Decision trees provide an instant visual summary of the outcomes of your peer review process, offering details that may otherwise get buried when data is presented in a tabular form. Decision trees offer an overview of a process and details of that process all at once. They can be displayed either vertically or horizontally. The figure above is an example of a vertical decision tree that outlines the initial decision ratios for a journal. When discussing a vertical decision tree, you should move from the top to the bottom and discuss the elements of each row in comparison.
In this example, the blue box shows the number of manuscripts that received a decision during the set time frame, in this case the year 2019, as shown in the chart’s Parameters section.
The next row separates the manuscripts that received immediate decisions from the manuscripts that went through peer review. Immediate decisions are decisions that are rendered by Editors without any input from peer reviewers.
When creating an initial decision tree, you must decide how you want to handle manuscripts that do not have an actual “immediate reject” or “immediate accept” decision but did not go through peer review. This might occur because your journal does not have an immediate reject or immediate accept decision type or due to the manuscript making it through your journal’s initial triage only to be later rejected by the assigned editor without it being sent on to peer reviewers for full peer review. In Origin Reports, the user is given the option of deciding where to place these types of manuscripts by answering a series of questions before generation of the chart.
In this example that shows 159 manuscripts that received an initial decision in 2019, the journal rendered 44 immediate rejections and no immediate accept decisions with the remaining 115 manuscripts being sent through peer review.
The percentage of manuscripts that end up in each of these categories is also helpful to understand. These percentages can help a journal get a feel for editorial staffing needs. If there is a very high percentage of immediate decisions, this might indicate that your journal has very high standards for manuscripts that they are willing to send to peer review. This closer scrutiny will usually take more time and could indicate that your journal might need a larger triage team to ensure a bottleneck at this stage does not emerge. A large percentage of manuscripts being fed into full peer review, could indicate that your team might need to continually monitor the number of editors available to handle the stream of manuscripts, as well as a robust reviewer pool capable of handling the volume of material sent to review, so that these manuscripts can move seamlessly through the peer review process.
The final row of boxes in this decision tree shows how manuscripts that were subjected to full peer review fared. In our example, another 30% (n=35) of the manuscripts were rejected after peer review, while the remaining 80 manuscripts were either accepted or received some type of a revision decision. A very high percent rejected in the full peer review box might indicate that your journal could benefit from a more rigorous, or forceful, initial look (triage) process prior to assigning the manuscripts to editors. Removing clearly unpublishable or unwanted manuscripts prior to editor assignment and peer review keeps these resources from being overburdened.
Though the Immediate Reject box and the Peer Review box show the rejection rate at various stages, it is good to show an overall rejection rate for your journal. This can be calculated by adding together the number of manuscripts rejected at each stage and dividing it by the total number of manuscripts that received a decision in that time period (multiplied by 100 to form a percentage). This is usually a final bit of information that your editorial staff will want to know, since the rejection rate is typically reported as a single value.
Decision trees are easy to interpret and a fast way to absorb complex information, such as the breakdown of initial decisions. They can also be helpful at giving you a better understanding of how your processes are running and provide signposts to where initial assessment criteria or staffing levels require adjustment.