There is a widespread perception in the academic community that peer review is subject to many biases and can be influenced by the identity and biographic features (such as gender) of manuscript authors.
We examined how patterns of authorship differ between men and women, and whether author gender influences editorial and peer review outcomes and/or the peer review process for papers submitted to the journal Functional Ecology between 2010 and 2014.
Women represented approximately a third of all authors on papers submitted to Functional Ecology. Relative to overall frequency of authorship, women were underrepresented as solo authors (26% were women). On multi-authored papers, women were also underrepresented as last/senior authors (25% were women) but overrepresented as first authors (43% were women). Women first authors were less likely than men first authors to serve as corresponding and submitting author of their papers; this difference was not influenced by the gender of the last author. Women were more likely to be authors on papers if the last author was female.
Papers with female authors (i) were equally likely to be sent for peer review, (ii) obtained equivalent peer review scores and (iii) were equally likely to be accepted for publication, compared to papers with male authors. There was no evidence that male editors or male reviewers treated papers authored by women differently than did female editors and reviewers, and no evidence that more senior editors reached different decisions than younger editors after review, or cumulative through the entire process, for papers authored by men vs. women.
Papers authored by women were more likely to be reviewed by women. This is primarily because women were more likely to be invited to review if the authors on a paper were female than if the authors were male.
Patterns of authorship, and the role undertaken as author (e.g., submitting and serving as corresponding author), differ notably between men and women for papers submitted to Functional Ecology. However, consistent with a growing body of literature indicating that peer review underlying the scholarly publishing process is largely gender-neutral, outcomes of editorial and peer review at Functional Ecology were not influenced by author gender.

Introduction

Scholarly publication is the primary means by which scientific research is communicated among researchers. An individual's research contributions are therefore measured by their authorship on peer-reviewed scholarly articles, and authorship is a major determinant of success when scientists compete for research grants (Kaltman et al. 2014), apply for promotion or tenure (Wren et al. 2007), or receive recognition by their peers (Merton 1973). Though the specifics of what authorship order signifies vary by discipline (Zuckerman 1968; Frandsen & Nicolaisen 2010; Waltman 2012), author order (Shapiro, Wenger & Shapiro 1994; Zbar & Frank 2011; Cleary et al. 2012) and role (e.g., serving as corresponding author; Bhandari et al. 2014) commonly signify either the magnitude or type of contribution made and so can affect the success of scientists (Sonnenwald 2007).

Patterns of authorship commonly differ between male and female scholars. Globally, women comprise <30% of authorships on scientific publications (Larivière et al. 2013; Burns 2015) and publish less frequently than do men (Symonds et al. 2006; Ledin et al. 2007; Long et al. 2015). The proportion of authors that are women has been increasing (Jagsi et al. 2006; Baethge 2008; Dotson 2011), but the degree of change varies substantially among journals and disciplines (Sidhu et al. 2009; Kongkiatkamon et al. 2010; Oertelt-Prigione 2012; Mauleón et al. 2013; West et al. 2013). In particular, men have historically been over-represented as first and last authors while women have been over-represented as middle authors (Martin 2012), though the under-representation of women at the first and last positions, especially first authors, has improved over the past couple of decades (West et al. 2013). Findings such as these suggest either differences between genders in preference for roles in the scientific research process or societal biases that affect the choices men and women make in their careers (Xie & Shauman 1998), either of which may result in different opportunities and rewards afforded to women vs. men in the scientific publishing process (Fox, Burns & Meyer 2015).

Peer review, in some form, has been a key feature of the scholarly communication system for nearly three centuries (Spier 2002; Hunter 2010). Despite this, there is a widespread perception among researchers that systemic biases influence peer review outcomes (Lee et al. 2013). In particular, the gender of authors may influence manuscript or grant peer review due to subconscious, or sometimes conscious, biases (Tregenza 2002). Bibliometric analyses consistently demonstrate differences in publication rates between men and women (Abramo, D'Angelo & Caprasecca 2009; Larivière et al. 2011; Barrios, Villarroya & Borrego 2013), and analyses of grant funding rates commonly demonstrate that men get more grants and/or more money (Bornmann, Mutz & Daniel 2007; Ley & Hamilton 2008; but see Marsh et al. 2009). However, the evidence on whether and to what extent these differences are generated by systemic gender discrimination is limited and inconsistent; differences in funding and/or publication rate can be due to gender disparity in rank and institutional prestige (Nakhaie 2002), research specialization (Leahey 2006), or manuscript/grant submission rates (van Arensbergen, van der Weijden & van den Besselaar 2012) between men and women, rather than discrimination per se.

A few early studies of peer review demonstrated that papers and grants authored by women were less well reviewed or less likely to be funded or published (Wenneras & Wold 1997), and suggested that women reviewers tend to be more favourable towards papers written by female authors (Lloyd 1990). However, few subsequent studies have found gender bias in peer review. Controlled experiments in the life sciences (Borsuk et al. 2009) or other fields (Knobloch-Westerwick, Glynn & Huge 2013), and correlational studies of editorial (or granting) decisions or reviewer ratings in the life sciences (Tregenza 2002; Primack et al. 2009) but primarily in the health sciences and other fields (Lane & Linden 2009; Heckenberg & Druml 2010; Walker et al. 2015 2015 for manuscript review; Dickson 1997; Grant, Burden & Breen 1997; Friesen 1998; Ley & Hamilton 2008; Leemann & Stutz 2008; Sandström & Hällsten 2008; Marsh, Jayasinghe & Bond 2008, 2011; Mutz, Bornmann & Daniel 2015 for grant reviews), generally find that peer review is gender neutral, though there are some exceptions (e.g., Knobloch-Westerwick, Glynn & Huge 2013; Krawczyk & Smyk 2014). However, most manipulative studies use student populations as their raters, and the majority of analyses of peer review done by professional researchers look at grant applications; few have looked at manuscript submissions to academic journals. This is likely because the journal peer review process is typically quite secretive, whereas granting agency data on application and success rates is commonly made public.

The purpose of this study is to test (i) whether patterns of authorship differ between men and women for papers submitted to the ecology journal Functional Ecology and (ii) whether author gender influences the peer review process at this journal. A previous study (Fox, Burns & Meyer 2015) found that editor gender influenced the gender ratio of scientists invited to review (an effect mediated by editor seniority), and that responses to review invitations differed between male and female invitees to review, but that gender did not influence peer review scores given by reviewers or decisions made by editors. Here, we extend our analyses to the influence of author gender on the peer review process, and test for possible interactions between author and either editor or reviewer gender.

Materials and methods

Data Set

Functional Ecology uses ScholarOne Manuscripts (previously Manuscript Central) to manage manuscript submissions and peer review. We extracted data on the editorial and peer review process from ScholarOne on 19 December 2014 for all ‘standard’ papers submitted to Functional Ecology between 1 January 2010 and 30 June 2014 (inclusive). ‘Standard’ papers include all typical research studies (empirical or theoretical), but exclude review papers, commentaries, perspectives, editorials and other types of papers not considered typical research manuscripts. We chose 30 June 2014 as our cut-off to ensure that all papers examined had completed the editorial process at the time the data were extracted from ScholarOne. Over this time period (January 2010–June 2014) the journal received 3528 submissions of standard papers. Of these standard papers, 2298 were assigned to a handling editor, 1770 papers were sent out for peer review, and 551 were invited to revise and/or were accepted for publication.

