INTRODUCTION
The COVID-19 pandemic may have exacerbated gender gaps in academia. Emerging data suggest that womenFootnote 1 have had disproportionate reductions in research time and productivity coupled with increased time spent on childcare and housework (Deryugina, Shurchkov, & Stearns, Reference Deryugina, Shurchkov and Stearns2021). Studies of scientific authorship during COVID-19 show that – despite higher overall rates of production during the early months of the COVID-19 pandemic (Cui, Ding, & Zhu, Reference Cui, Ding and Zhuin press) – the relative proportion of manuscript submissions (Cui et al., Reference Cui, Ding and Zhuin press) and publications (Muric, Lerman, & Ferrara, Reference Muric, Lerman and Ferrara2021) was lower for women than for men. These gender disparities are also evident in articles about COVID-19, on which women represent only one-third of authors (Lerchenmüller, et al., Reference Lerchenmüller, Schmallenbach, Jena and Lerchenmueller2021). The underrepresentation of women during this critical period could have longstanding negative effects on academic advancement in promotion systems based heavily on publications (Wren, et al., Reference Wren, Kozak, Johnson, Deakyne, Schilling and Dellavalle2007).
We do not know the extent to which the pandemic has affected the research productivity of women in clinical neuropsychology (CN), who represent 60% of the field (Sweet, Klipfel, Nelson, & Moberg, Reference Sweet, Klipfel, Nelson and Moberg2021). CN has a history of gender disparities in multiple domains, including scientific publishing (Rohling et al., Reference Rohling, Ready, Dhanani and Suhrin press). A recent study of 10,531 articles published between 1985 and 2019 in six CN journals revealed that women were underrepresented as authors in the 2000s, despite their growing prevalence in the field; for example, 43.3% of the authors listed in CN article bylines were women (Matchanova et al., Reference Matchanova, Avci, Babicz, Thompson, Johnson, Ke and Woodsin press), yet women comprised approximately 52% of clinical neuropsychologists during this period (Sweet, Lee, Guidotti Breting, & Benson, Reference Sweet, Lee, Guidotti Breting and Benson2018). The present study extends this work by examining whether the COVID-19 pandemic affected the representation of women as authors on initial submissions to four representative CN journals. In addition, we report the representation of women on articles about the pandemic (e.g., COVID-19, teleneuropsychology) published in all major CN journals. The latter analysis provides insights on the extent to which women were productive in novel, emergent research projects during the pandemic and expands upon the limited representation of journals in the primary study by including all major journals in the field.
METHOD
Study 1: Initial Submissions to CN Journals
Invitations to participate in this study were sent by SPW and KOY to the editors of nine major journals in CN (Sweet, Meyer, Nelson, & Moberg, Reference Sweet, Meyer, Nelson and Moberg2011) via electronic mail in April 2020. The editors of Archives of Clinical Neuropsychology (ACN), Journal of Clinical and Experimental Neuropsychology (JCEN), Neuropsychology (NP), and The Clinical Neuropsychologist (TCN) all agreed to participate by confirming their ability to secure the relevant permissions and data from their respective journal publishers. The remaining editors declined to participate, primarily due to limited resources or difficulties securing the relevant data and/or permissions. The institutional review board (IRB) at Wayne State University approved the coding and analysis of the data from JCEN, whereas IRBs at the University of Houston and University of Utah both independently determined that this retrospective, de-identified study was exempt. All aspects of the study were conducted in accordance with the Declaration of Helsinki.
Representatives from the four journals coded the probable genderFootnote 2 of each author on every consecutive initial submission received from March 15 through September 15 in the years 2019 and 2020. Each author was coded as either a man or woman for submissions to ACN, JCEN, and TCN based on the OpenGenderTracking Project database (Ros, Matias, & Hyland, Reference Ros, Matias and Hyland2013), which contains binary gender probabilities for over 90,000 first names that are derived from census data in the USA and UK. Where possible, authors whose names were not listed in the database or who were gender atypical were coded by verifying their gender on an institutional or personal website. This method has excellent interrater reliability (Matchanova et al., Reference Matchanova, Avci, Babicz, Thompson, Johnson, Ke and Woodsin press; Odic & Wojcik, Reference Odic and Wojcik2020). The gender coding for Neuropsychology was conducted by the publisher as part of a broader project that used https://genderize.io/, which provides reliable estimates of authors’ binary gender from their first name based on over 100 million data points collected from social media across over 200 countries.
The author byline coding from 1,070 records were sent in a deidentified file to the corresponding author (SPW). Fifty-two records were excluded because the lead or corresponding author names were not codable, which included submissions from Asia (65.4%), Europe/UK (17.4%), USA (7.7%), Central/South America (5.7%), Middle East (1.9%), and Canada (1.9%). The remaining 1,018 records were from primarily from author groups in the Europe/UK (31%) and U.S. (39%). The following variables were generated from the eligible records: (1) a continuous variable representing the number of women on the author byline (range = 0–27); (2) a continuous variable representing the percentage of women on the author byline (adjusted for any un-codable authors; range = 0–100%); (3) a binary variable indicating whether the lead author was a woman or a man; and (4) a binary variable indicating whether the corresponding author was a woman or a man.
