A response to the report on gender bias in publishing in the chemical sciences

The Royal Society of Chemistry launched a report in November 2019 that summarised an analysis of publications in Society journals between January 2014 to July 2018 (RSC, 2019). Their report “Is publishing in the chemical sciences gender biased?” and corresponding research article (Day et al., 2019) explored the presence and extent of gender bias in publishing in Society journals. In his foreword to the report, Robert Parker, Chief Executive of the Royal Society of Chemistry summarised:

“We found that there is a complex interaction of subtle biases occurring throughout the publishing pipeline, which combine to put women at a disadvantage when disseminating their research.”

This finding about publishing in the chemical sciences necessitates close scrutiny of author actions and editorial practices in each of the chemical sciences sub-disciplines, including our own practices in Chemistry Education Research and Practice (CERP). Below I set out a response to the report including actions that will arise. Before I do so I want to state that I think the Society should be acknowledged for publishing this analysis of their own journals – and indeed I join their call for other major publishers to do so – and to state that the views expressed below are my own as Editor of CERP, and not those of the Society.

The key message in this Editorial is that CERP will be enthusiastic participants in any activities the Society proposes to eliminate gender bias in publishing. We will seek to lead our peers by showcasing exemplary editorial practices, both in the managing of submissions as well as documenting and explaining aspects of the publishing to our community. CERP has already a long tradition of improving awareness about various aspects of publishing, with a series of articles from our previous Editor, summarised in a recent Editorial offering guidance for those looking to submit manuscripts to this journal (Seery et al., 2019).

The analysis summarised in the report was conducted on the Society's full suite of journals of which CERP is one. 47 journals are named in the Supporting Information to the corresponding research article (Day et al., 2019), and at the time of writing, data on individual journals is not in the public domain. Therefore it is not possible to label any specific conclusions to this journal. However, on the basis that a gender bias was found for the complete suite of RSC journals in this study, and gender biases have been found in other large scale studies (Freeman and Huang, 2015; Elsevier, 2017; Murray et al., 2018), this Editorial is written with the hypothesis that there is a gender bias in publishing. Furthermore, while outside the scope of this Editorial, the report is set in the context of two previous reports published by the Society: “Diversity Landscape of the Chemical Sciences” and “Breaking the Barriers”, and readers are encouraged to read those reports to gain insight into systematic barriers in place for women in the chemical sciences (RSC, 2018a, 2018b). Even if individual journal data were to show CERP having better metrics than those achieved overall by other journals in the suite of RSC publications, the wider landscape shows there is no room for complacency.

The Editorial Processes at CERP

Our Editorial processes have been previously outlined (Seery et al., 2019), but are summarised again here as it will be important to consider the findings from the report in that context. RSC journals have a range of Editorial models, with some journals having full time editors, and others, such as CERP, using the “Associate Editor” model. In this model, the publishing process is:

• Submission: Authors submit a manuscript to a journal.

• Initial Assessment: Editor makes an initial assessment to consider whether the manuscript is within scope and whether it is of sufficient quality and selects one of two options:

– Assignment to Associate Editor: The manuscript is deemed by the Editor to be within scope and of sufficient quality (or potential quality) to undergo peer review. At CERP, Associate Editors may also conduct their own pre-screening judgement.

– Reject without Peer Review: The Editor decides that the manuscript is not within scope or is not of sufficient quality (the context of what is meant by “quality” is defined elsewhere (Seery et al., 2019)).

• Peer Review: Associate Editors select two reviewers, including up to one of the author-suggested reviewers where appropriate.

• Decision: The Associate Editor considers the two reviews and also their own judgement of the manuscript and selects one of the following options. In some cases, a third review may be sought to inform the decision:

– Minor Revision: The manuscript is accepted in principle with minor changes or additional clarity and/or detail required. The revision is not usually sent to reviewers again unless there are specific details that warrant reconsideration.

– Major Revision: The manuscript needs substantial revision and will be sent for review again on resubmission.

– Reject and Resubmit: The manuscript needs substantial revision usually involving additional data or reconsideration that is not feasible within the timeline of major revision (3 months).

– Reject: There is a (usually unanimous) decision that the manuscript is not suitable for CERP because of either scope or quality.

