(엠바고 해제 발송) 의약 분야 여성 연구자, 동료평가 시스템에서의 차이 전문가 의견 26-008 '여성 연구자 논문, 남성 저자 논문보다 심사 기간 길어 (PLOS Biology)'
2026.1.21. **엠바고 21일 04시 해제**
|
|
|
-
배경
-
동료 평가 제도가 여성에게 불리하게 작용하는지에 대해서는 여러 논쟁이 있었습니다.
-
연구팀은 전 세계 연간 발표 논문의 36%를 차지하는 의학 및 생명과학 분야 논문 전체를 대상으로 성별 격차를 연구했습니다. 연구 결과는 21일 04시 'PLOS Biology'에 발표됐습니다.
-
그 결과, 여성 저자 논문의 심사 소요 시간 중앙값이 남성 저자 논문보다 7.4%~14.6% 더 길었으며, 여러 요인을 통제한 후에도 차이가 유의미하게 유지됨을 확인했습니다.
-
연구팀은 성별 격차가 광범위하게 존재하며, (그 분야에서 여성의 대표성이 얼마나 높은지와 무관하게) 대부분의 학문 분야에 영향을 미친다고 지적했습니다.
-
논문 제목: David Alvarez-Ponce, et al., Biomedical and life science articles by female researchers spend longer under review (10.1371/journal.pbio.3003574)
-
- 논문의 맥락, 관련 과거 논쟁이나 논란
- 의의, 현재 시점에서 조명 받아야 하는 이유
- 분석 결과 중 특기할 만한 내용(결론, 해석, 연구 방법 등)
- 보고서의 한계, 취약점, 또는 결과 해석 시 주의해야 할 점
- 한국에 시사하는 점, 우리가 특히 주목해야 할 부분 등
- 해외 SMC(대만, 일본, 뉴질랜드, 스페인)에서도 의견을 받고 있습니다.
|
|
|
기자 여러분은 아래 주의사항을 참고해 활용해주시길 부탁드립니다.
- 엠바고는 21일 04시 해제됐습니다. 자유롭게 활용 가능합니다.
- 되도록 원문을 그대로 활용해주시길 부탁드립니다.
- SMCK를 꼭 인용할 필요는 없습니다. 만약 인용 출처가 필요한 경우, 아래 형식을 따를 수 있습니다.
- "ㅇㅇㅇ(전문가)는 한국과학기술미디어센터에 ㅁㅁㅁ라고 말했다."
|
|
|
홍성욱 서울대 과학학과 교수
*2026.1.20.
논문 저자들은 여성이 제1, 교신 저자가 된 논문의 심사가 오래 걸린다는 사실에서 의약(biomedicine) 분야의 시스템이 여성 저자에게 편견을 갖는 방식으로 고착되어 있다고 주장하는 듯하다. 저자들은 출판된 논문에서 성별을 해석하고, 역시 같은 논문에서 심사 데이터를 추출해서 위의 결론을 냈다. 실제로 시간이 오래 걸린다는 위의 사실에는 문제가 없어 보인다. 그런데 보통 학술지는 심사를 할 때 저자 이름을 뺀 채로 논문을 심사위원에게 보내는데(anonymous peer review), 이에 대해서는 논의가 없다. 논문 중간에도 언급되지만, 여성 저자의 경우 심사 결과를 받고 논문을 고치는 데 더 오랜 시간을 쓸 가능성이 있다. 이 부분이 분명하지 않아서 논문의 주장에 충분히 설득이 되지 못했다.
comenius@snu.ac.kr
|
|
|
홍진규 연세대 대기과학과 교수
*2026.1.20.
논문 출판 과정에서 다양한 소수 연구자가 겪을 수 있는 구조적 불이익은 비교적 오래전부터 문제의식으로 제기되어 왔습니다. 과학기술 분야는 여성 비율이 낮은 영역이 적지 않은데, 성별 이슈에 더해 연구자의 국적, 소속 기관, 경력 단계 등 다층적인 정체성이 심사 및 편집 과정에서 암묵적 편향의 단서로 작용할 수 있다는 우려를 지속시켜 왔습니다.
