Information for Reviewers
Reviewers play an extremely important role in the operations of the IEEE Transactions on Information Theory. Reviewers volunteer their time and expertise to assess the suitability of a submission for publication, serving as trusted advisors to the Associate Editor handling the paper. Reviewers are selected for their expertise in the field and their ability to exercise fair judgement.
The suitability of a paper for publication is often assessed against three criteria (two of them objective and one subjective), which, loosely speaking, can be phrased as follows: is the paper new?; correct?; interesting?1
Novelty and correctness are obvious attributes that any newly published paper must possess. Novelty is a measure of the results and ideas of the paper: a paper can be considered novel if it contains a new result (even if the result flows from the application of well-known techniques) or it contains a new idea (even if the idea merely reveals a simpler way to understand an old result). When a reviewer discovers that a paper is not novel, they should provide the Associate Editor with clear and specific references to the prior literature where the results or ideas of the paper may be found. When a reviewer discovers an error in a paper, they should give a clear and specific reason (a counterexample, perhaps) that elucidates the issue.
The third question—is the paper interesting?—requires judgement and familiarity with the community served by the Transactions. The question might also be phrased as "does it matter?" or "will it make a difference?" Of course this is a subjective criterion (after all, what is interesting to one person may not be interesting to another); what is intended by this question is whether or not the paper, if published, will serve the needs of the readership of the Transactions. The information for authors states that
"novelty alone does not assure publication; the significance of a paper and its usefulness to this Transactions' readership will also be assessed."
For example, a paper with new and correct results that are, however, purely of mathematical interest, without any engineering significance, may be deemed outside the scope of this Transactions, and the authors re-directed to a suitable Mathematics journal.
Reviewers have specific responsibilities towards the authors. Reviewers must:
- provide unbiased feedback in a timely manner, giving comments supported by a suitable rationale;
- avoid personal comments or criticism;
- maintain confidentiality of the review process and of the content of the paper.
Reviewers also have specific responsibilities towards the Associate Editor. Reviewers must:
- alert the AE about possible conflicts of interest;
- notify the AE promptly if circumstances arise that make it impossible to complete the review in a timely manner;
- notify the AE promptly with any ethical concerns;
- refrain from contacting the authors directly.
In turn, the Associate Editor has the responsibility to preserve the anonymity of the reviewer, and to communicate the reviewers' comments to the authors without editing them.
Reviewers should keep in mind that they serve in an advisory capacity only; they are not themselves accepting or rejecting a paper, but, rather, are providing a recommendation for the Associate Editor. Ideally the recommendation should be crystal clear: should the paper be accepted or should it not? The best reviews are those which the authors ultimately view as being helpful (even when the review recommends rejection). Indeed, as evidenced by the many acknowledgments in published papers to comments made by anonymous reviewers, the IEEE Transactions on Information Theory has a proud tradition of excellent and helpful reviews.
Finally, as Halmos suggests [1, p. 121], "Papers to be refereed don't improve by sitting for a few months at the bottom of the pile on your desk, and you don't save time and energy by postponing the day when you refuse the job." Thus, when deciding whether or not to accept a review request, a potential reviewer should take on the review only if it is something they have time for immediately. As Halmos writes, "Zero or infinity. Do the job now or do it never, and, in either case, say which, now."
The reviewer's task is often a difficult one. The following links provide some helpful advice.
- The Council of Science Editors, on Reviewer Roles and Responsibilities
- Niklas Elmquist, on Mistakes Reviewers Make
- Jennifer Raff, on How to become good at peer review: A guide for young scientists
- Alan Meier, on How to Review a Technical Paper
Policy on reviewing
The Board of Governors (BoG) of the IEEE Information Theory Society passed the following resolution at its July 1, 2012, meeting:
"In view of its concerns about excessive reviewing delays in the IT Transactions, the BoG authorizes the EiC in his sole judgment to delay publication of papers by authors who are derelict in their reviewing duties."
Reviewers may be considered to be derelict if they habitually decline or fail to respond to review requests, or accept but then drop out; or if they habitually submit perfunctory or excessively delayed reviews.
In applying this rule, the EiC will take into account the authors' overall record of service to the Transactions. This policy is effective immediately. The EiC will inform authors prior to imposing penalties.
1P. R. Halmos [1, p. 119] and R. P. Boas, Jr. [2, p. 10], attribute these criteria to mathematician G. H. Hardy, who, as editor of the Journal of the London Mathematical Society used to tell referees to ask three questions: Is it new? Is it true? Is it interesting? (back)
 P. R. Halmos, I Want to be a Mathematician: An Automathography, Springer-Verlag, 1985.
 R. P. Boas, Jr., Lion Hunting and Other Mathematical Pursuits, The Mathematical Association of America, 1995.