be considered a 'bias'.
Thank you very much for your reply, but my concern is about the self-plagiarism not about self-citation. As my methodology is same as my previous published work, so a major portion is same as my previous paper. When I checked its plagiarism, it shows a high amount of plagiarism from my previous paper. Should I re-write my paper again?
Plagiarism checker would ordinarily detect any section of writing that has been published previously. When that piece of work is done by you and it's being still being used in another material written by you, then you have done self-plagiarism. However, you can rewrite that section in a different style and reference it; that becomes self-citation. If you use the same method in several of your articles, then you must rewrite it differently and do self-citation. Journals have different levels of perception and judgement when it comes to accepting materials with the same methodology, and self-plagiarism detection limits. You must double check with the guidelines to authors for clarity.
Rajesh - what you refer to I would call 'salami-slicing'. It is owned upon by editors in that, when you sign journal copyright declarations, you are not just stating that it is your own work - but that it hasn't been published elsewhere. Often authors think that means they are okay if they are not submitting a whole published paper to another journal. However, the copyright also often covers no 'significant part' has been published elsewhere as well - hence the salami-slicing i.e trying to get multiple publications from the same study reporting similar outcomes.
Just to add a bit to the conversation, I've seen a situation where results are new and methods are new but the setup / lead-in / tools from another recent work are used. In this situation, I would advise rephrasing things (to get around the copyright issues previously discussed) but as long as the point of the work is not just "salami slicing" (thank you Dean), it should not be a problem.
Looking back at the origin of this problem, automated checking of texts was introduced to detect cheating in the form of stealing other people's intellectual property and claiming it as one's own, and has made it possible to reduce salami slicing. So-called self-plagiarism is identified through the same process as copying other people's work and there is no doubt that it can exist at an unacceptable level, so it needs to be curbed. What authors tend not to realise is that as we are all inundated with papers to read to the extent that we cannot keep up with the literature, one really good paper with substantial results has much more value to the reader than five or ten weak ones.
Common sense should always dictate what is and what is not acceptable, and the copyright terms associated with the work published in journals and books explain what is allowed. However they also state that proper acknowledgement of the original source is important, so this should not be forgotten. If you need to repeat a description of a method or some other standard procedure that is difficult to explain, refer to the relevant text in the opening sentence: "This [xxx] has already been described in earlier work, and I paraphrase [/or even reproduce] this here for the ease of the reader." If you cite directly, always use quotation marks.
Generally reasonable reproduction of one's own work only needs proper acknowledgement to the original source rather than permission, however where blocks of 100s of words are required [say for a review where you have played an important part in the history of the field], I would obtain permission, add an acknowledgement and possibly even point out the copied text on submission.
Native English speakers do not have any excuse for what has become known as "self-plagiarism" – we can always rewrite text, however the fact that we are so fortunate should oblige us to be tolerant of authors who struggle to express themselves, asking is this person deliberately copying in an attempt to cheat, or battling a language barrier.
I agree with the submissions above. @Pritam Kumar Panda, that is a better way of reporting. Even if it is your own work, you still need to c to previous work.
Don't worry about the plagiarism percentage as its just a number. It doesn't give much insights on whether there's actually any copied content or not. Let me explain... The plagiarism level usually acceptable is 25 - 30%.
But this limit is because of the natural requirement to quote something, cite a reference, or include a part of results from previous researches and build upon that. This limit is not for directly ripping off lines from a paper or research work or any other source. If your plagiarism is about 25% and there aren't any cases of completely ripped off paragraphs then your paper is just fine. If your plagiarism is about 10% but there are complete paragraphs ripped off from some other source then its not so cool. I hope this helps. All the best!
I have learnt that when you are conducting research for a project and you refer previous work it is is called recycling so you need to check with your supervisor. I am completing an MBA and I referred work from a some publications I have and I was marked down for it so do your due diligence. Make sure you have access to a good tool for checking similarities like turn-it-in. Do not sell your self short.
The acceptable level may not vary wildly with fields of study because plagiarism means the same thing in all the fields. If you have used plagiarism checker, depending on the version, you would have out that any similarity above 35 words in a line is indicated in red. That is to inform you that it is below 30% percent acceptable as Osama Rahil pointed out. Besides, some journal publishers or schools said nothing more than 20%. Find out the acceptable point in your school having in mind that you may wish to publish your work outside your school at the end. So you may work towards 20% to be sure of acceptance in any journal.
Why self plagiarise? In fact that is called recycling in academia. With Universities it is a definite no, no. Any credible Peer Review Journals are very strict about these things.
Besides, some journal publishers or schools said nothing more than 20%. Find out the acceptable point in your school having in mind that you may wish to publish your work outside your school at the end. So you may work towards 20% to be sure of acceptance in any journal