Q & A
Q: What about the situation where a revision required either removing a significant portion, or adding a significant portion, that either eliminates an author's contribution, or the new addition included the contribution of a new author? is authorship change ok in that case?
A: I suppose this is OK. It would have to make sense and the authors would have to make it clear (probably in the cover letter) that one of the two cases is happening and why.
Q: What happens if we have sufficient reviewers and the remaining reviewers are not responding. No automated email to notify them that their reviews are no longer required.
A: The system is not currently configured to generate and send such an email. The system is configured to BCC only the reviewers who have completed their reviews. In the rare circumstance that the non-responsive reviewer tried to submit a review after you have made a decision, the reviewer would be presented with a message that indicates his/her review is no longer needed.
Q. How many times can the 'major revision' decision be made? If the first decision is 'major revision' and reviewers are not satisfied with the first revision, we may have to make another 'major revision' decision if the paper has substantial amount of material. Is there a limit for the number of times that the 'major revision' decision should be made?
A: Let us bound the question first. For a manuscript that has passed the initial screening stage, the lower bound for the number of times it must be reviewed is one. IEEE policy defines no upper bound. The practical upper bound defined by your time as an associate editor or the number of times reviewers are willing to review it. Clearly, we do not want to “explore” the upper bound. I think the ideal case is as follows:
Manuscript sent to reviewers.
Decision = major revision (revise and re-review).
Revised manuscript sent to reviewers.
Decision = Accept/Reject
Sometimes the revision is close but requires some minor changes before it can be accepted. In this case the steps are:
Manuscript sent to reviewers.
Decision = major revision (revise and re-review).
Revised manuscript sent to reviewers.
Decision = minor revision
Revised manuscript evaluated by AE
Decision = Accept/Reject
Note that in these two scenarios, the reviewers see the manuscript two times. I call these scenarios ideal for two reasons: (1) we do not exhaust the reviewers; and (2) the final decision is timely (assuming the reviewers are also timely).
The question is what happens if the second review (by the reviewers) finds major flaws. Some suggestions:
- If the issue is the authors have not fully addressed the concerns of the reviewers, then it is probably best to reject the revision. The justification is that progress is not sufficient, and reviewer and editor resources are limited.
- If the reviewers see publishable contributions, then you may elect to issue a second Major Revision decision (which means the reviewers will review it a third time). But, to show respect for our reviewers’ time, you should probably make it clear that this is the authors’ last chance to get it right.
I have seen manuscripts with seven R’s be rejected. This means that the reviewers have evaluated the manuscript eight times! I think it would have been better to reject the manuscript earlier. It saves your time, it saves the reviewers’ time, and it saves the authors’ time.
Three final thoughts. I worry that we “wear out” our reviewers by asking them to review revision after revision of a manuscript that they are not enthusiastic about. I think many of you are too nice: you are trying to help authors generate a publishable manuscript. This is a worthy goal, but the review process is not the place for helping authors develop a publishable paper. Finally, when an author asks for more time to complete a major revision, this could be a sign that the better decision would have been a rejection.
Q: I have a paper that's been praised for its technical content, but reviewers have pointed to very poor use of English language (and I agree). It is too much to ask of reviewers, to expect them to do the language editing. I have done some of it myself, but it's taking a long time and, quite frankly, I'm not qualified to do that. What is the IEEE policy on prescribing professional services? How much "cleaning up" of the writing happens after we are involved?
A: Accepted manuscripts are edited by IEEE Publishing’s editorial staff. AESS pays for the full editing service. (There is a less-expensive option called moderate editing.) With the full editing service, the text in each accepted manuscript is checked and corrected for grammar, punctuation, spelling and style. Consequently, this service is intended for articles where intelligibility is not the issue, but style and grammar.
For manuscripts where intelligibility is an issue but the technical content is reasonably good, it is appropriate for you to request the author to consult with a native speaker of English or use his/her favorite editing service. In the cases where the aforementioned resources are not available, major editorial resources are available via IEEE. The link to this resource is available through the IEEE Author Center. The “navigation path” is the following:
https://ieeeauthorcenter.ieee.org → Journals → Create your IEEE Journal Article → Refining the Use of English in Your Article
Here one finds links to language editing services. The authors have to pay for this service, but there is a discount for IEEE authors.