Authorship on Papers

Data in ScholarOne is author-entered and so author lists in the database are sometimes incomplete and usually ordered incorrectly. We thus determined authorship on papers from the cover page of the submitted manuscript and not from the data extracted from ScholarOne. PDFs of manuscripts submitted to Functional Ecology were available in ScholarOne only from 2010 onwards, which limited the time frame of our study. PDFs were only available for one version of the paper. For papers that were rejected, the PDF available was always the original version of the paper. For papers that had been revised, the PDF available was usually the last version viewed by the handling and/or senior editors. This difference in which version of a paper is available will only introduce error in our analyses if authorship on the paper changed during the manuscript revision process; though minor authors are sometimes added to papers in revision, it is exceedingly rare for the first, corresponding or last authors to change between versions.

Authors were categorized as First Author (the first name listed in the author list), Last Author (the last name listed in the database), Corresponding Author (the author identified on the cover page of the manuscript as author to whom correspondence should be sent) and/or Submitting Author (the author who actually submitted the manuscript in ScholarOne). Corresponding and submitting authors were usually but not always the same. A small percentage of papers (2·6%) had no corresponding author listed on the paper, or had two corresponding authors listed; these were excluded from analyses of corresponding authorship. The last author is included in our analyses because it is common in ecology for the last author to be the senior author, i.e., the person overseeing the laboratory and advising students, post-docs or other scientists on the project. That this is commonly the case is suggested from the observation that the last author (on papers with two or more authors) was the submitting author 18% of the time. However, we cannot know for any specific paper whether the last author is the senior author.

Author and Reviewer Gender

Author gender was determined using the online database genderize.io. This database includes >200 000 unique names and assigns a probability that each name is male or female given the distribution of genders for these names in the database. If the name of the first, corresponding or senior author was not listed in genderize.io, we used an internet search to determine gender (we searched for individual web pages or entries in online databases that included a photograph of the individual or other language indicating their gender). We had the most difficulty genderizing names of authors from Asian countries. This is because genderize.io includes few names from Asian cultures, transliteration into Roman alphabet may obscure differences in names, and because we could not find websites for many of these authors.

Our data set includes 14280 author entries (a mean of 5·2 authors per paper). 128 authors listed no first name on their submission and so cannot be genderized; these individuals are not included in our analyses. Of the remaining authors, 13284 author names (93%) were present in genderize.io; 94% of these names were ≥ 0·70 probability of being one gender or the other, and 85% of names were categorized as ≥ 0·90 probability of being one gender or the other. Considering only authors in non-Asian countries, 97% of author names were listed in genderize.io (with half of the remaining authors successfully found in an internet search), with 97% of these identified to gender with ≥ 0·70 probability and 94% identified to gender with ≥ 0·90 probability.

We also know the gender of almost all reviewers in our database. Reviewer gender was determined similarly to author gender, first using genderize.io, followed by an internet search for all reviewers whose probability of being a specific gender in the genderize.io database was < 0·99. Additional details are provided in Fox, Burns & Meyer 2015.

Author and Reviewer Geographic Location

Our data set contains the geographic location (country) of most authors who submitted a paper to Functional Ecology. To categorize localities, we used the M.49 area codes and their continental regions as defined by the United Nations' Statistical Commission (unstats.un.org). There were two exceptions: (i) we divided the Americas into Latin America (which includes North America south of the United States–Mexico border) and North America (the United States and Canada), and (ii) we divided Europe into the United Kingdom and ‘other Europe’. This second change reflects that a British learned society (the British Ecological Society) owns Functional Ecology and thus the journal receives a large number of papers from British authors (greater representation of authors from the United Kingdom than expected from the distribution of ecologists in the world).

We also know the geographic location of most reviewers in our database. As for authors, reviewer geographic locations are reviewer-entered and reflect the most recent location of the reviewer according to their last database entry. Some scientists move between countries/continents during their careers, and such changes may not be reflected in our database. Nonetheless, we expect most reviewer location data to be correct, with the errors creating random noise in the data.

Editor Gender, Geography and Seniority

The journal editors are all known by the journal staff, so their gender and geographic location are also known. Editor Seniority was calculated as the number of years between when an editor got their PhD and the year they handled each specific paper. Because PhD degrees are awarded at various times of year and we only know the year (not month) of graduation, our estimates of academic seniority should be considered plus-or-minus ~2 years of the actual time between the date of interest and the actual PhD graduation of the editor.

Editorial and Peer Review Outcomes

We limit our examination of the editorial and peer review process to the initial submission of a paper to the journal. Editorial and peer review of revised manuscripts is not considered.

Throughout our analysis, we distinguish three stages of the editorial and peer review process. Firstly, papers are screened by senior editors before being assigned to a handling editor (Paper Assigned to Editor [yes/no]); a large proportion of papers are declined at this stage (Fox & Burns 2015). Associate editors then perform their own screening and decide whether to send the paper out for review (Paper Reviewed[yes/no]. Once reviews are obtained, a final decision is made on the paper (Decision[reject/revision request]).

For each review invitation, the journal also tracks whether the reviewer responds to the invitation email (yes/no), how long each invitee takes to respond to the invitation (in days), whether they agree to review [yes/no], how long each reviewer takes to submit their review (in days) and the review score (range 1–4; details in Table 1 of Fox, Burns & Meyer 2015 and Table S1, Supporting information).

Analyses

For statistical analyses, each manuscript represents a single data point that includes one first author, one corresponding author, one author gender ratio, a single handling editor, and so on. The exception is the analysis of the interaction of author and reviewer gender, for which each manuscript has two data points, one for each gender of reviewer (see below).

Most of the variables we examine in this study have binary responses: the gender of the author (f/m), whether a paper is sent for peer review (yes/no), the decision on a paper (reject/not reject). For analyses in which our dependent variables are binary we used logistic regression with models of the form: DependentVariable = Year + IndependentVariables + Interactions. Year was included in all models (except those sorted by year) as a categorical variable that represents when the paper was submitted. Logistic regression analyses were performed using SAS PROC LOGISTIC or PROC SAS GLIMMIX, both of which provide the same results but allow different modelling options. We used general linear models to analyse time data (time to respond to invitation, time from acceptance to submission of review) and reviewer scores (using SAS PROC GLM). Time to respond to the review invitation and time from acceptance to submission of review were log-transformed before analysis to meet the assumptions of analysis of variance. All categorical variables were treated as fixed effects except that Handling Editor Identity was included as a random effect (nested within Handling Editor Gender) for analyses that include the effect of Handling Editor Gender, and Senior Editor Identity was included as a random effect in analyses of senior editor pre-review rejection rates.

For analyses of authorship patterns on papers we used the full data set (all papers) with each paper treated as a single data point. However, for analyses testing for differential treatment of papers written by male and female authors (i.e., all sections of the results except ‘The distribution of male and female authorships on papers’) we excluded papers for which the first and/or corresponding author is from Asia (511 submission, 14·5% of total submissions). For bias to occur, consciously or subconsciously, it is necessary that editors and reviewers be able to identify author gender from their names. Asian names are less familiar to most editors and reviewers working with Functional Ecology, most of whom are in non-Asian countries. 52% of the names of Asian authors in our data set are either not listed in genderize.io or return a probability of ≤70% one gender suggesting that these names are either not commonly encountered outside of Asia or are commonly androgynous. By contrast, only 6% of non-Asian names are either not listed in genderize.io or return a score of ≤70% of one specific gender.

To examine whether there was an interaction between author and reviewer gender, we limited our analysis to papers that had at least one reviewer of each gender (n = 351).

Results

The Distribution of Male and Female Authorships on Papers

Women occupied 34·5% of all authorship positions on manuscripts submitted to Functional Ecology (averaged across years; Fig. 1). On papers with just one author (n = 122 paper), that author was female just 26·0% of the time. On papers with two or more authors, women occupied a substantially larger proportion of first author positions (43·1% of first authors were women) than expected from their overall frequency of authorship, but a much lower proportion of last authorships (24·6% of last authors were women). Women were also listed as the corresponding author on the cover page for 37·6% of manuscripts and served as submitting author for 37·8% of manuscripts, both less often than they were first authors (Fig. 1). Some of these distributions of authorship varied statistically, but none varied substantially, across the authors' geographic regions (Table 1).