Study 2: Pandemic-Related Articles Published in CN Journals
We identified all articles that were published in nine major neuropsychology journals between March 2020 and March 2021. Specifically, two authors reviewed the table of contents and advanced access sections of the four journals from Study 1, plus Applied Neuropsychology-Adult, Child Neuropsychology, Journal of the International Neuropsychological Society, Journal of Neuropsychology, and Neuropsychology Review. We selected papers that were related to the pandemic (e.g., COVID-19, teleneuropsychology) based on the contents in the title and abstract. We then used the methods described above for the OpenGenderTracking Project database to code the gender of all authors of every study in this sample. There were zero uncodable authors.
RESULTS
Study 1: Initial Submissions to CN Journals
Initial manuscript submissions to CN journals rose by 27%, from 448 submissions in 2019 to 570 submissions in 2020, (Pocock’s z = 3.9, p < .001; Pocock, Reference Pocock2006). The total submissions in 2019 and 2020 did not vary significantly by journal (X 2 (3, n = 1018) = 1.90, p = .595].
Results showed that 2,912 of the 5,304 authors across all articles (n = 1,018) were women (54.9%). The total number of authors who were women rose from 1,306 in 2019 to 1,606 in 2020, which is a significant increase of 23.0% (Pocock’s z = 5.6, p < .001). The total number of authors who were men rose from 1,061 in 2019 to 1,331 in 2020, which is also a significant increase of 25.4% (Pocock’s z = 5.5, p < .001). The increase in women authors from 2019 to 2020 did not differ from the increase in men authors during this timeframe (Pocock’s z = 0.34, p = .37).
Next, we conducted a multiple regression predicting the percentage of authors who were women on any given submission byline across all articles (n = 1,018). The dummy-coded categorical predictors were: (a) year (2019 or 2020); (b) journal; and (c) the interaction of year and journal (see Figure 1). The overall model was not significant [F (7, 1010) = 1.74, p = .095] and explained a very small amount of the variance (adjusted R 2 = .01). No main effects or interactions in this model were significant (all ps > .05).
We then conducted a logistic regression predicting the likelihood of women as lead authors, again using year, journal, and their interaction as predictors (see Figure 2a). The overall model was not significant [X 2 (7, 1010) = 5.40, p = .617], nor were any of the individual predictors in the model significant (ps > .05, odds ratios < 1.3). Figure 2b shows similar null findings in a parallel logistic regression conducted with corresponding author as the criterion [X 2 (7, 236) = 12.00, p = .102]. Although this analysis was conducted conservatively on the 244 records for which a different person was lead and corresponding author, the overall model also was not significant when the analysis was conducted in the full sample [X 2 (7, 1010) = 8.3, p = .310].
Study 1 post hoc analyses
Epochs. Given the evolving nature of COVID-19, we also conducted a post hoc analysis to check for differences between the early (March 15–June 15) versus later (June 16–September 15) part of 2020. We repeated the multiple and logistic regression analyses described above and added main effects and interaction terms for a binary epoch variable that was dummy coded for early (0) versus late (1). No main effects or interactions for epoch emerged as significant predictors of the percentage of women on the byline or the likelihood of submissions with women as first or corresponding authors (all ps > .05)
Sponsorship. We also conducted post hoc analyses to examine the role of sponsorship in manuscript submissions, given its importance for authorship among women in CN (Matchanova et al., Reference Matchanova, Avci, Babicz, Thompson, Johnson, Ke and Woodsin press). Specifically, we conducted a series of one-way ANOVAs on the 570 initial submissions from 2020, with the continuous percentage of women authors as the dependent variable and the presence of a woman as either lead or corresponding author (binary yes/no) as independent variables. Results showed a higher percentage of women on author bylines when women were lead [70.0% vs. 32.9%; F (1, 568) = 302.32, p < .001, d = 1.59] or corresponding authors [71.2% vs. 37.8%; F (1, 568) = 269.94, p < .001, d = 1.52]. Similarly, among the 136 manuscripts submitted in 2020 in which the first and corresponding author were not the same person, the first author was 2.3 times more likely to be a woman when the corresponding author was also a woman, X 2 (1, n = 136) = 4.55; CI 95% 1.05, 5.05).
Study 2: Pandemic-Related Articles Published in CN Journals
Forty papers were identified across five journals, with author groups primarily located in North America. In total, 161 of the 237 authors of pandemic-related articles published in CN journals were women (67.9%). The following series of one-sample tests was anchored by the estimated 60% prevalence of women in CN (Sweet et al., Reference Sweet, Klipfel, Nelson and Moberg2021). On average, 64.4% (SD = 29.90, range = 0.0–100) of the authors on the bylines of the 40 articles were women, which did not differ statistically from the estimated prevalence of women in the field (one-sample t = 0.93, df = 39, p = .357, Cohen’s d = 0.13). Likewise, 67.5% and 65.0% of the lead and corresponding authors were women, respectively, which did not differ significantly from the estimated prevalence of women in the field (one-sample X 2 s < 1.0, ps > .05).