이러한 문제의식의 연장선상에서 2008년 Nature1)는 편집글을 통해 저자 신원을 심사자가 알 수 없도록 하는 이중맹심사(double-blind peer review) 도입의 타당성을 공개적으로 논의한 바 있습니다. 최근에는 일부 주요 출판사와 저널이 출판 과정에서의 성별 대표성 및 편집·심사 단계의 지표를 공개하며 점검을 시도하고 있습니다. 최근 공개된 Nature Portfolio의 보고서2)에 따르면, 편집 및 심사 결정 단계에서 여성 저자에 대한 명시적인 차별 증거는 발견되지 않았으나, 여성 교신저자의 투고 비중 자체가 현저히 낮다는 대표성 격차가 있음을 시사합니다. 연구 설계 단계부터 성별 및 젠더 관점을 체계적으로 반영하기 위해 2016년 제정된 SAGER 가이드라인3)은 출판 윤리의 새로운 이정표를 제시했습니다. 이러한 제도적 장치가 연구의 투명성을 제고하고 인지적 편향을 완화하는 데 실질적으로 기여하고 있음을 시사하는 연구들도 보고되고 있으나, 효과의 크기와 방향은 분야·저널 운영 방식·측정 지표에 따라 달라질 수 있어 추가적인 검증이 필요합니다.
국내에서도 관련 조사 분석을 기반으로 국내 고유의 연구·출판 환경을 반영한 가이드라인 마련과 더불어, 제도의 효과성 평가를 토대로 학술지 평가 기준에 형평성·투명성 관련 지표를 보완하고 그 효과를 측정할 방법에 대해 검토할 필요가 있습니다.
참고 문헌
- https://www.nature.com/articles/451605b
- https://www.nature.com/immersive/content/gender-gap-report/executive-summary/index.html
- https://link.springer.com/article/10.1186/s41073-016-0007-6
jhong@yonsei.ac.kr
|
|
|
박종화 울산과학기술원 바이오메디컬공학과 교수
*2026.1.20.
이 논문의 가치는 전례 없는 샘플 규모에 있습니다(논문 수천만 건). 그러나 여성 저자 논문의 7~14% 심사 지연을 성차별로 볼 수 있는 인과관계는 증명하지 못합니다. 역설적으로, 이런 리뷰 지연은 일부 심사자들이 여성 저자에게 신중하고 격려적인 피드백을 준 '긍정적 차별'의 결과일 가능성마저 있습니다. 여성 저자 논문 인용률이 더 높다는 일부 선행 연구1)도 이 가능성을 뒷받침합니다.
진짜 문제는 논문이 부차적으로 다룬 나라별 GDP 효과 내용입니다. 저소득 국가 저자들의 25~44% 지연은 성별 효과의 3배 이상으로, 명백한 지역/언어/문화/인종 편향을 증명합니다. 과학계에 큰 객관성 문제가 있다는 뜻입니다. 해결책은, 현재로서는 완전한 이중맹검 심사(double blind)와, AI와 인간의 혼합 리뷰 체제를 과학계가 적용하는 것입니다. AI는 통계 오류, 방법론, 이미지 조작을 객관적으로 검증하고, 인간 전문가는 과학적 중요성, 창의성을 평가하는 식입니다. 역사적으로는, 수백 년간 서구에서 유지된 과학계의 동료평가(peer-review) 대신, 출판 후 평가(post-publication review) 체제 도입 등 대체 시스템을 검토할 때가 온 것이라는 의미가 있습니다.