Details are in the caption following the image — **Figure 1**
Open in figure viewer PowerPoint

The distribution of authorships on papers submitted to *Functional Ecology* from 1 January 2010 to 30 June 2014. The gender ratio across all author positions is calculated first across all authors on a paper, and then averaged across papers. The gender ratio at different authorship positions is only for multi-authored papers (excluding single-authored papers).

Table 1. Variation in authorship on papers submitted to Functional Ecology according to the geographic location of the first author on the paper. Means are calculated by first averaging across papers (one data point per paper) within years, and then averaging across years. Proportions for first, corresponding and last authors are for multi-authored papers only

	Number of authors (mean ± std err)*	Proportion female authors (all positions)†	Proportion of authors on single-authored papers that are female¶	Proportion of first authors that are female‡	Proportion of corresponding authors that are female‡	Proportion of last authors that are female‡
Asia	3·67 ± 0·23	0·33	0	0·37	0·40	0·30
Africa	4·31 ± 0·23	0·37	0·26	0·41	0·30	0·29
Europe	4·39 ± 0·11	0·35	0·28	0·45	0·39	0·24
Latin America	4·18 ± 0·11	0·35	0	0·45	0·37	0·19
North America	3·52 ± 0·13	0·35	0·20	0·42	0·39	0·28
Oceania	3·72 ± 0·20	0·32	0·20	0·46	0·39	0·19
United Kingdom	4·23 ± 0·15	0·33	0·33	0·42	0·39	0·22
Variation among years§	F_4,3387= 3·53**	Χ²₄= 3·72ns	–	Χ²₄= 4·87ns	Χ²₄= 5·88ns	Χ²₄= 4·25ns
Variation among regions§	F_6,3387= 13·28***	Χ²₆= 21·2**	–	Χ²₆= 4·24ns	Χ²₆= 12·4ns	Χ²₆= 15·8*
Year-x-region interaction§	F_24,3387= 1·11ns	Χ²₂₄= 16·9ns	–	Χ²₂₄= 26·1ns	Χ²₂₄= 25·7ns	Χ²₂₄= 18·2ns

*Analysis of variance, model: NumberOfAuthors = Year + Region + Interaction, with Year and Region as fixed effects and each paper treated as a single data point (SAS Proc Mixed).
†Logistic regression, model: NumberOfFemaleAuthors/TotalNumberOfAuthors = Year + Region + Interaction, with Year and Region as fixed effects and each paper treated as a single data point (SAS Proc Logistic).
‡Logistic regression, model: GenderOfAuthor(f/m) = Year + Region + Interaction, with Year and Region as fixed effects and each paper treated as a single data point (SAS Proc Logistic).
§ns = P > 0·05; * P < 0·05; ** P < 0·01; *** P < 0·001.
¶Sample sizes for single-authored papers are too small for meaningful statistical analysis (Year-x-Geography cells); Africa (n = 3 papers across all years), Asia (n = 21); Europe (n = 30); Latin America (n = 3); North America (n = 44); Oceania (n = 11); United Kingdom (n = 9).

Women occupy a larger proportion of first author positions on papers if the last author is female (Fig. 2a; Χ²₁ = 4·54, P = 0·03; considering only papers with two or more authors). This appears to be because a higher proportion of all authors on a paper are women (the gender ratio of non-last authors) when the last author is female (Fig. 2b; Χ²₁ = 16·5, P < 0·001). In a logistic regression including both the senior author gender and author gender ratio (FirstAuthorGender = Year + SeniorAuthorGender + AuthorgenderRatio[excluding senior author] + Interactions), the gender of the first author of the paper is predicted by the gender ratio of all non-last authors (Χ²₁ = 627·3, P < 0·001) but not by the gender of the last author (Χ²₁ = 0·0, P = 0·98). This analysis suggests that women are more often first authors on papers with female last authors because there are more female authors (across all positions) on papers that have female last authors.

If the first author was female, she was less likely to be listed as the corresponding author on the cover page (Fig. 3a; Χ²₁ = 18·0, P < 0·001) and less likely to be the submitting author (the author who actually submitted the manuscript in ScholarOne; Χ²₁ = 33·1, P < 0·001) than if the first author was male. Neither of these differences were influenced by the gender of the last author (last author gender and last author-x-first author gender interactions were all non-significant at P > 0·25 for each). The probability that the first author was also the corresponding and/or submitting author also varied across geographic regions (Fig. 3b; corresponding author: Χ²₆ = 264·7, P < 0·001, submitting author: Χ²₆ = 132·6, P < 0·001); most notably, first authors from Asia were least likely to serve as corresponding authors of papers. Importantly, there was no evidence of an author gender-x-geographic region interaction (corresponding author: Χ²₆ = 4·42, P = 0·62, submitting author: Χ²₆ = 8·34, P = 0·21); i.e., female authors are similarly less likely to serve as corresponding or submitting authors than are men across all geographic regions (the apparent switch in this pattern for authors from Africa is accompanied by large errors due to small sample sizes; Fig. 3b).

There was no evidence that the number of authors varied with the gender of the first, corresponding or last author (P > 0·27 for each), or with the gender ratio of all authors on the paper (F_4,3514 = 0·31, P = 0·58). The number of authors on papers increased over the five years, from an average of 3·7 ± 0·07 in 2010 to 4·4 ± 0·09 in 2014 (2011: 4·0 ± 0·07, 2012: 4·0 ± 0·07, 2013: 4·2 ± 0·12)(Year effect in univariate model: F_4,3519 = 5·84, P < 0·001). The average number of authors varied across geographic regions - it was highest in Europe, the United Kingdom and Africa (mostly South Africa) and lowest in North America and Asia (Table 1).

The Relationship between Author Gender and Outcomes of the Editorial and Peer Review Process

All papers submitted to the journal are first evaluated by one of our Senior Editors and only a subset of these are assigned to a handling editor. We found no evidence that gender of the first, corresponding or last author influenced the probability that a paper was sent to a handling editor (model: SentToHandlingEditor (yes/no) = Year + AuthorGender + Interaction, with SeniorEditor as a random effect; P > 0·14 for all author gender and interaction effects; note that all senior editors are male so SeniorEditorGender is not included in the analysis) (Fig. 4). There was a suggestion in the data that papers with a higher proportion of female authors were slightly less likely to be sent to a handling editor (AuthorGenderRatio: Χ²₁ = 5·9, P = 0·02; Year-x-AuthorGenderRatio Interaction, Χ²₄ = 4·39, P = 0·36). However, the effect size was quite small (one percentage point for every 20% increase in the proportion of women on a paper) and becomes non-significant when correcting for multiple comparisons or if the non-significant interaction is removed from the model. Also, the best fit model (that with the lowest AIC) includes only Year and not AuthorGenderRatio.

After a manuscript has been assigned to a handling editor, that editor can decide whether to send the paper out for review or decline it without review. The probability that the handling editor sent a paper out for review varied significantly among individual editors, but there was no evidence that the gender of the first, corresponding or senior author, or the gender ratio of all authors on the manuscript, affected this probability (Fig. 4) (model: SentForReviewByHandlingEditor (yes/no) = Year + AuthorGender + Interaction, with SeniorEditor as a random effect; P > 0·14 for all author gender and interaction effects).

