A recent post by Elsevier on a study by Chetty et al (http://www.elsevier.com/reviewers/reviewers-update/how-small-changes-can-influence-reviewer-behavior) and a follow-up post on the Scholarly Kitchen (http://scholarlykitchen.sspnet.org/2014/05/28/what-motivates-reviewers-an-experiment-in-economics/) addressed an issue core to the Rubriq team – motivating reviewers. The study confirmed what we already knew about the effects of shorter deadlines and paying for reviewer time. However, the social piece discussed could have been an experiment on its own. Instead of just using email reminders, there could be a peer critique element introduced depending on the system used. We hope to pursue that further with our network of Rubriq reviewers, but as a supplement to payments, not a replacement.
It’s always interesting to see the comments that come from discussions regarding paying reviewers. As some of the comments seen on the Scholarly Kitchen article suggest, the act of providing a $100 honorarium/stipend is quickly extrapolated into a commercialized market steeped in corruption. However, the bias that money=evil conflicts with many other discussions we’ve seen about how reviewers are expected to work for free, not receiving any reward for their time in a monetary model that favors the journal/publisher. There is a big difference between providing a moderate payment as an incentive to meet a deadline on a single review, and paying a full-time salary with large bonus incentives.
When we first launched Rubriq, we worried that we would face community criticism about our policy to pay reviewers. However, we have received very little negative feedback from the researchers we have contacted. Less than 10% of our reviewers have either opted-out of receiving payment or are ineligible due to residence or employment restrictions. But most are happy to accept a monetary reward for their time and effort.
We recently surveyed our reviewers to find out more about their motivations. The leading reason was to receive financial compensation for their work/time. But this 33.2% was a narrow victory over intrinsic motivations such as gaining more peer review experience (32.5%), supporting the mission behind Rubriq (15.1%), to help create standards for their field (12.8%), and even to get to read new papers in their fields (3.2%). Like most decisions, we assume that a researcher decides to review an article based on multiple factors, not just a single motivator. In our survey we asked the reviewers to select just one reason as their primary motivation, but in future research we may change to a ranked sort or percent distribution. Because the decision to review has many potential sources of influence, we feel it is unlikely that these same researchers would accept a shift to a completely commercialized review market.
Payment could help compensate for some other elements that might be lacking – papers that are submitted to a journal, but that appear to be low on the novelty/interest scale might languish for months at a journal waiting to attract attention. Outsourcing peer review of “challenging/problem” papers to a service like Rubriq would make journals more efficient but not require across the board reviewer payments – they could still get interested reviewers for the sexier papers to do it for free, and pay for the benefit of faster/efficient process for others. For reviewers, knowing that they are being compensated for their time can help ease frustration when they are reviewing some of these more challenging papers.
Despite the positive acceptance of our policy to pay reviewers, we are still exploring other types of reward options. Recognition is one aspect we are pursuing, and we hope to work with the CASRAI/ORCID project to integrate reviews done through Rubriq into a researcher’s record. In addition to recognition programs, we asked our reviewers about alternate types of payment. Although our current direct payment option is the most preferred, reviewers have also expressed interest in models in which an equivalent donation is made to the charity of their choice, or in which the funds are deposited into a central account for use by the researcher’s lab/department/institution.
Quality of review is another compelling topic. We were happy to see that our own findings that quality was not adversely affected by speed were upheld by the Chetty et al study. At Rubriq we have even seen an increase in review quality within our system, which is attributed to both our standardized scorecard structure and our in-house quality check of all reviews. The scorecard provides reviewers with a tested instrument that effectively guides them through a comprehensive assessment of the paper. The standardized format enables reviewers to become more efficient over time and to maintain a consistent quality of review. Our internal QA check assesses each review for its ability to highlight the strengths and weaknesses of the manuscript. In addition to ensuring that the customer receives a high-quality assessment, this also adds built-in control to select out any reviewers who are simply filling in the blanks to get a check. In the event a reviewer “phones it in” the reviewer is given the chance to amend the scorecard to provide a more detailed, informative assessment, but is replaced on that manuscript if they chose not to do so.
Some have expressed concern that compensating reviewers for their time will result in the loss of objectivity and honesty, but this fear is based on a very shortsighted perspective. Researchers and publishers do not come to us for glowing reviews; they come to us for honest assessments. Accordingly, we happily pay reviewers for comprehensive, thoughtful reviews regardless of the reviewer’s overall opinion of the work. Our future hinges on our ability to consistently provide this honest feedback to the research community, and because we are not a publisher, there is no financial incentive to bias our reviews or reviewer selection towards positive opinions.
One of the other concerns regarding compensating reviewers is the associated cost to the system. We have found that 15 million hours of researcher’s time is lost each year to performing redundant reviews (and under the current system that number will grow). However, the costs associated with having this amount of time distracting researchers from their primary mission seems to be willfully ignored in the discussion of compensating reviewers. “Free peer review” isn’t free, it is simply paid for in a different currency. This is one of the primary motivations behind Rubriq: if a standardized, in-depth review is performed and it is transferable from journal to journal, then costs associated are apparent and only paid for once. In addition, this system is not simply pulling money from the research community, but is also putting compensation back into the hands of the researchers in the community that are performing the evaluations.
My favorite quotable from the Elsevier post was “… as editors, we shouldn’t believe that the performance of our journals is something we can’t change. We can greatly improve the quality of our journals’ review process through simple policy changes and active editorial management.”
I hope more journal editors adopt this attitude because it could be one of the most impactful and meaningful shifts for the industry. When journals treat all authors as valuable customers and act accordingly to improve services, I think everyone wins. Overall these articles offer a nice validation of our efforts at Rubriq, confirming that we can improve turnaround speed without sacrificing quality. We welcome your comments or questions.
Lisa Pautler, Rubriq | firstname.lastname@example.org