• Accept: After any required revision cycles, manuscripts are accepted and passed to production for publication.

Summary of report findings

The report summarises an analysis of the gender profile of 717 108 manuscript submissions to Society journals between January 2014 and July 2018, and 141 073 citations between journals from August 2011 to September 2018. The process involved assigning gender to authors according to established protocols, described in the methodology in the accompanying research article, removing from the dataset any data where gender could not be assigned (Day et al., 2019).

Population of women in the chemical sciences

The analysis reports that the proportion of female authors submitting to Society journals was 35.8%, which they state tallies with other measures of gender balance in the chemical sciences community. Using this benchmark, the report notes that the proportions of first authors who are female is slightly higher at 36.9%, but the proportion of corresponding authors who are female is 29.2%. The proportion of reviewers who are women is much lower than the benchmark at 24.5%, while the proportion of editors who are women is higher than the benchmark, at 40.5%. A note regarding the headline figure relating to editors is that this includes those journals where the journal model is based on full time editors who are RSC staff members, among whom there is a much higher proportion of women than that of the external associate editors, which has a female percentage in line with the baseline for chemistry researchers (Day et al., 2019).

Pre-screening and acceptance rates

The report states that across all journals published by the Society, initial submissions from female corresponding authors are slightly more likely to be rejected without peer review. The success rate for female first authors to progress to peer review is 69.78%, while that for males is 70.74%, with the difference being noted as significant (Day et al., 2019). Furthermore, when considering the overall acceptance rate for female corresponding authors is 47.38%, whereas for male corresponding authors the success rate is 50.1%; a difference also reported as statistically significant.

Citations

The report summarises the extent to which men and women cite articles, and states that papers by male authors are cited more (average citation of 7.2 times) than papers by female authors (5.6 times). Male authors tend to cite female authors less (17.8%) than female authors cite female authors (20.6%).

Interpretation of findings in the context of CERP

The report details other conclusions but the above three categories are extracted in this Editorial as they point to different aspects of the review process as well as the actions of different stakeholders involved in the journey of a manuscript from submission through acceptance to citation. The main data extracted for consideration here is summarised in Table 1. For the purposes of discussion, I will offer commentary below on each these categories.

Table 1 Data extracted from RSC report (RSC, 2019) and corresponding research study (Day et al., 2019) for all journals published by the RSC

Category	Main finding
Population	Female population in study: 35.8%
	Female first author: 36.9%
	Female corresponding author: 29.2%
Involvement in review process	Female reviewers: 24.3%
	Female editors: 40.5% (including in-house RSC editors for certain journals)
Manuscript success	Manuscripts that progress to peer review
	• Female corresponding author: 69.78%
	• Male corresponding author: 70.74%
	Overall acceptance rate (including pre-screening)
	• Female corresponding author: 47.38%
	• Male corresponding author: 50.1%
Citations	Average citations for:
	• Papers where corresponding author is female: 5.9
	• Papers where corresponding author is male: 7.2
	Citations by:
	• Female corresponding authors to papers that have female corresponding authors: 20.6%
	• Male corresponding authors to papers that have female corresponding authors: 17.8%

---

Stage: submission to peer review

The number of papers submitted by women is lower than the population of women in the chemical sciences. The report also notes that women are less likely to submit to high impact journals than men. While there is only one education journal in the Society's suite of publications, CERP is fortunate to have an excellent reputation and impact with comparable disciplinary journals and within the field of chemistry education is high impact. The first proposed action then is to continue to make clear the submission requirements and peer review processes, in a constructive manner, so that all authors can work on the basis of informative and accessible guidelines that encourage submission of high quality work to CERP.

Action 1: Continue the policy of explaining guidance for quality in editorials to empower authors to submit high quality work to CERP.

Manuscripts that are rejected without peer review are slightly but significantly lower for female corresponding authors, and journal specific information will be valuable in monitoring this process. An Editorial challenge in pre-screening is balancing the desire of authors to receive a critical overview of their work from peer reviewers against not demanding too much of reviewers in the extent of work asked of them in completing a review. However, there is certainly potential in offering authors some more detail on pre-screening decisions by way of specific editorial feedback. This is often the case now – especially when manuscripts are within scope and show potential for the quality required (and indeed in these cases we do, on balance send manuscripts out for review, or at least a second consideration by an Associate Editor) – but there is potential for improvement in the standard pre-screening decision email to offer more guidance to authors that is journal specific.