Manuscripts sent for review were not, on average across all papers, scored differently depending on the gender of the first, corresponding or last author or the gender ratio of all authors (analysis of variance, model: ReviewScore = Year + AuthorGender + Interaction; statistics in legend of Fig. 5). Following review, the review score was a highly significant predictor of whether a paper was rejected vs. invited for revision (P < 0·001 in all analyses) but there was no evidence that any of the author gender variables affected the handling editor's decision (Fig. 4) (including review score as a covariate in the model; RevisionInvitedIfReviewed (yes/no) = Year + AuthorGender + Interaction, with HandlingEditor as a random effect; P > 0·27 for all author gender and interaction effects when ReviewScore; excluding review score from the analysis, P > 0·16 for all author gender and interaction effects). Cumulative across all stages of the editorial process, we found no evidence that gender of the authors (first, last or corresponding) or author gender ratio on a paper (across all author positions) affected the probability that the paper is declined from the journal (Fig. 4) (Model: RevisionInvited(yes/no) = Year + AuthorGender + Interaction, with HandlingEditor as a random effect; AuthorGender: P > 0·11 for each; AuthorGender-x-Year interaction: P > 0·27 for each).

Interestingly, papers with more authors were less likely to be rejected at the senior editor review stage, and thus more likely to be sent to handling editors; each additional author increased the probability of success at the senior editor stage by 1·96 ± 0·87 percentage points (averaged across years) (Χ²₁ = 16·9, P < 0·001; this analysis deletes one manuscript with 87 authors; the next highest was 20 authors). However, there was no evidence that the number of authors was predictive of the success of a paper at any subsequent stage of the editorial process [paper sent for review by the handling editor (Χ²₁ = 0·85, P = 0·36), the peer review score (F_1,1505 = 0·32, P = 0·57) or the probability a paper is rejected after peer review (Χ²₁= 1·06, P = 0·30)] and no evidence that number of authors affected the outcome cumulative across all parts of the editorial process (Χ²₁ = 1·19, P = 0·28).

Interactions between Handling Editor Gender and Author Gender

In the previous section, we included handling editor in our analyses (as a random effect) to control for among-editor variation in decisions. The handling editor effect was always statistically significant. We thus explored whether the gender of the handling editor influenced whether papers authored by male vs. female authors have different fates.

We found no evidence that female vs. male editors were assigned a greater proportion of papers with more female authors (P > 0·18 for all author gender variables). We also found no evidence of an interaction between editor and author gender on how likely editors were to send papers out for review (model: SentForReviewByHandlingEditor [yes/no] = Year + HandlingEditorGender + AuthorGender + AuthorGender-x-EditorGender Interactions, with HandlingEditor[nested within HandlingEditorGender] included as a random effect; P > 0·15 for all AuthorGender effects; P > 0·27 for all interactions). Lastly, there was no evidence of an interaction between editor and author gender on how likely editors were to request revisions of a manuscript after peer review (RevisionInvitedIfReviewed [yes/no] = Year + HandlingEditorGender + AuthorGender + AuthorGender-x-EditorGender Interactions, with HandlingEditor[nested within HandlingEditorGender and ReviewScore included as a covariate; P > 0·23 for all AuthorGender effects; P > 0·19 for all interactions).

Interactions between Reviewer and Author Gender

Over the 4·5 years for which we have data, 351 papers were reviewed by both one male and one female reviewer. The mean scores given to papers, averaged across all papers, did not differ between male and female reviewers in any year (Wilcoxon signed-rank test, P > 0·54 for each year). The score given by women reviewers on papers was positively correlated (across manuscripts) with the score given by male reviewers (analysis of covariance, ReviewScore_Female = Year + ReviewScore_Male, + Interaction; covariate: F_1,341 = 9·92, P = 0·002; Interaction: F_4,341 = 0·13, P = 0·97). Notably, this relationship did not differ depending on the gender of the authors (i.e., there were no significant author gender or author gender-x-male reviewer score interactions in the model FemaleReviewerScore = Year + MaleReviewerScore + AuthorGender + Interactions; AuthorGender: F_1,342 < 1·77, P > 0·18, Interaction: F_1,342 < 1·11, P > 0·28 for all author positions).

As an alternative analysis, we tested whether the difference between scores given by male reviewers and female reviewers (male reviewer score - female reviewer score) differed depending on whether the authors of a paper were male or female. If men gave generally more negative or women generally more positive review scores, or vice versa, to papers authored by women, we expect that the difference in review scores between male and female reviewers to differ for male vs. female authors. We found no such difference; the average difference in reviewer score given by male and female reviewers did not depend on the gender of the first author (F_1,344 = 1·11, P = 0·29), the corresponding author (F_1,344= 0·19, P = 0·66), the senior author (F_1,344= 0·29, P = 0·59), or the author gender ratio (F_1,344= 1·05, P = 0·41).

Editor Seniority

In a previous analysis (Fox, Burns & Meyer 2015), we found that senior editors differed from younger editors in the proportion of women they selected to review (EditorSeniority = years since PhD at the time of handling a manuscript). We thus examined whether editor seniority differentially affected the decisions editors made on papers authored by males vs. females. We found no evidence that more senior editors were more or less likely to send papers for review if the first or last/senior author was female (model: PaperReviewed = Year + CorrAuthorGender + EditorSeniority + AuthorGender-x-HandlingEditorSeniority interaction with HandlingEditor as a random effect; non-significant interaction, P > 0·64, for first and last author gender, and for author gender ratio). The only suggestion of an editor seniority effect was a significant interaction between the gender of the corresponding author and the seniority of the handling editor (CorAuthorGender-x-HandlingEditorSeniority) influencing the likelihood that a paper is sent for peer review (Χ²₁ = 6·2, P = 0·01). However, though this interaction was significant when included in the model, removing it from the model reduces model AIC. More importantly, this interaction between corresponding author gender and editor seniority (if real) did not carry through to the final decisions on papers; we found no evidence that more senior editors reached different decisions after review, or cumulative through the entire process, for papers authored by men vs. women (AuthorGender-x-HandlingEditorSeniority interaction: P > 0·47 for all author gender variables).

The Relationship between Author Gender and other Aspects of the Peer Review Process

In previous analyses (Fox, Burns & Meyer 2015) we found that editor gender affected responses to review invitations (the Associate Editor handling a manuscript is identified in peer review invitations); invitees to review for Functional Ecology were less likely to respond to review invitations, and less likely to agree to review, if the handling editor for a paper was female. This suggests that reviewers were influenced, albeit only slightly, by the gender of participants in the editorial process. Here, we examine whether author gender likewise influences reviewer responses to review invitations.

We found no evidence that the gender of the first or corresponding author influenced how likely invitees responded to review invitations, how quickly they responded to review invitations, whether they agreed to review (if they responded), whether they returned their review (if they agreed to review) or how quickly they returned their review (if they agreed) (model: ReviewVariable = Year + AuthorGender + Interaction; P > 0·29 for AuthorGender and the Interaction for all response variables). There was a significant effect of senior author gender on the probability that invitees agreed to review; invitees were more likely to agree to review (if they responded to the review invitation) if the last author was female rather than male (Χ²₁ = 5·27, P = 0·02). However, the effect size was very small (73·5 ± 0·7% of reviewers that replied to review invitations agreed to review if the last author was female, averaged across years, compared to 70·0% if the last author was male) and there was no evidence that last author gender influenced any other aspect of the peer review process (P > 0·16 for how likely invitees responded to review invitations, how quickly they responded to review invitations, whether they returned their review, and how quickly they returned their review).

Interestingly, the gender ratio of the reviewers that agreed to revie w differed for papers authored by female vs. male authors (P < 0·03 for corresponding author gender, last author gender, and author gender ratio, but P = 0·15 for first author gender). Specifically, reviews were more likely to be written by women if the paper had a female corresponding or senior author (proportion of reviewers that are women when corresponding author is female: 29·5 ± 1·4%, male: 24·9 ± 1·4%; proportion when the last author is female: 30·0 ± 1·3%, male: 25·7 ± 1·1%). This suggests that either (i) editors are inviting more women as prospective reviewers if the authors on a paper are female, or (ii) male and female reviewers are responding differently to invitations to review if the authors are male vs. female. We thus tested these two hypotheses.

