On January 31 2023, we launched our new model for publishing, combining the speed and transparency of preprints with the expert scrutiny and evaluation of peer review. We committed to publish everything that we reviewed. We would publish preprints together with public reviews and assessments. By doing so we would reshape the purpose of a journal.
In adopting our new model we hoped to persuade the research community to embrace change. We could help researchers choose alternative ways to share their research, convince institutions and funders to look beyond one dimensional measures, and encourage journals to offer publishing models that embody the principles of open science.
We know change can be hard. So to everyone who has been a part of our journey and our mission – thank you.
The editorial leadership of eLife share their reflections on our first year and the road ahead here.
12 months of our New Model: 3 key findings
- We received over 6,200 submissions and published more than 1,300 Reviewed Preprints
- Reviewed Preprints help communicate your research more quickly
- Authors rate the quality of our public reviews and eLife assessments highly
Over 6,000 submissions
More than 6,200 teams of researchers have submitted their research (fig.1), choosing our journal and publishing model. Our month-on-month submissions have been stable since launch, but this January marks the highest number so far with 615.
At present, 27.7% of submissions (fig. 2) to our new model are sent for review. This is compared to 31.4% of submissions sent for review in our legacy model (from February 1, 2022–January 31, 2023). It is important to note that we don’t ascribe value to the decision to review. Our aim is to produce high-quality reviews that will be of significant value but we are not able to review everything that is submitted.
The demographic characteristics of corresponding authors (and peer reviewers) – self-declared, by gender and whether someone identifies as a member of an underrepresented or minority group – is very similar between the two models (tables 1 and 2).
Time in review
Reviewed Preprints help researchers share their peer-reviewed and assessed research faster. For a Reviewed Preprint, the median time from submission to publication of the first version with reviews and eLife assessment is 91 days (fig. 3). This is over two and a half times faster than the median submission to publication time in the legacy model. This ensures that peer-reviewed research is available to read, cite and share, where it can begin to have tangible impact, as soon as possible.
Reviewed Preprints, Assessments, and Revisions
By the end of January 2024 we had published 1,332 Reviewed Preprints, and this figure rises to 1,836 when we include revised versions. These articles have been viewed over 850,000 times by more than 320,000 readers.
eLife assessments use a common vocabulary to describe the significance of the findings and the strength of the evidence reported in a Reviewed Preprint. On significance, 50% of Reviewed Preprints were described as important or stronger during the first round of peer review and the strength of evidence was described as solid or stronger for more than 70% (table 3).
For the revised versions published to date, the changes in significance have been modest, but the changes in the strength of evidence have been larger. In particular, articles originally described as incomplete improved in 77.5% of cases after revision (62/80) and articles first described as inadequate improved in 80% of cases (12/15).
Moreover, based on the terms used by reviewers and editors to describe the significance and strength of evidence, the quality of submissions under the new model is very similar to the quality under the previous (legacy) model (figures 4 and 5).
This shows the importance of peer review and revision in helping researchers make substantive improvements to their papers. (Though, peer review also steers more subtle improvements which may not result in different terms being used in an eLife assessment but are still of value to authors and readers.)
That this all happens publicly allows greater accountability, and our model removes the onus on peer reviewers to make binary accept/reject recommendations for papers when the reality is often far more nuanced.
Author experience
According to 325 responses to our author survey collected from April to September last year, the quality of our journal remains the top reason for authors choosing to submit with us. The second reason for authors choosing to publish in eLife is to experience our new model, closely followed by the strength of our editorial board and consultative review process. Authors rated the overall quality of our public reviews and assessments highly (4.4 out of 5).
Introducing a new way to publish has revealed many complex challenges, both for how we engage researchers on what we’re doing and why, and for how we work day to day as an organisation. But eLife was created to innovate and improve the way research is communicated. Building on eLife’s strong reputation gave us a platform from where we could lead much-needed reforms in publishing. It demonstrates our commitment to change, and it proves to scientific and publishing communities that change is possible.
“We are encouraged by the early success of our model, shown through the number of authors trusting us to review their work and we have demonstrated the value of a publishing system that facilitates, rather than impedes, scientific progress.”
– Fiona Hutton, Head of Publishing at eLife
We will build on this work by encouraging wider community support for preprints and open research evaluation, and by creating a shared infrastructure that enables other parts of the academic community to move towards models that put more open and meaningful discussion of research first.
Supporting Data
Unless indicated otherwise the data pertains to February 2023 - January 2024.
Figure 1: New model submissions, submissions sent for review, and published Reviewed Preprints
Figure 2: Percentage of new model submissions sent for peer review
The values shown are calculated using the number sent for peer review and the number of submissions declined each month, and excludes submissions under consideration but not sent for peer review.
Figure 3: Median number of days between submission and publication of the first version of the Reviewed Preprint
Figure 4: A comparison of the evaluation of significance as a proportion of total publications in eLife’s new model and eLife’s legacy model
Figure 5: A comparison of the evaluation of strength of evidence as a proportion of total publications in eLife’s new model and eLife’s legacy model
Table 1: New model |Demographic characteristics of corresponding authors and peer reviewers
Table 2: Legacy model|Demographic characteristics of corresponding authors and peer reviewers
Table 3: Use of terms in Reviewed Preprints (v1)
Term | Number of times used | % times used |
Significance | ||
No term used | 2 | 0.1% |
Useful | 196 | 13.8% |
Valuable | 508 | 35.6% |
Important | 617 | 43.3% |
Fundamental | 99 | 6.9% |
Landmark | 3 | 0.2% |
Evidence | ||
Inadequate | 50 | 3.2% |
Incomplete | 407 | 26.3% |
Solid | 504 | 32.5% |
Convincing | 376 | 24.3% |
Compelling | 209 | 13.5% |
Exceptional | 3 | 0.2% |
#
We welcome comments and questions from researchers as well as other journals. Please annotate publicly on the article or contact us at hello [at] elifesciences [dot] org.
For the latest in published research sign up for our bi-weekly email alerts. You can also follow eLife on Twitter, Mastodon, Facebook or LinkedIn.