Action 2: Monitor pre-screening rates for CERP and act to improve feedback on pre-screening.

Stage: peer review

One of the most striking statistics is the low percentage of women who act as reviewers compared to the population as a whole. Some data specific to CERP will be valuable, but a core action must be to ensure that our reviewer population matches the field in which it sits. The research study gives more detail on the review process (Day et al., 2019) and the awareness that this report brings means that we can be more mindful of the nature and tone of reviews, interjecting as necessary. Again this is often the case in the normal course of the editorial process, and CERP is very fortunate to have reviewers that are very proactive in providing constructive feedback, but the report encourages a more systematic consideration to be established. We can assist this process by informing reviewers on the aspects of a constructive and valuable review, and this will be the subject of a future editorial. One additional aspect that can be considered in this phase is to task reviewers, who will be experts in the topic they are reviewing, with ensuring that relevant key literature is cited.

Action 3: Prepare an Editorial on the provision of useful and constructive reviews; monitor statistics of reviewer selection and set out standards for high quality reviewing.

As well as these actions that we will consider locally within the CERP editorial processes, the Society has committed to four key areas of action across all of its journals:

(1) To increase transparency by publishing annually an analysis of authors, reviewers, and editorial decisions by sub-discipline;

(2) To recruit and train reviewers, editorial board members, and associate editors to reflect the current gender balance of the research community;

(3) To provide training and resources to empower editors to eliminate bias;

(4) To encourage intervention to drive change in the academic publishing industry more generally.

The report concludes with the statement from Professor Melinda Duer and Dame Athene Donald which sets out a challenge to all of us in this community:

“Our challenge to publishers, editors and referees alike is to do more to check at every stage that there is no lurking bias, implicit or explicit – and to think about the knock-on effects, for gender equality, of everything they do.”

The Society brings together all of the journal editorial boards in February 2020 at its Editorial Symposium where we will hear how practical actions can begin to address the findings of this report, and the actions proposed in this Editorial will be presented to the journal's Editorial Board. CERP will seek to be at the forefront of activity that contributes to eliminating any bias in publishing, both in our own journal, and as a means to demonstrate exemplary practice for other publications within the RSC suite of publishing as well as the wider publishing landscape.

References

1. Day A., Corbett P. and Boyle J., (2019), Is There a Gender Gap in Chemical Sciences Scholarly Communication?.
2. Elsevier, (2017), Gender in the Global Research Landscape, retrieved from https://www.elsevier.com/_data/assets/pdf_file/0008/265661/ElsevierGenderReport_final_for-web.pdf.
3. Freeman R. B. and Huang W., (2015), Collaborating with people like me: ethnic coauthorship within the United States, J. Labor Econ., 33(S1), S289–S318.
4. Murray D., Siler K., Lariviére V., Chan W. M., Collings A. M., Raymond J. and Sugimoto C. R., (2018), Gender and international diversity improves equity in peer review, BioRxiv, 400515.
5. RSC, (2018a), Breaking the Barriers: Women's retention and progression in the chemical sciences, retrieved from https://www.rsc.org/globalassets/02-about-us/our-strategy/inclusion-diversity/womens-progression/media-pack/v18_vo_inclusion-and-diversity-_womans-progression_report-web-.pdf.
6. RSC, (2018b), Diversity landscape of the chemical sciences, retrieved from https://www.rsc.org/globalassets/02-about-us/our-strategy/inclusion-diversity/cm-044-17_a4-diversity-landscape-of-the-chemical-sciences-report_web-2.pdf.
7. RSC, (2019), Is publishing in the chemical sciences gender biased? retrieved from https://www.rsc.org/globalassets/04-campaigning-outreach/campaigning/gender-bias/gender-bias-report-final.pdf.
8. Seery M. K., Kahveci A., Lawrie G. A. and Lewis S. E., (2019), Evaluating articles submitted for publication in Chemistry Education Research and Practice, Chem. Educ. Res. Pract., 20(2), 335–339.

---