Q: Maybe also tell us why the training has been deemed necessary?
A: When we (the AESS publications committee) go through the manuscripts, we find many are “stuck” at one stage or another and the Associate Editor is not aware that the manuscript is stuck. This is one of main contributors to our longer than average review times. The hope is that this training will help the associate editors understand the process, the various manuscript stages, and avoid the “getting stuck” problems. In addition, the more familiar an Associate Editor is with IEEE policy, the more confident an Associate Editor can be in making decisions.
Q: What is your view on self-plagiarism?
A: There are a number of issues embedded in the phrase “self-plagiarism”. The first is reuse of an author’s own words and work and the second is citing one’s own work. Let me answer these separately.
- Reuse of an author’s own work. This can be a tricky issue. IEEE’ policy is
IEEE’s technical publications shall include original material which appears only once in the archival literature.
One way to think about this is to ask, “is the reused material original material?” From this point of view, it is probably OK when an author reuses sentences, equations, maybe even large portions of paragraphs, if the reuse is in the introductory material, problem statement, and problem formulation. We should see less reuse as the new material is introduced and developed. However, the reuse outlined above can be abused. I think this one is a judgment call.
I usually try to catch cases where that I think cross the line and either open a consultation session or attach the Cross Check report to the manuscript record. In the second case, I usually send an email to the Technical Editor to make him/her aware of the Cross Check report. Again, I ask that you review this to see if it crosses your threshold.
Another case of reuse is when an author reuses material from a conference paper version of the work. Within limits, this sort of reuse is anticipated by IEEE’s “evolutionary publication process”. This is defined in PSPB 8.1.7.F “PUBLICATION OF ARTICLES IN IEEE PERIODICALS BASED UPON CONFERENCE ARTICLES”. This section was highlighted in the training session. I repeat it here for convenience:
IEEE Policies, Section 6.4, specifies that IEEE’s technical publications shall include original material, which appears only once in the archival literature, and Subsection 1.3.A of this Manual presents a definition of an archival publication. Nevertheless, it is common in technical publishing for material to be presented at various stages of its evolution, such as publishing early ideas in a workshop, more developed work in a conference, and fully developed contributions as journal or transactions articles. IEEE recognizes the importance of this evolutionary publication process as a significant means of scientific communication and fully supports this publishing paradigm, but IEEE requires that this evolutionary process be fully referenced. Accordingly, the publication in an IEEE periodical of original technical material contained in a conference article or articles is permitted provided that:
- All the articles have undergone the standard peer review for the specific periodical in question.
- With the exception of that specified in Subsection 8.1.7.A, all the articles published in the periodical contain substantial additional technical material with respect to the conference article or articles of which they represent an evolution.
- Each article in the periodical satisfies the requirements of Subsection 8.2.4.G.2.
- If the organizational unit or units sponsoring the periodical desire to set a quantitative threshold for the additional technical material mentioned in 8.1.7.F.2, the requirements shall be clearly specified in the instructions for the authors of the periodical.
It seems to me that some “reuse” is anticipated, but that the reused material does not constitute a significant portion of the content. As suggested by item 4, many journals establish numeric thresholds for additional technical material (e.g., reuse of conference paper content cannot exceed 30% of the journal version). AESS has not established a numeric threshold. Instead, AESS relies on the judgment of the editors.
- The responsibility to cite one’s own work. The responsibility to cite one’s one work is the same responsibility to cite someone else’s work. In other words, if one is using previously-published material, the previously-published material needs to be cited, independent of who authored it. So, I tend to view the responsibility to cite one’s own work as part of the responsibility to cite other’s work: the same standard applies. With one exception: it is one thing to fail to cite another’s work if that other’s work is unknown to the author (this is often discovered in the review process). It is another thing entirely to not cite one’s own work. The author cannot claim it was unknown.
The two most common scenarios where authors fail to cite their own work are when their own work is a conference paper or when their own work appears in a non-IEEE venue. I am somewhat sympathetic to an author who fails to cite his/her own conference paper. When I started my career, it was common practice to not cite one’s own conference papers. When I encounter this scenario during the quality control process, I usually send the manuscript back to the authors with instructions to cite the conference paper and to describe the substantial additional technical material contained in the journal submission.