Hypothesis A: We found that editors do invite more women to review if the first, corresponding or last author is female and if the gender ratio of authors on the manuscript is more female (Fig. 6; statistics in figure legends). Though we also found that female editors invited more female reviewers than did male editors (averaged across all papers; as previously reported by Fox, Burns & Meyer 2015; P < 0·001 for all models), this editor gender effect did not explain the result that more women were invited to review papers with female authors; including handling editor gender in our model did not account for the AuthorGender effect described above (AuthorGender remained significant in all analyses), and there was no significant AuthorGender-x-EditorGender interaction (P > 0·27 for all author gender metrics).

Hypothesis B: In a previous study, Fox, Burns & Meyer (2015) found that men invited to review were less likely to respond to review invitations, and less likely to agree to review, if the handling editor for the manuscript was female. Women invited to review showed no such difference in response depending on the gender of the handling editor. We thus analysed our author data separately for male and female invitees. We found no evidence that women that were invited to review were more likely to respond to review invitations (P > 0·28) or agree to review if they responded (P > 0·21) if the author was male or female. In contrast, men who were invited to review were less likely to respond to review invitations if the authors on a paper were female (Χ²₁ = 16·2, P < 0·001 for author gender ratio; P < 0·03 for first and corresponding author gender; P = 0·56 for senior author gender). However, the effect size is very small (e.g., averaging only two percentage points difference between papers corresponded by women vs. men).This effect also appears to be largely, but not entirely, due to the editor gender effect described above (men respond to review invitations less often if the handling editor is female); when we include editor gender in the statistical model (and editor identity nested within editor gender as a random effect), the only author gender effect that remains marginally significant is author gender ratio, and the proportion variance explained by author gender drops substantially (Χ²₁ = 4·75, P = 0·03). Of men that responded to review invitations, invitees were not less likely to agree to review (if they responded) if the authors were female (P > 0·10 for all author gender variables).

Discussion

In this study, we examined whether patterns of authorship on papers submitted to Functional Ecology differed between male and female authors, and tested whether the gender of authors influences the peer review process or peer review outcomes. We find that women were less likely to be sole or last author, but more likely to be first author, relative to their overall frequency of authorship. Women were also less likely to serve as corresponding author of their papers than were men when they were first author. Despite these clear differences in pattern of authorship, we found no compelling evidence of gender discrimination in the peer review process – papers with female authors were equally successful at all stages of the editorial and peer review process.

Patterns of Authorship

Women occupied about a third of all authorships, averaged across all positions, on papers submitted to Functional Ecology. Relative to this, they were over-represented as first authors (43%) but underrepresented as last authors (25%) and underrepresented on single-authored papers (26%). That women are overrepresented as first authors (relative to authorship across all positions) has been observed in some disciplines (e.g., van der Weijden & Medina 2014), whereas women are underrepresented as first authors in other disciplines (van Arensbergen, van der Weijden & van den Besselaar 2012); this variation is likely in part due to variation in conventions on authorship order across fields (van Arensbergen, van der Weijden & van den Besselaar 2012; West et al. 2013). Generally, though, the representation of women as first authors has been increasing (West et al. 2013).

Women are better represented as first authors on submissions to Functional Ecology than in the global scientific literature (Larivière et al. 2013) and better represented in submissions to Functional Ecology during the time period of our study than they were in the broader literature in ecology and evolution between 1990 and 2011 (women represent ~24% of first authors in ecology and evolution for papers archived in JSTOR; West et al. 2013; http://www.eigenfactor.org/gender/#). That women are more commonly first author than last author is a typical result for analyses of authorship in biology and medical journals (e.g., Jagsi et al. 2006; Feramisco et al. 2009; Kongkiatkamon et al. 2010; Dotson 2011; West et al. 2013; Erren et al. 2014). This is likely due to demographic differences between individuals who occupy the first vs. last author positions on papers. First authors on papers are commonly students and post-docs, populations for which female representation is quite high in the sciences (Jagsi et al. 2006). In contrast, last author is commonly (though not always) the senior scientist for the project; e.g., the lab supervising professor or grant primary investigator, populations for which women remain underrepresented (Jagsi et al. 2006).

Women are more likely to co-author papers with other women, and men with other men, compared to expectations if the genders associate randomly (Shah et al. 2013; Long et al. 2015). We observed this at Functional Ecology; in particular, we see that papers with female last authors include a higher proportion of women authors at other positions, including first author, than do papers with male last authors. This likely reflects two phenomena. First, women tend to (on average) have slightly different research interests than do men, both in ecology (Bonnet, Shine & Lourdais 2004; West et al. 2013; details at www.eigenfactor.org/gender/#) and in other fields (Lipetz 1999; Rigg, McCarragher & Krmenec 2011; Long et al. 2015). This leads to a non-random association of women across sub-disciplines. Second, women students may be more likely to enter graduate school under female mentors (Tenenbaum, Crosby & Gliner 2001). This association of female students with female mentors may be driven by commonality of research interests (as noted above) but may also reflect a preference by women for female mentors (Blake-Beard et al. 2011). Either mechanism would generate a non-random association between gender of the senior author and gender of the other authors. Importantly, although we do find a non-random association between senior author gender and the gender of other authors, we find no evidence of bias specifically against female first-authorship when the senior author is male; instead, the proportion of female authors is lower across all positions when the senior author is male, with this reduced the proportion of women authors across all positions leading to a lower proportion of female first authors on papers with male senior authors.

Women first authors were less likely to serve as the corresponding author or submitting author of their paper than were men first authors for papers submitted to Functional Ecology. A similar difference between men and women in the frequency of corresponding authorship was observed for a European medical journal (Heckenberg & Druml 2010). Women tend to leave science at a higher rate than do men (Fox 2008), and we suspect that the ‘missing’ female corresponding authors are women who left science, or at least left academic research, following graduate school or their post-doctoral research, and that the corresponding authors on those papers are primarily their graduate or post-doctoral advisors. Alternatively, women may expect to face a gender-biased editorial process and so defer corresponding authorship to their collaborators more frequently than men. However, our data do not support this latter hypothesis; we saw no evidence that women were less likely to serve as corresponding author if the senior author was male. Instead, women may more frequently defer corresponding authorship to their coauthors for other reasons: they may more frequently (i) lack the confidence necessary to serve as corresponding author (or have advisors that lack confidence in them), (ii) be less assertive than men in taking charge of the manuscript submission process (e.g., in competition with strong-willed coauthors or advisors), or (iii) have less time available (such as due to non-work responsibilities, or additional non-research responsibilities at work) than do male first authors.

The number of authors on papers submitted to Functional Ecology increased noticeably in just the 5 years of this study. Similar increases have been observed for a variety of academic fields (Wren et al. 2007; Larivière et al. 2013). Research performed by larger research teams, producing papers with more authors, tends to be more impactful (or at least more highly cited) than research done by individuals or small groups (Wuchty, Jones & Uzzi 2007), even after controlling for self-citations (Larivière et al. 2014). The increasing number of authors on papers may reflect the increasingly interdisciplinary and collaborative nature of research (Larivière et al. 2014). Alternatively, it may reflect an increase in the inclusion of contributors with minor roles in the project as authors (Danell 2014) possibly in response to incentives related to impact, productivity measures or career advancement. As the number of authors increases, ambiguity over author contributions, and how authorship order reflects these contributions, necessarily increases (Zuckerman 1968). It is commonly recognized in the biological sciences that first and last author are the key positions, but the roles of middle authors, and even the last author, can vary substantially among papers (Wren et al. 2007). Of note for our analysis is that the number of authors on papers was not correlated with gender of the authors. This contrasts with an observation by Shah et al. (2013) who found that papers with female first authors (but not female last authors) tended to have more authors, and that of Bozeman & Gaughan (2011) who found that, when controlling for other variables, women tend to have more collaborators than do men.

