Speeding Up The Review Process: SISC Editorial Policies and Procedures

This document contains an update on editorial policies from SIAM for reducing the reviewing time of papers. The various types of reminder emails and services that SISC has implemented are thereafter described.

SIAM Polices (2012)

In 2012, SIAM attempted to implement many of the suggestions in a report by Nick Trefethen on how to shorten the average time from submission to acceptance of papers (an update to the original report appeared as a SIAM News article). The new suggested SIAM policies, also adopted by SISC, are summarized below.

Editors are encouraged to turn around manuscripts promptly. In particular, they should assign referees as quickly as possible, and are reminded that it is entirely appropriate to reject obviously weak manuscripts without sending them to referees. It is also often appropriate to make decisions based on two contradictory reports if the editor has sufficient information.
SIAM's journals depend on the work of thousands of volunteer referees. Words like "deadline" and "overdue" should never be used in making requests of such volunteers. Gentler expressions like "target date" are preferred instead.
By default, referees seeing a manuscript for the first time should normally be asked to respond within at most 2 months, except in the case of Survey and Review manuscripts for SIAM Review. Editors are free to make exceptions for individual manuscripts, and referees are free to request extensions.
Shorter response times should be requested when papers that have undergone minor revision are sent to referees for a second or third time. Some minor revisions can be approved by editors without sending back to referees.
By default, authors should normally be asked to produce minor revisions within at most 1 month, and major revisions within at most 2 months. Extensions will be readily granted if they wish to take longer. Unrevised manuscripts won't be removed from the system until many months later, but we should encourage more rapid action in all phases of the process.
SIAM will investigate the ongoing development of systems that might be put in place so that editors and possibly referees are reminded automatically of approaching and passed target dates, always employing reasonable language (item #2). The details will depend on what is technically feasible and will be worked out by the SIAM office in consultation with Editors-in-Chief.
SIAM's aim is that the average interval between submission of a manuscript and final acceptance for publication should be no longer than about 8 months.

Summary of SISC's Reminder Emails

Regarding item 6 above, SISC has been proactive about implementing various types of reminders to reduce delays:

There is a reminder to the AE if the referees are not assigned after 7 days.
There is an autogenerated reminder to the AE when a referee has not responded after 7 days.
There is an autogenerated reminder to the AE if the referee report is not in at the day after the due date. A repeated reminder to the AE is sent three weeks later if the report is still missing.
The AE is notified in email when all reports are in and a decision can be made.
Brittni sends an email to the AE 14 days after the last referee report arrived in case no decision has yet been made.
Every two months, Brittni sends each AE a status report with an overview of all the papers the AE is responsible for in the web system (this report was sent every three months in the past).
Karen Støverud identifies delayed manuscripts and sends personal emails to the relevant AEs about four times a year.

Desired Time Schedules for Paper Handling

The desired time periods for various actions are implicit in the first five points above. Our aim is to have potential referees assigned within one week and then have acceptance from the referee within another week. Referees should be reminded when the target date for the review has passed, and a decision should be made at most two weeks after the last report has arrived. Very often, decisions can be made quickly after all reports are in. We know, of course, that there are many natural exceptions to this ideal scheme in the real world. However, we hope reminders can help all of us to reduce unnecessary delays in the handling of SISC papers.

There is one circumstance in which we ask all AEs to always respond quickly, and that is when authors write back about referee reports and decisions. Even though more time may be necessary to respond properly to the inquiry, please make sure to confirm receipt of the email and indicate when you will come back with a more complete answer. Do not hesitate to contact me or the section editors if you need to discuss feedback from authors and the possibility for resubmission.

Chaser Emails

All follow-up measures and autogenerated reminders described above are with regard to AEs only. The way in which referees are reminded is up to each AE, but using the chaser emails in the web system is strongly recommended, as this simplifies the task of tracking paper handling. In general, we recommend communication with authors, referees, section editors, or myself through the web system (click on "Send Manuscript Correspondence" under the heading "More Manuscript Info and Tools" in the manuscript main page).

Sometimes the target date of a review is adjusted as a result of communication between the AE and the referee. Since the reminders to the AEs about missing referee reports are generated based on the recorded due date in the web system, it is important to adjust this date accordingly. Under the "Advanced Manuscript Tasks" heading at the bottom of the manuscript main page, you can click on "Extend Review Deadline" and fill in the number of days from "Review Start Date". The system is not capable of reading absolute dates; only time intervals are possible. Knowing well that even scientific computing experts might struggle with "date plus days" calculations, I have included some help at the end of this email. You can also ask Brittni (email sisc@siam.org) to a referee's due date on your behalf.

Let me remind you that the journal's communication with authors and referees is performed by the AE; however, the SE or the Editor-in-Chief can take over such communication whenever the AE wants it.

SISC's long-serving editor Margaret Wright previously brought attention to the wording of chaser letters to referees: such letters should be phrased in a very positive tone and avoid demands to deliver the report as well as absolute words such as "deadline". Her comment resulted in item 2 in the new SIAM guidelines above. And as Nick says in his SIAM News article, "We are relying on expert labor contributed gratis by busy colleagues". Please have this in mind when writing chaser emails.

Personal Email Templates

Different AEs and different referees represent different cultures. We therefore strongly recommend that you edit the default SIAM/SISC chaser template the way you desire, and that you adapt the template to the referee in question. As an example, my personal templates start with an appreciation of the willingness to referee before asking, in a soft tone, when she or he expects the report to be ready. In the exceptional cases when collecting a report proves really difficult, it can sometimes be effective to send friendly reminders weekly.

To edit the template emails, find the heading "General Tasks" on the editor's main page, and choose "Individual Preferences: My Email Templates". Any of the template letters can now be edited by clicking on "edit".

One important point: for those of you who have already customized the templates for minor and major revisions, notice that we now want the author to submit the revised manuscript within one and two months, respectively. Please find your templates and make the corresponding edits! All journal-wide default templates have been updated by SIAM, but your personal versions of minor and major revision decision letters will override the journal-wide template.

Rejection Letters

Template letters can be sufficient in many cases, but rejection letters normally deserve an edit and extension of the template. Unless all the referee reports are very clear about the reasons for rejection, the editor should briefly explain the background for the decision. Quite often there is a positive and a negative report, and since acceptance in the end requires two positive reports, the editor must judge whether it is likely possible to revise the manuscript in a way that the opinion of the negative referee can be changed. Also, remember that we always attach all available reports, positive or negative, to the rejection letter. The exception occurs only when a report does not meet a reasonable scientific standard.

As we speak of rejection, the first point in the SIAM/SISC policy above says that "it is entirely appropriate to reject obviously weak manuscripts without sending them to referees". Therefore, I encourage all of you to invest some effort in assessing manuscripts where you spot weak elements, and see if early rejection is a good choice. It is much better for the authors to get the paper quickly back and then consider other journals or perform substantial changes than to wait several months for a decision that most likely will be negative. A good portion of the incoming submissions to SISC are immediately rejected, either by the SEs or me. However, an AE often has expert competence on the topic that may uncover major weaknesses we have not seen.

Appendix: Date + Days

To calculate the due date for an extended review period of 90 days, starting on July 15, 2012, you can write in MATLAB:

datestr(datenum('15-July-2012')+90)

The corresponding Python statements are:

import datetime as d
print d.date(2012, 07, 15) + d.timedelta(90)

Those who prefer C++ and a page of code can take a look at http://stackoverflow.com/questions/2344330/algorithm-to-add-or-subtract-days-from-a-date.