Peer review is a critical part of the publishing process at JM3, as it is for most science journals. Yet for many authors, the editorial review process might seem intimidating, and maybe even a bit mysterious. Since there are many variations on the basic peer-review paradigm, in this editorial I’ll explain in some detail how the process works at JM3.
Peer review is defined as “the critical assessment of manuscripts submitted to journals by experts who are usually not part of the editorial staff.”1 It supports the scientific process by providing authors with constructive criticism of their work, and by filtering out less valuable work and thus providing a “stamp of approval” from editors and peers for published scientific work. The mere prospect of peer review prompts authors to improve both the science and its presentation in a submitted manuscript.
The Peer-Review Process at JM3
JM3 practices an editor-driven external peer review of author-submitted manuscripts. Reviewers (also called referees or assessors) are anonymous, meaning that authors never know the identity of the reviewers. This single-blind approach is not the only style in use at science journals. Some journals practice double-blind reviewing, where the reviewers are not told the names or institutions of the authors (in an attempt to avoid bias). Other journals practice open review, where the names of the reviewers are published along with their reviews when the paper is published. Other journals take a middle road, where reviewers are given the option of signing their reviews before they are sent to the authors. The single-blind process used by JM3 (and described in some detail below) is by far the most common style of peer review in scientific publishing.2
Journals should have a well-documented process for peer review. In the spirit of transparency, here is a step-by-step description of the manuscript review process used by JM3.
1. Authors submit their manuscript online, along with a cover letter and various other information. During this submission the author selects either a currently open special section, or a regular paper or letter category [lithography, microelectromechanical systems (MEMS), micro-optoelectromechanical systems (MOEMS), microfabrication, or metrology] for their manuscript.
2. The manuscript goes through a quality control check by journal staff. If there were problems with the submission, journal staff works with the corresponding author to address them.
3. The manuscript is processed through the CrossCheck plagiarism screening service, comparing the submission to a large database of previously published papers. If there is sufficient content in the submission that is identical to that found in a previously published paper, the authors will be contacted for an explanation and the manuscript may be rejected and sanctions imposed if egregious problems are confirmed.
4. Based on the category selection made during submission, the manuscript goes to either the special section guest editors or the senior editor (SE) associated with the regular paper category. The senior or guest editor performs a first editorial review, reading the title and abstract and skimming through the paper. This editor checks to see if the scope of the paper properly matches the scope of the journal, and if the writing is sufficiently good to allow for an effective review. If not, the SE may decide to decline the manuscript without review.
5. For a regular submission, the senior editor decides on an associate editor (AE) with suitable expertise to handle the submission. The AE isn’t necessarily an expert on the topic, but will have enough familiarity to be able to find reviewers and interpret their reviews. For a special section submission, the guest editors will decide which guest editor will serve the role of the AE for this submission.
6. The AE does the bulk of the editorial work for JM3. They begin by performing a second editorial review of the paper, checking for scope, novelty, significance, and quality. They may skim the paper quickly, or read it in great detail. The AE must decide if the paper has a chance of being accepted for publication and thus is worth sending out for review.
7. If the AE does not decide to decline the manuscript without review, he/she will search for and assign qualified reviewers. At least two reviews are required to accept a manuscript for publication, but some AEs may choose to seek three reviews. Often the reviewers are chosen to have complementary skills (experimental, theoretical, mathematical, etc.) so that the full range of topics of the manuscript can have expert analysis. Authors have the opportunity to supply a list of suggested reviewers at the time of submission, but it is the AE’s decision whether to use anyone from that list. Finding qualified reviewers is the often most difficult and problematic step in the process, and sometimes 10–20 candidates must be asked before two reviewers accept the task.
8. When the reviews have been returned, the AE evaluates the reviews and makes a decision (usually request author revisions or reject). While the reviewers may provide accept/reject advice, the AE makes the final decision based on his/her reading of the manuscript and the substance of the reviews.
9. If the author revises the manuscript, it is sent back to the same AE. The AE looks over the revised manuscript and the author’s point-by-point response to the reviewers’ comments, and either decides to send the manuscript out for re-review or makes an accept/reject decision at this point. Multiple rounds of re-review are possible, depending on the extent of the revisions. Generally the manuscript would be sent back to the same reviewers, but it is possible that new reviewers would be chosen if an original reviewer was unavailable, or if significant added material required a reviewer with an additional area of expertise.
10. Finally, once a manuscript decision has been made the proposed decision is sent to the Editor-in-Chief for approval. The Editor-in-Chief performs a final quality check on the overall editorial process, possibly making suggestions for changes or improvement. At JM3, it is rare that I change in any way the decision made by the AE.
11. If the manuscript is accepted, the authors receive instructions on how to make a final submittal of the manuscript and its figures. No changes to the manuscript should be made following acceptance.