Does Author Gender Influence Peer Review?

For Functional Ecology we found little evidence that author gender influenced outcomes at any stage of the editorial and peer review process, and no evidence that female reviewers responded to author gender differently than did male reviewers. Papers with female authors were equally likely to be sent for peer review, obtained equivalent peer review scores (regardless of reviewer gender), and were equally likely to be accepted for publication, compared to papers with male authors. These results are consistent with the majority of studies that tested for evidence of systemic biases in peer review based on author gender for journal manuscript submissions. A couple of early studies of journal peer review presented evidence that papers authored by women were more harshly reviewed or less likely to be accepted (controlling for review scores; Petty, Fleming & Fabrigar 1999), and that female reviewers responded differently to author gender than did male reviewers (e.g., Lloyd 1990). However, subsequent studies generally find no evidence of differences in peer review scores or editorial outcomes for papers authored by men vs. women (Tregenza 2002; Primack et al. 2009; Lane & Linden 2009; Heckenberg & Druml 2010; Buckley et al. 2014; but see Walker et al. 2015 2015), and no author gender-x-editor gender interactions (Lane & Linden 2009; Walker et al. 2015 2015). The difference in results between early studies and more recent studies may be reflect the increased number of women in science and better social attitudes about women in science (Walker et al. 2015 2015). Controlled manipulative studies, in which author identity is manipulated on text provided to a study population, have more mixed results. Most find that text authored by men and women is rated the same (Blank 1991; Borsuk et al. 2009), and that there is no evidence of same-gender bias (Borsuk et al. 2009; Knobloch-Westerwick, Glynn & Huge 2013; Krawczyk & Smyk 2014), but a few find evidence that works authored by women are rated lower. For example, students rated conference abstracts with male authors to be of greater scientific quality (Knobloch-Westerwick, Glynn & Huge 2013), and rated economics papers authored by men as more publishable (Krawczyk & Smyk 2014; though economics as a field may be more male-biased than the biological sciences, van Arensbergen, van der Weijden & van den Besselaar 2012). Overall, though gender biases are clearly evident in many aspects of academic research and careers (e.g., recruitment of editors, symposium participation invitations, evaluation of academic job applicants; Moss-Racusin et al. 2012; Fox, Burns & Meyer 2015 and citations therein), and there are occasional instances where gender discrimination by specific editors or reviewers influences scientific publishing, on balance most evidence indicates that the peer review process underlying scientific publication is gender-neutral, or close enough to gender-neutral for biases to be difficult to detect.

Our analysis tests for gender differences in manuscript outcomes throughout the entire editorial and peer review process. It thus avoids the problem of bibliometric analyses (analysis of gender distributions on published papers) that confounds actual gender discrimination with author behaviour. For example, comparing author gender distributions before vs. after adoption of double-blind peer review (Budden et al. 2008) requires consideration of changing author behaviour over time, including the changing proportion of women publishing in the field (Webb, O'Hara & Freckleton 2008) and changes in behaviour specifically stimulated by changes in peer review models (e.g., women who perceive that there is gender discrimination may preferentially submit papers to journals with double-blind peer review). However, our study is entirely correlational with no manipulative elements. We thus do not (cannot) control for variation in manuscript quality or other variables that might differ between male- and female-authored papers and that could obscure biases that might be present. If, for example, papers authored by women are generally higher quality than papers authored by men (Symonds et al. 2006), the absence of a difference in peer review scores or editorial outcomes would demonstrate a discriminatory rather than gender-neutral process.

We found a non-random association between author gender and reviewer gender - reviews obtained by Functional Ecology were more likely to be written by women if the authors on the paper were women. This was primarily because editors invited more women to review if the authors were female, though there was also a small influence of author gender on whether male invitees respond to review invitations. The former effect makes sense given the known gender variation among research areas within ecology (Bonnet, Shine & Lourdais 2004; West et al. 2013); given that reviewers are chosen to have expertise overlapping with the topic of the manuscript, and that the proportion of women varies among subfields, we expect some gender association between authors and reviewers even if reviewers are chosen at random (regarding gender) from the pool of people with appropriate expertise for each paper. Testing whether there is a non-random relationship between author and reviewer gender beyond that caused by gender structuring in the subfields of ecology requires that papers be categorized by subject area compared to actual author gender ratios for each subfield, a project well beyond the scope of this paper. The second effect - that men were less likely to respond to review invitations if the authors were female - is a concerning result, suggestive of real gender discrimination. This effect is similar to one observed previously for editor gender - men (but not women) invited to review for Functional Ecology were less likely to respond to review invitations, and less likely to agree to review, if the editor was female (Fox, Burns & Meyer 2015; see also Gilbert, Williams & Lundberg 1994). As in the current study, the effects in Fox, Burns & Meyer (2015) were small (just a couple percentage points), but present most years and in the direction predicted if there is gender discrimination (albeit very slight) against women authors (this study) or editors (Fox, Burns & Meyer 2015).

Conclusions

Women and men differ in their patterns of authorship on papers submitted to Functional Ecology. Women represent about a third of all authors (across all positions), but are overrepresented (relative to overall authorship) as first authors and underrepresented as last and solo authors. Notably, female first authors were less likely than male first authors to serve as the submitting or corresponding author of their papers, possibly reflecting well-documented gender differences in attrition from scientific careers. The overrepresentation of women as first authors is a positive sign that women are well-represented at the less senior levels in ecology; however, for this to translate into growth in the proportion of women at more senior positions, the scientific community needs to reduce the rate of attrition of women from science so that they play more dominant roles in the research process.

Though authorship patterns differ between men and women, we found little evidence that author gender influenced peer review scores or editorial decisions. The few analyses that suggest gender differences in editorial decisions are at most marginally statistically significant (and thus non-significant after controlling for multiple comparisons) and of very small effect size. Our evidence here, and in Fox, Burns & Meyer (2015), is consistent in indicating that peer review outcomes are gender-neutral, but that gender does influence some aspects of the peer review process. For example, Fox, Burns & Meyer (2015) found that editor and reviewer gender influence reviewer recruitment and the time span of the review process, although the influences were generally small. We thus conclude from our pair of studies that gender of the participants influences the process of manuscript review - it influences who is chosen to review papers, the response of invitees to review invitations, and the time span of the review process. But, at the end of the process, gender of the participants does not affect editorial and peer review outcomes at this particular ecology journal.

Acknowledgements

We thank the British Ecological Society (BES), owners of the journal Functional Ecology, for permitting us to use their peer review database for this project. We thank Liz Baker, the previous Managing Editor of Functional Ecology, for discussions about peer review and for her enthusiastic support for this project. Jason Wolf (at the University of Bath) provided office space and statistical advice (to CWF) during the analysis of these data. Alison Bennett, Katie Field, Julia Koricheva, Emma Sayer, and members of the Fox laboratory at the University of Kentucky (Joseph Deas and William Licht) provided comments on an earlier draft of this paper. This work was reviewed and approved by the Internal Review Board at the University of Kentucky, IRB 14-0570-P4S. This research was funded in part by the Kentucky Agricultural Experiment Station.

Data accessibility

Data for this project are deposited in the Dryad Digital Repository (doi: 10.5061/dryad.0mv3q). However, because the data set contains information on human subjects, the data available at Dryad are anonymized and lacking variables that allow parts of the data set to be de-anonymized. Please see the metadata accompanying the Dryad submission for additional details.