Authors who fail to cite their non-IEEE journal papers is a little more problematic. Sometimes, authors do not realize this is a requirement. (I do not know why they believe this.) In other situations, I worry that authors are trying to hide closely related work in an effort to artificially inflate the perceived novelty of the current submission. AESS has no policy on how to distinguish between these two cases. Consequently, we adopt a case-by-case approach. If you encounter this situation in the review process, please open a consultation session with your Technical Editor to discuss the particulars of the situation.
Q: I've been doing this for a few years now and the number of invitations to review required in order to get that minimum of two is growing over time - is there any action being taken to deal with people being less and less inclined to review?
A: I am not sure what our “lever” would be in dealing with people who do not accept invitations to review submissions. Suggestions are welcome. On the flip side of the issue, the AESS publications committee currently is exploring the possibility of some sort of reviewer incentives. The discussion is in the early stages.
Q: Chinese reviewers often accept with little or no comments.
A: No comment.
Q: It's also super hard to get timely review...
A: This is a challenge for all journals. But it seems to be a huge challenge for TAES. Historically, we have allowed more time for reviews than other journals because many of our reviewers worked in industry. The feeling was industry people needed more time to complete a review than academics. Today, I am not sure the situation is the same: everyone is overworked. It is challenge for all to find the time to perform a thorough review.
From my perspective, there are two things that help this:
- Invite the “right people” to be reviewers. Those who are very familiar with a topic and have published on the topic recently are more inclined to accept review invitations and to perform insightful reviews. This may seem obvious, but it is harder than it might appear. It can be difficult to do this in an area you are not familiar with. I offered some ideas in the training slides for finding the right people. When I was an Associate Editor and I had to find reviews for submissions too far outside my expertise area, it was very very hard. So, it is important for Technical Editors to be aware of the specialties of Associate Editors assigned to the area. This will help.
- Submission quality. Reviewers are more likely to accept an invitation to review a high-quality submission than a low-quality submission. I would like to see the overall quality of submissions increase. One way to do this for each of you to invite the best people in your areas to submit to TAES. This is an easy and relatively pain-free way to start the improvement process.
A corollary to this issue is the large number of low-quality submissions. Not all of these submissions have to be reviewed. Given the scope of TAES, it is sometimes hard for me to tell if a submission is good enough to be reviewed. If a low-quality submission is assigned to you, you have permission to open a consultation session with your Technical Editor and explain your views of the submission. The Technical Editor will likely expand the consultation session to include another Associate Editor or two to obtain a concurring opinion. In this case, the manuscript may be rejected without review because it does not contain enough “technical substance”. Please review PSPB Section 8.2.2.A.3 “Prescreening of Articles by Editors-in-Chief”. (This section was highlighted in the training session.)
Q: How long do I have to wait for the 3rd review?
A: First, it might help to understand how our manuscript management system responds to late reviewers. The system is configured to send three automatically generated emails to reviews:
Review Almost Due – sent 15 days before the review due date
Review Past Due #1 – sent 3 days after the review due date
Review Past Due #2 – sent 10 days after Review Past Due #1
Review Past Due #3 – sent 10 days after Review Past Due #2
I have the sense that many people ignore machine generated emails, so I am not convinced the last two are effective. If you have not received a review from the 3rd reviewer after the Review Past Due #1 email has been sent, then I suggest you contact the reviewer personally. Perhaps you know this person. Send an email from your email account to the reviewer asking if he/she will be able to submit a review. If you do not receive a response, then it is probably time to move on either make a decision with the two reviews you have received or secure a new reviewer (but this needs to be done fast with a shortened timeline).
Q: What about the 4th?
A: Same answer as the 3rd.
Q: May authors reuse paragraphs or subsections from other papers?
A: The reuse of paragraphs and subsections from other papers (either by the author or by others) is a complicated issue. The best practice is for authors to summarize and cite the previous work. But we must be careful. When an author is listing equations describing, say, the measurements used to formulate a problem and there exists a series of canonical equations describing the measurements and how the measurements are initially processed, it may appear that the author is reusing a paragraph. But there are so few ways to say something like this, that we should probably view this as acceptable.