12. The final submitted manuscript goes through copyediting and professional composition steps. These important and often unheralded steps can have a major impact on the level of professionalism of the paper, fixing typos and grammatical errors, improving the exposition and presentation of the paper, and ensuring that the graphics are of sufficient quality.
13. Page proofs are sent to the corresponding author for approval, and possibly to supply missing information. Authors should return these proofs promptly.
14. The finalized paper is published online immediately, and in the print version of the journal at the end of the quarter.
JM3 has a specific process for handling submissions by members of the editorial board (myself included) to ensure an impartial review, treating the editorial board member as any other author, with no access to the internal editorial process for that submission. Additionally, JM3 accepts appeals from authors who disagree with an editorial decision. And I am always available to hear from authors or reviewers who wish to lodge complaints or make suggestions for improving the publication process.
Here are some of the major statistics for JM3 in 2014:
• The just-released JM3 2-year impact factor for 2014 is 1.43 (up from 1.20 in 2013).
• 151 manuscripts were received (102 regular papers, 45 special section papers, and 4 letters).
• For regular submissions (papers and letters),
- 8% were declined without review.
- 23% of manuscripts were rejected after being reviewed.
- Overall acceptance rate was 69%.
- The average time to first decision was 6.9 weeks (median time was 5.7 weeks). Our goal at JM3 is to reach a first decision in under 6 weeks.
- No papers were accepted without revision, 66% of accepted papers were revised by the authors once, 30% were revised twice, and 4% were revised three times.
- For papers that were accepted, the average time to acceptance was 16.6 weeks (which includes the time for author revisions). Each additional revision cycle added about 2 weeks on average to the final acceptance time.
All parties in the peer-review process (authors, editors, and reviewers) must work in an environment of mutual trust and cooperation. Honesty and integrity are of course required in all aspects of the process. Additionally, each participant in the peer-review process has specific responsibilities that must be fulfilled.
• Ensure that the work is original and has not been previously published or submitted for publication elsewhere (for a submission based on an SPIE conference proceedings paper, see the SPIE policy: http://spie.org/x85029.xml#Proceedings). Cite your own prior and overlapping work properly.3
• Choose the most appropriate journal and submit the best manuscript possible.4 Never knowingly submit a poor manuscript hoping for the editors and reviewers to help you fix it.
• Spend the time to understand the submission requirements and comply with those requirements.
• Select the list of authors appropriately, with full approval of the submission by all authors.5
• Identify all funding sources and make the editors aware of any potential conflicts of interest.
• Provide a transparent process for editorial review, and deviate from that process only under exceptional circumstances.
• Deal fairly and respectfully with all parties in the publishing process.
• Recuse yourself when dealing with a manuscript for which you have a conflict of interest—let a nonconflicted editor handle the submission and make the decisions.
• Ensure that all details of a submission are kept confidential.
• Work assiduously for timely decisions.
• Choose reviewers who are likely to provide fair, unbiased, high-quality, and timely reviews.
• Disclose any conflicts of interest (arising from competitive, collaborative, financial, or other relationships) that might bias your opinions of the manuscript. If you are chosen to review despite a conflict of interest, do your best to provide an unbiased review.
• Return the review quickly. If you are unable to return a quality review in a timely manner for any reason, let the editors know as soon as possible.
• Provide a constructive, professional review — it should never get personal.
• Provide a detailed review, supporting all opinions with evidence; your goal should be to help the authors improve their paper even if you recommend rejection.4
• Hold information gained from reviewing confidential. Never disclose or use knowledge gained from reviewing a manuscript until that manuscript has been published.
Criticisms of the Peer-Review Process
• It is unreliable, frequently failing to find major flaws in the work, including fraud and plagiarism.
• It is neither consistent nor objective, and is often biased in several ways.12
• It is expensive and delays publication.
• There is little evidence that it is effective, let alone the best method available.
• Most rejected articles are eventually published in another peer-reviewed journal.
I have to admit that each one of these points has some validity. The peer-review process is not, and never will be, perfect. However, there is a growing body of evidence that peer review works in its intended goals of filtering and improving papers.13–15 A recent survey found that 91% of authors thought the peer-review process had improved their last published paper.16 There are many flaws in the process, but as former BMJ editor Stephen Lock wrote, “We have no better way of distinguishing between the promising and the meretricious or for improving the scientific and linguistic qualities of an article.”8
Peer review has evolved significantly since it was first introduced in the mid-eighteenth century,17,18 and it continues to evolve today. Technology has drastically sped the process, with email, web-based submissions, and online publishing. Search-engine-style document comparisons do a reasonable job of detecting plagiarism. But in the end, it is the careful reading of a manuscript by editors and expert reviewers that makes the whole process work. Science is a human endeavor, with scientific quality dependent on the attitude, training, and work ethic of the scientists involved. Likewise, scientific journal publishing depends on the efforts of well-trained and hardworking scientists and engineers who choose to give back to their scientific community by volunteering for their journal. To the editors and reviewers of JM3, I can never say thank you enough.