Supporting Information

References

Abramo, G., D'Angelo, C. & Caprasecca, A. (2009) Gender differences in research productivity: a bibliometric analysis of the italian academic system. Scientometrics, 79, 517–539.
van Arensbergen, P., van der Weijden, I. & van den Besselaar, P. (2012) Gender differences in scientific productivity: a persisting phenomenon? Scientometrics, 93, 857–868.
Baethge, C. (2008) First authors in Deutsches Ärzteblatt: women are catching up. Deutsches Ärzteblatt International, 105, 507–9.
Barrios, M., Villarroya, A. & Borrego, Ã. (2013) Scientific production in psychology: a gender analysis. Scientometrics, 95, 15–23.
Bhandari, M., Guyatt, G.H., Kulkarni, A.V., Devereaux, P.J., Leece, P., Bajammal, S. et al. (2014) Perceptions of authors' contributions are influenced by both byline order and designation of corresponding author. Journal of Clinical Epidemiology, 67, 1049–1054.
Blake-Beard, S., Bayne, M.L., Crosby, F.J. & Muller, C.B. (2011) Matching by race and gender in mentoring relationships: keeping our eyes on the prize. Journal of Social Issues, 67, 622–643.
Blank, R.M. (1991) The effects of double-blind vs. single-blind reviewing: experimental evidence from the American Economic Review. The American Economic Review, 81, 1041–1067.
Bonnet, X., Shine, R. & Lourdais, O. (2004) Does gender affect a scientist's research output in evolutionary ecology? Journal of Women and Minorities in Science and Engineering, 10, 353–360.
Bornmann, L., Mutz, R. & Daniel, H.D. (2007) Gender differences in grant peer review: a meta-analysis. Journal of Informetrics, 1, 226–238.
Borsuk, R.M., Aarssen, L.W., Budden, A.E., Koricheva, J., Leimu, R., Tregenza, T. et al. (2009) To name or not to name: the effect of changing author gender on peer review. BioScience, 59, 985–989.
Bozeman, B. & Gaughan, M. (2011) How do men and women differ in research collaborations? An analysis of the collaborative motives and strategies of academic researchers. Research Policy, 40, 1393–1402.
Buckley, H.L., Sciligo, A.R., Adair, K.L., Case, B.S. & Monks, J.M. (2014) Is there gender bias in reviewer selection and publication success rates for the ‘New Zealand Journal of Ecology’? New Zealand Journal of Ecology, 38, 335–339.
Budden, A., Tregenza, T., Aarssen, L., Koricheva, J., Leimu, R. & Lortie, C. (2008) Double-blind review favours increased representation of female authors. Trends in Ecology and Evolution, 23, 4–6.
Burns, C.S. (2015) Characteristics of a megajournal: a bibliometric case study. Journal of Information Science Theory and Practice, 3, 16–30.
Cleary, M., Jackson, D., Walter, G., Watson, R. & Hunt, G.E. (2012) Editorial: location, location, location - the position of authors in scholarly publishing. Journal of Clinical Nursing, 21, 809–811.
Danell, R. (2014) The contribution of authors: A study of the relationship between the size and composition of author teams. In Proceedings of the Science and Technology Indicators Conference 2014, 3–5 September, Leiden, the Netherlands (pp. 123–131).
Dickson, D. (1997) Female scientists wanted—apply to UK research councils. Nature, 390, 431–431.
Dotson, B. (2011) Women as authors in the pharmacy literature: 1989–2009. American Journal of Health-System Pharmacy, 68, 1736–1739.
Erren, T.C., Groß, J.V., Shaw, D.M. & Selle, B. (2014) Representation of women as authors, reviewers, editors in chief, and editorial board members at 6 general medical journals in 2010 and 2011. JAMA Internal Medicine, 174, 633–635.
Feramisco, J.D., Leitenberger, J.J., Redfern, S.I., Bian, A., Xie, X.J. & Resneck, J.S. (2009) A gender gap in the dermatology literature? Cross-sectional analysis of manuscript authorship trends in dermatology journals during 3 decades. Journal of the American Academy of Dermatology, 60, 63–69.
Fox, M.F. (2008) Institutional transformation and the advancement of women faculty: the case of academic science and engineering. Higher Education: Handbook of Theory and Research, 23, 73–103.
Fox, C.W. & Burns, C.S. (2015) The relationship between manuscript title structure and success: Editorial decisions and citation performance for an ecological journal. Ecology and Evolution, 5, 1970–1980.
Fox, C.W., Burns, C.S. & Meyer, J.A. (2015) Editor and reviewer gender influence the peer review process but not peer review outcomes at an ecology journal. Functional Ecology, 30, 136–149.
Frandsen, T.F. & Nicolaisen, J. (2010) What is in a name? Credit assignment practices in different disciplines. Journal of Informetrics, 4, 608–617.
Friesen, H.G. (1998) Equal opportunities in Canada. Nature, 391, 326–326.
Gilbert, J.R., Williams, E.S. & Lundberg, G.D. (1994) Is there gender bias in JAMA's peer review process? JAMA, 272, 139–142.
Grant, J., Burden, S. & Breen, G. (1997) No evidence of sexism in peer review. Nature, 390, 438–438.
Heckenberg, A. & Druml, C. (2010) Gender aspects in medical publication–the Wiener klinische Wochenschrift. Wiener Klinische Wochenschrift, 122, 141–145.
Hunter, M. (2010) The great experiment: the Royal Society. History Today, 60, 34–40.
Jagsi, R., Guancial, E.A., Worobey, C.C., Henault, L.E., Chang, Y., Starr, R. et al. (2006) The ‘gender gap’ in authorship of academic medical literature — a 35-Year perspective. The New England Journal of Medicine, 355, 281–287.
Kaltman, J.R., Evans, F.J., Danthi, N.S., Wu, C.O., DiMichele, D.M. & Lauer, M.S. (2014) Prior publication productivity, grant percentile ranking, and topic-normalized citation impact of NHLBI cardiovascular R01 grants. Circulation Research, 115, 617–624.
Knobloch-Westerwick, S., Glynn, C.J. & Huge, M. (2013) The Matilda Effect in science communication: an experiment on gender bias in publication quality perceptions and collaboration interest. Science Communication, 35, 603–625.
Kongkiatkamon, S., Yuan, J.C.C., Lee, D.J., Knoernschild, K.L., Campbell, S.D. & Sukotjo, C. (2010) Gender disparities in prosthodontics: authorship and leadership, 13 years of observation. Journal of Prosthodontics, 19, 565–570.
Krawczyk, M. & Smyk, M. (2014) Author's gender affects the rating of academic articles: Evidence from an Incentivized, deception-free laboratory experiment. University of Warsaw, Working Paper no. 7/2014 (124).
Lane, J.A. & Linden, D.J. (2009) Is there gender bias in the peer review process at Journal of Neurophysiology? Journal of Neurophysiology, 101, 2195–2196.
Larivière, V., Vignola-Gagné, E., Villeneuve, C., Gélinas, P. & Gingras, Y. (2011) Sex differences in research funding, productivity and impact: an analysis of québec university professors. Scientometrics, 87, 483–498.
Larivière, V., Ni, G., Gingras, Y., Cronin, B. & Sugimoto, C. (2013) Global gender disparities in science. Nature, 504, 211–213.
Larivière, V., Gingras, Y., Sugimoto, C.R. & Tsou, A. (2014) Team size matters: collaboration and scientific impact since 1900. Journal of the Association for Information Science and Technology, 66, 1323–1332.
Leahey, E. (2006) Gender differences in productivity: research specialization as a missing link. Gender and Society, 20, 754–780.