참고문헌
- https://doi.org/10.1093/qje/qjz035
jongbhak@genomics.org
|
|
|
아래는 대만 사이언스미디어센터(UK SMC)에서 수집한 전문가 반응입니다. 엠바고는 21시 04시로 동일합니다. 이후로도 다수의 전문가 의견이 취합되고 있습니다. 전체 코멘트는 대만 SMC 홈페이지에서 확인할 수 있습니다. |
|
|
2025/01/16
Distinguished Professor Yu Ting Liu, Department of Soil and Environmental Sciences, National Chung Hsing University
This study analyzes biomedical and life science articles indexed in PubMed and estimates the number of days from manuscript submission to acceptance. The results suggest that, across most biomedical journals, papers with women as first authors or corresponding authors tend to have longer review cycles than those authored by men, and manuscripts with authors based in low-income countries also tend to take longer than those from other countries. Such differences may have implications for career development and cumulative productivity.
Several factors may contribute to these patterns. Women scientists often face additional external burdens (e.g., household and childcare responsibilities), which can make it more difficult to attain senior or leading positions and to access flexible research resources. Lower average seniority or constrained resources may lead to more extensive revision requests and slower mobilization when additional experiments or analyses are required. For national differences, non-native English speakers may face higher costs in academic writing and in responding to reviewers; researchers in low-income settings may incur greater time and financial barriers to generating additional data or conducting further analyses; and reviewers may be less familiar with local research communities, potentially leading to more cautious and prolonged review processes.
Important limitations should be noted. Inferring gender from names may involve misclassification and cultural bias, and the analysis does not account for key factors such as authors’ seniority/age/rank, editors’ gender, whether journals use double-blind review, or whether authors are on parental leave. More importantly, overall submission-to-acceptance time cannot distinguish whether delays arise from editorial handling, reviewer turnaround, or author revision time; therefore, the findings cannot be used to directly infer “reviewer or editor discrimination.” Longer review time should also not be equated with higher quality; it may reflect additional revisions, but it may also indicate unnecessary or unequal demands that slow dissemination, affecting career-level efficiency more than any single paper’s quality.
Nevertheless, women scientists and researchers in low-income countries may be disadvantaged in international visibility and access to flexible resources, which could expose them to more stringent requests while limiting their ability to respond quickly. Beyond journal-level reforms (e.g., double-blind review and publicly reported process indicators), universities and research institutions could provide more concrete support to reduce structural workload disparities and improve women researchers’ scientific performance.
yliu@nchu.edu.tw
2025/01/16
Associate Investigator and Attending Physician Chia-Yu Chi, National Institute of Infectious Diseases and Vaccinology, National Health Research Institutes
In biomedical research, one of the hardest parts after submission is waiting—waiting for an editor’s decision, reviewers’ comments, and (hopefully) an acceptance email. But is that waiting time the same for everyone?
A study published in PLOS Biology analyzed millions of PubMed-indexed papers with traceable submission and acceptance dates. It reports that manuscripts whose first or corresponding authors were inferred to be women based on their names tended to spend longer in the overall submission-to-acceptance process.. A similar pattern appeared for authors in lower-income settings (by GDP per capita), whose papers also tended to take longer to reach acceptance. For any single article, the gap may seem modest—roughly a few extra days to about one or two weeks. Over a career, however, small delays can accumulate into a substantial “time cost.” The study estimates that across 50 papers, women authors may spend an additional 350 to 750 days waiting for review, editorial decisions, or the completion of revisions compared with male peers.
Scientific progress depends on diverse perspectives and lived experiences. If certain groups face delays more often in the publication pipeline, their findings may take longer to become visible—and, in turn, influence when evidence enters media coverage, clinical guidance, and public discussion. Over time, persistent delays could also affect promotion, access to resources, and visibility, and may narrow the range of questions, methods, populations, and clinical priorities represented in the biomedical literature.
The study makes these potential “institutional delays” visible at scale, reminding us that inequality in science may arise not only in laboratories, but also in the publication process. The authors suggest that double-blind peer review may help reduce disparities—at least early in the process—by masking author and institutional identities and limiting stereotypes linked to names or prestige.
