A Date is a Date is a Date…. Maybe Not
March 30, 2022
When you create charts to present your journal data, it is important to note which of the many possible dates you used. For example, when creating a chart to show initial decisions, are you using the submission date or the date of the initial decision as your reference date?
Does it really matter?
Actually, the reference date you select is very important because the date determines the way that you will group the submissions. If you were tasked with putting apples into groups and counting the apples in each group, you can easily see how different your counts would be depending on how you set up your groups. If you grouped the apples by color, you would get different counts than if you grouped them by apple variety or by size. The dates that we use to group our submission data works the same way. If we group all the manuscripts by submission date, we will get different counts than we would if we used different dates, such as initial decision date, final decision date, date the editor was assigned, or date the reviewer was invited. So, what reference date should you select when creating your charts? It really depends on what questions you are trying to answer. We will look at a variety of scenarios.
When reporting the number of submissions received, you will want to use the submission date. However, you should be aware that for many manuscript handling systems, there are two submission dates available, a static (non-changing) original submission date and a dynamic original submission date, which might be updated. When an author submits their manuscript for the first time, the static and the dynamic submission dates are set to the same date value; however, the dynamic submission date will be updated if the submission is returned to the author for any reason and resubmitted prior to assigning it to an editor.
When this occurs, the dynamic original submission date will be updated to the date that the author re-submits the manuscript. Many editorial offices prefer to use the dynamic original submission date as the reference date for submissions charts, since the journal has no control over the length of time that the submission is with the author. In your manuscript handling system, the static submission date may be called: Submission Date – Original, First Receipt Date, etc. These systems use terms such as Submission Date or Initial Date Submitted for the dynamic submission date. In the example above, First Receipt Date is the static submission date and Initial Date Submitted is the dynamic submission date.
Initial Decisions & Final Decisions
Grouping your decisions by the submission date or by the decision date will lead to a different group of manuscripts and, therefore, a different mix of decision types and changes to timing data. This is especially true for journals with large numbers of submissions per year and journals that take a longer time to reach decisions, since they will have more manuscripts that are submitted in one year but don’t receive their decisions until the following year.
Many journals prefer to use the Submission Date as the reference date for their decision charts so that their submissions charts and decisions charts are reporting on the same group of manuscripts. While this is logical, depending on what question you are trying to answer it may not be the best choice.
If you are trying to answer the question “How did manuscripts submitted in 2021 perform at initial decision?”, then using the Submission Date is the best option. Since the static and dynamic submissions dates can be different, it is important to use the same submission date column for all your charts so that your charts are describing the exact same set of manuscripts and can be accurately compared to each other. Also, you need to include a category for submissions that have not yet received a decision. You can call this category “in-process” or some other term that makes it clear that there are manuscripts not included in the other decision groups. Without the in-process category, you could easily give a false impression of the results for that year, and your total decisions values will not match your submissions chart values.
If you want to answer questions about policy or staffing changes, such as “Has time to initial decision improved since we instituted a particular policy?”, then it is much better to use the decision date for your reference date. To understand why that is important, consider the example shown below.
In this example, the journal instituted a new decision type of “Desk Reject” in early 2021. For Manuscript X and Manuscript Z, it would not matter whether the journal was grouping the manuscripts by the Submission Date or the Initial Decision Date since each of these manuscripts would clearly fall in the same year either way. However, consider Manuscript Y. If the submission date is used to group the data, this submission would show an initial decision of Desk Reject as part of the 2020 counts, which would be confusing since the Desk Reject decision type was not put in place until 2021. If the data are grouped by the Initial Decision Date, then this manuscript would be counted in 2021 when this new Desk Reject decision was available.
You also want to be careful not to artificially soften possible trends. For example, if you cleaned up the journal’s reviewer database in early Q1 of 2021, which enabled your editors to find reviewers quicker than was possible in 2020, then there might be 2020 manuscripts that benefitted from that improvement. The average 2020 timings would then appear to be somewhat faster. So, the improvement from the change would not appear as significant in your chart as it would if the data were grouped by decision date, which is appropriate because the date of the decision would be after that change was put in place.
It is a very good idea to keep a record of significant policy or staffing changes so that changes in trends can be fully understood and explained.
Reject and accept decision data also can be affected when grouping by submission date instead of the decision date, especially when reported early in the calendar year. If you are reporting your Q1 rejection rate using the submission date as your reference date, then you can get an artificially inflated number of immediate reject and/or immediate accept decisions because these are quicker to reach. Using the decision date will provide data for all manuscripts that received a decision in a given time period, regardless of submission date.
When looking at editor performance, we want to know the mean (average) or median time it took each editor to reach their initial decision. The time to initial decision is typically calculated as the interval between the date that the editor accepted the manuscript assignment until the editor made their initial decision. Since this important statistic uses the editor assignment date, it is beneficial to base the other editor performance charts on this same date variable so that all the performance charts can be used together to give an overall picture of the editor’s performance.
Most journals look at reviewer data to see how long it is taking the reviewers to submit their review comments. Just as we used the editor assignment date for the editor charts, we use the reviewer agreement date for the reviewer charts. When using the reviewer agreement date, make it clear to your audience that these are stand-alone charts that can be compared only to each other and not to other groups of charts, such as the numbers in the total submissions or total decisions charts for the year.
When creating charts for editorial board meetings or responding to questions asked, be sure you are clear on what your audience needs answered. Then decide which date variable will allow you to answer those questions. The date should be noted on your chart, in the parameter section, so that everyone has a clear understanding of how the data was collected and how to interpret the results.