DISCUSSION
Emerging research suggests that women in academia were disproportionately affected during the early part of the COVID-19 pandemic in both home and professional activities, including publishing (Deryugina et al., Reference Deryugina, Shurchkov and Stearns2021; Krukowski et al., Reference Krukowski, Jagsi and Cardel2021). Encouragingly, the present findings show no evidence of gender disparities in authorship or author byline position in manuscripts submitted to four major neuropsychology journals. In fact, submissions authored by men and women both increased during the pandemic. Similarly, women were well represented among 40 COVID-19 related publications in nine major CN journals. Thus, our findings suggest that women in CN were not underrepresented in research activity as indexed by journal submissions and publications, despite the notable stressors and challenges that accompanied the COVID-19 pandemic. Representation of women as authors in manuscript submissions during the early stages of the COVID-19 pandemic was largely comparable to that of women as authors in published manuscripts both within the larger field of psychology (see Odic & Wojcik, Reference Odic and Wojcik2020) and within the subspecialty of CN (Matchanova et al., Reference Matchanova, Avci, Babicz, Thompson, Johnson, Ke and Woodsin press) over the past few years.
We considered several possible explanations for these encouraging findings. First, the research productivity of women clinical neuropsychologists may have kept pace with pre-COVID-19 research productivity levels because of the surge of young women in the field over the past two decades (Hilsabeck & Martin, Reference Hilsabeck and Martin2010; Sweet et al., Reference Sweet, Lee, Guidotti Breting and Benson2018). Younger professionals might have experienced a reduction in clinical and teaching responsibilities due to halted in-person services and might also have been less affected by increased childcare and household responsibilities as compared to mid-career women. In fact, an exploratory search revealed that among the 40 articles related to COVID-19 in CN journals, 77.7% of women as first authors were trainees or early-career scientists, as determined by their institution or personal website. Second, the sponsorship and mentorship provided to these early-career women by mid- and late-career women clinical neuropsychologists may have been protective during this tumultuous period (de Vries et al., Reference de Vries, Webb and Eveline2006). Third, compared to other fields, psychologists may be better suited to acknowledge existing biases in the professional field and challenge them (APA, 2017). Relatedly, clinical neuropsychologists benefit from training in clinical psychology, including learning effective coping strategies, which in turn may have built added resilience to the stress of COVID-19. Interested readers are referred to a recent series of publications in The Clinical Neuropsychologist (Hilsabeck & Rivera Mindt, Reference Hilsabeck and Rivera Mindt2020) for recommendations on ways we can act as individuals and as a discipline to bolster women in the field of CN.
The current study is not without its limitations. Gender coding systems were binary and based on estimated probabilities from national databases that may have resulted in gender assigned to authors that does not match their identity. The lead or corresponding authors were uncodable for 4.5% of the articles, which were therefore excluded from analyses. This rate of exclusion is slightly lower than reported in prior studies using similar coding methods (Matchanova et al., Reference Matchanova, Avci, Babicz, Thompson, Johnson, Ke and Woodsin press; Odic & Wojcik, Reference Odic and Wojcik2020), with Asia being the geographic region of origin for most of these articles. A minority of authors were from regions other than Europe/UK and North America. Only four of nine CN journals participated in Study 1; however, all nine journals were included in Study 2 and its similar results help to bolster the representativeness and robustness of the findings. Lastly, information about the race and ethnicity of the authors was not available, and we therefore were unable to comment on the potential intersectional impact of gender and race/ethnicity on research productivity during early stages of the COVID-19 pandemic. Given the underrepresentation of people who identify as Black and Latinx in CN (Sweet et al., Reference Sweet, Klipfel, Nelson and Moberg2021), prospective studies are needed to identify specific risk and resilience factors related to scientific productivity and academic advancement in individuals from these ethnoracial backgrounds.
ACKNOWLEDGEMENTS
The authors thank all of the women who have served as our mentors, sponsors, and collaborators over the years and in years to come. The authors also thank Steph Pollock and Lois Jones at the American Psychological Association for their efforts in securing and coding the data for Neuropsychology.
FINANCIAL SUPPORT
There were no funding sources for this project.
CONFLICTS OF INTEREST
Several authors serve major editorial roles for the journals that were included in this study. Dr. Lee is the Editor-in-Chief of Archives of Clinical Neuropsychology and is on the editorial board of Neuropsychology Review. Dr. Rapport is the co-Editor-in-Chief for Journal of Clinical and Experimental Neuropsychology and is on the editorial board of The Clinical Neuropsychologist. Dr. Suchy is the Editor-in-Chief for The Clinical Neuropsychologist and is on the editorial board of Journal of the International Neuropsychological Society and Neuropsychology Review. Dr. Yeates is Editor-in-Chief for Neuropsychology and is on the editorial board of Archives of Clinical Neuropsychology, Child Neuropsychology, and Journal of Head Trauma Rehabilitation. Dr. Woods is an Associate Editor for both The Clinical Neuropsychologist and Neuropsychology and is on the editorial boards of Archives of Clinical Neuropsychology, Journal of Clinical and Experimental Neuropsychology, and Neuropsychology Review. The other authors report no other conflicts of interest.