Ledin, A., Bornmann, L., Gannon, F. & Wallon, G. (2007) A persistent problem. Traditional gender roles hold back female scientists. EMBO Reports, 8, 982–987.
Lee, C.J., Sugimoto, C.R., Zhang, G. & Cronin, B. (2013) Bias in peer review. Journal of the American Society for Information Science and Technology, 64, 2–17.
Leemann, R.J. & Stutz, H. (2008) Gender and research funding (GEFO). Final report and synthesis, Swiss National Science Foundation. Retrieved from http://www.snf.ch/SiteCollectionDocuments/wom_ber_gefo_synthesis_report_e.pdf
Ley, T.J. & Hamilton, B.H. (2008) The gender gap in NIH grant applications. Science, 322, 1472–1474.
Lipetz, B.-A. (1999) Aspects of JASIS authorship through five decades. Journal of the American Society for Information Science, 50, 994–1003.
Lloyd, M.E. (1990) Gender factors in reviewer recommendations for manuscript publication. Journal of Applied Behavior Analysis, 23, 539–543.
Long, M.T., Leszczynski, A., Thompson, K.D., Wasan, S.K. & Calderwood, A.H. (2015) Female authorship in major academic gastroenterology journals: a look over 20 years. Gastrointestinal Endoscopy, 81, 1440–1447.e3.
Marsh, H.W., Jayasinghe, U.W. & Bond, N.W. (2008) Improving the peer-review process for grant applications: reliability, validity, bias, and generalizability. American Psychologist, 63, 160–168.
Marsh, H.W., Jayasinghe, U.W. & Bond, N.W. (2011) Gender differences in peer reviews of grant applications: a substantive-methodological synergy in support of the null hypothesis model. Journal of Informetrics, 5, 167–180.
Marsh, H.W., Bornmann, L., Mutz, R., Daniel, H.D. & O'Mara, A. (2009) Gender effects in the peer reviews of grant proposals: a comprehensive meta-analysis comparing traditional and multilevel approaches. Review of Educational Research, 79, 1290–1326.
Martin, L.J. (2012) Where are the women in ecology? Frontiers in Ecology and the Environment, 10, 177–178.
Mauleón, E., Hillán, L., Moreno, L., Gómez, I. & Bordons, M. (2013) Assessing gender balance among journal authors and editorial board members. Scientometrics, 95, 87–114.
Merton, R.K. (1973) The normative structure of science. The Sociology of Science: Theoretical and Empirical Investigations (ed. R.K. Merton), pp. 267–278. University of Chicago Press, Chicago, IL, USA. (Original work published 1942).
Moss-Racusin, C.A., Dovidio, J.F., Brescoll, V.L., Graham, M.J. & Handelsman, J. (2012) Science faculty's subtle gender biases favor male students. Proceedings of the National Academy of Sciences, 109, 16474–16479.
Mutz, R., Bornmann, L. & Daniel, H.D. (2015) Does gender matter in grant peer review? Zeitschrift für Psychologie, 220, 121–129.
Nakhaie, M.R. (2002) Gender differences in publication among university professors in Canada*. Canadian Review of Sociology/Revue Canadienne de Sociologie, 39, 151–179.
Oertelt-Prigione, S. (2012) Sex and gender in medical literature. Sex and Gender Aspects in Clinical Medicine (eds S. Oertelt-Prigione & V. Regitz-Zagrosek) pp. 9–15. Springer, London, UK.
Petty, R.E., Fleming, M.A. & Fabrigar, L.R. (1999) The review process at PSPB: correlates of interreviewer agreement and manuscript acceptance. Personality and Social Psychology Bulletin, 25, 188–203.
Primack, R.B., Ellwood, E., Miller-Rushing, A.J., Marrs, R. & Mulligan, A. (2009) Do gender, nationality, or academic age affect review decisions? An analysis of submissions to the journal Biological Conservation. Biological Conservation, 142, 2415–2418.
Rigg, L.S., McCarragher, S. & Krmenec, A. (2011) Authorship, collaboration, and gender: fifteen years of publication productivity in selected geography journals. The Professional Geographer, 64, 491–502.
Sandström, U. & Hällsten, M. (2008) Persistent nepotism in peer-review. Scientometrics, 74, 175–189.
Shah, D.N., Huang, J., Ying, G.S., Pietrobon, R. & O'Brien, J.M. (2013) Trends in female representation in published ophthalmology literature, 2000–2009. Digital Journal of Ophthalmology, 19, 50.
Shapiro, D.W., Wenger, N.S. & Shapiro, M.F. (1994) The contributions of authors to multiauthored biomedical research papers. JAMA, 271, 438–442.
Sidhu, R., Rajashekhar, P., Lavin, V.L., Parry, J., Attwood, J., Holdcroft, A. et al. (2009) The gender imbalance in academic medicine: a study of female authorship in the United Kingdom. Journal of the Royal Society of Medicine, 102, 337–342.
Sonnenwald, D.H. (2007) Scientific collaboration. Annual Review of Information Science and Technology, 41, 643–681.
Spier, R. (2002) The history of the peer-review process. Trends in Biotechnology, 20, 357–358.
Symonds, M.R., Gemmell, N.J., Braisher, T.L., Gorringe, K.L. & Elgar, M.A. (2006) Gender differences in publication output: towards an unbiased metric of research performance. PLoS ONE, 1, e127.
Tenenbaum, H.R., Crosby, F.J. & Gliner, M.D. (2001) Mentoring relationships in graduate school. Journal of Vocational Behavior, 59, 326–341.
Tregenza, T. (2002) Gender bias in the refereeing process? Trends in Ecology and Evolution, 17, 349–350.
Walker, R., Barros, B., Conejo, R., Neumann, K. & Telefont, M.l. (2015) Personal attributes of authors and reviewers, social bias and the outcomes of peer review: a case study. F1000Research 2015, 4, 21.
Waltman, L. (2012) An empirical analysis of the use of alphabetical authorship in scientific publishing. Journal of Informetrics, 6, 700–711.
Webb, T.J., O'Hara, B. & Freckleton, R.P. (2008) Does double-blind review benefit female authors? Heredity, 77, 282–291.
van der Weijden, I. & Medina, C.C. (2014) Gender, academic position and scientific publishing: A bibliometric analysis of the oeuvres of researchers. Proceedings of the Science and Technology Indicators Conference 2014 Leiden, 673.
Wenneras, C. & Wold, A. (1997) Nepotism and sexism in peer-review. Nature, 387, 341–343.
West, J.D., Jacquet, J., King, M.M., Correll, S.J. & Bergstrom, C.T. (2013) The role of gender in scholarly authorship. PLoS ONE, 8, e66212.
Wren, J.D., Kozak, K.Z., Johnson, K.R., Deakyne, S.J., Schilling, L.M. & Dellavalle, R.P. (2007) The write position. EMBO Reports, 8, 988–991.
Wuchty, S., Jones, B.F. & Uzzi, B. (2007) The increasing dominance of teams in production of knowledge. Science, 316, 1036–1039.
Xie, Y. & Shauman, K.A. (1998) Sex differences in research productivity: new evidence about an old puzzle. American Sociological Review, 63, 847–870.
Zbar, A. & Frank, E. (2011) Significance of authorship position: an open-ended international assessment. The American Journal of the Medical Sciences, 341, 106–109.
Zuckerman, H.A. (1968) Patterns of name ordering among authors of scientific papers: a study of social symbolism and its ambiguity. American Journal of Sociology, 74, 276–291.

Citing Literature

Volume30, Issue1

Special Feature: Mechanisms and Consequences of Facilitation in Plant Communities

January 2016

Pages 126-139

Gender differences in patterns of authorship do not affect peer review outcomes at an ecology journal

Summary

Introduction