In summary, the answer is generally “no”, but there are reasonable cases where reuse is to be expected.
Q: Do we as AE see the final version they submit after acceptance?
But we don't see the final version submitted for publication
A: The current workflow does not require the Associate Editor to approve the final files. If you perceive we have a problem with authors modifying accepted manuscripts, we can discuss changing the workflow. So, the question is: do you perceive we have a problem?
Q: Can we only use reviewers' comments? What if the AE notes an issue not identified by the reviewers? (I assume we cannot use ourselves as reviewers due to independence requirements.)
A: First, allow me to quote PSPB Section 8.2.2.A.3
It shall be clearly stated by the Editor-in-Chief when the Editor-in- Chief submits the article to the referee that the recipient is a formal reviewer and that his or her comments and opinions, with those of other formal reviewers, will form the basis upon which the Editor-in-Chief will decide whether or not to publish the article, and with what changes.
The editor examines the comments of the referees and exercises his or her own best judgment, in the light of the referees’ recommendations, on whether or not to publish. Reviewers’ comments and marked articles are normally returned to the author in any case. It is essential that the editor assure that the anonymity of the referees is protected during this process.
Everything within this review process must be done openly, except that the referees are protected from personal interactions with the authors by withholding their names. The editor in any event, should be guided in technical matters by the reviewers’ comments. The editor must not arbitrarily withhold the referees’ comments from the corresponding author, or vice-versa, unless the editor deems them clearly to be irrelevant, incorrect, or otherwise inappropriate. In particular, editors should not arbitrarily ignore referees’ suggestions for modifications of the article without sufficient technical cause to do so. If an article is returned for revision, it is important to make clear to the corresponding author whether on the one hand the article will be accepted if the indicated changes are made or, on the other hand, the article will be resubmitted to the referee for further review.
In writing a review, it is unacceptable for a referee to require additional citations for the sole purpose of influencing bibliometric measures of either an individual or a periodical. In such cases, referee comments shall be transmitted to the authors, but editorial judgment shall be exercised in explaining requirements to the authors regarding addition of citations. The editor’s decision is always based on all the reviews received, but mixed reviews present the need for the exercise of editorial judgment. Responsibility for the final decision of acceptance or rejection lies with the Editor-in-Chief.
In summary, the decision is based on the reviewers’ comments. “Based on” means the associate editor uses “editorial judgement” in interpreting the reviewers’ comments. The reviewers are not voting. They are providing information to you so you can make a decision.
The question here asks what to do if an Associate Editor identifies an issue not identified by the reviewers. This issue should absolutely be a part of the decision process. It is appropriate, when you are summarizing your decision, to explain that your own reading of the manuscript has identified additional issues and that these issues need to be addressed in a revision (or, if rejecting, are part of the reason for the rejection). In this case you are not anonymous, but your opinions are important. Your appointment as an editor is because your opinion matters!
Q: Experimental results of a previously simulation-only paper is generally enough?
A: It depends. It might be in some cases. It might not be enough in others. It really does depend on the area and the particular circumstances of the work. Here, we must exercise editorial judgement.
Q: A common numerical metric for "substantial" is 30-35% additional significant material (theory, numerical analysis, experimental study, etc.). Is this acceptable for TAES?
A: The 30 – 35% additional material rule is in line with what is used by many of the journals that set a numeric threshold. You are free to use that as an initial starting point in evaluating a manuscript. But, because this is not a formal policy, it cannot be the only reason a submission is rejected.
Allow me to construct a hypothetical circumstance where this rule might not serve TAES well. Suppose an author publishes a conference paper with some initial experiments that suggest a new idea has potential merit. Later, the author submits a “journal version” to TAES that includes two theorems and their proofs, some analysis in real environments, and the experiments from the conference paper whose results are reexamined in light of the analysis. It is common for theorems, proofs, and analyses to take up much less space than the description of an experiment with its accompanying block diagrams, plots, procedures, etc. I think we would all say the TAES submission has substantial additional material, but the new material might only account for 50% of the page area of the TAES submission.
Given the breadth of TAES and the different cultures that prevail in the different areas, I prefer sound editorial judgment to a numeric threshold.