Important limitations remain. The study defines “time under review” as the total days from submission to acceptance, even though the process includes multiple stages (editorial handling, reviewer recruitment, reviewer response time, author revision, and repeated rounds). Because stage-by-stage timestamps are usually not public, the analysis cannot identify where delays occur; the findings therefore cannot be simplified as proof of editor/reviewer bias, or as evidence that women authors inherently revise more slowly. Gender was inferred using Genderize.io, which is imperfect and can misclassify names. Finally, because the the dataset is limited to PubMed (biomedicine and life sciences), the patterns may not generalize to other fields such as mathematics, engineering, or the humanities and social sciences.
pedchi@nhri.edu.tw
2025/01/16
Associate Professor I-Chun Pan, Horticulture Department, National Chung Hsing University
This article draws on the PubMed database, primarily covering the life sciences and biomedical fields, and compiles publications from 1781 to 2024 to examine differences in review duration (from submission to acceptance) when male and female serve as first authors or corresponding authors. The results show that papers authored by female take approximately 7.4%–12.7% longer to be accepted than those authored by male, regardless of whether female are first or corresponding authors. In addition, manuscripts from authors in low-income countries experience substantially longer review times, increasing by about 25.8%–44.3%. These disparities are widely observed across most biomedical journals and subject categories.
Because the study uses “submission-to-acceptance” as the overall indicator of review duration, it cannot distinguish whether the prolonged timeline is mainly driven by delays on the journal side (e.g., editorial handling and peer review) or on the author side (e.g., time spent preparing revisions). Nevertheless, the findings indicate that this is not an isolated phenomenon confined to a small number of journals or disciplines, but rather a systemic structural disparity. This suggests that female scientists and authors from low-income countries may face longer waiting times during peer review, resulting in cumulative disadvantages.
The study also suggests that female authors may be more frequently required to conduct additional experiments or may face stricter scrutiny of their scientific argumentation and writing. Meanwhile, authors from low-income and non-English-speaking countries may be constrained by limited research resources, inadequate infrastructure, and the added costs of language editing, all of which can prolong the review process. However, review duration itself is not equivalent to paper quality and should not be used to infer differences in scientific value. In today’s academic environment, publication output is often a primary benchmark for evaluation, and delays in peer review can affect career competitiveness and academic achievement, including promotion, hiring, and grant applications. Ultimately, the value of academic research should be judged by its innovation, forward-looking contribution, and practical impact. Gender, economic status, and language background should not determine scientific merit; therefore, the academic community should improve peer-review practices to reduce these potential hidden barriers, enhance efficiency and fairness, and promote greater equity in academia.
icp@dragon.nchu.edu.tw
2025/01/16
Professor Yun-Hsuan Chang, Institute of Gerontology, National Cheng Kung University College of Medicine
1. What is this study about? What are its limitations?
This study analyzed more than 36.5 million biomedical and life science articles and found that papers by female researchers spend longer in peer review than those by male researchers. As the largest study on this topic to date, it indicates that review delays are not isolated to specific disciplines but represent a systemic pattern across biomedicine.
For researchers, publication speed is closely tied to promotion, funding, and academic impact. When small delays accumulate over a career, female corresponding authors may spend 350–750 more days waiting for review outcomes than male corresponding authors, potentially widening the gender gap in academia.
Gender was inferred from authors’ first names, which may misclassify gender-neutral or non-binary individuals. PubMed data records only submission-to-acceptance time, so the source of delay—slow reviewers, difficulty finding reviewers, or longer author revisions—cannot be distinguished. The study also examined only accepted papers and did not track manuscripts rejected because of bias or after multiple rounds of revision.
2. What may explain the difference?
Editors or reviewers may hold stereotypes about female authors or be less familiar with topics where women are more represented, leading to stricter evaluation—so-called implicit bias. Women also tend to carry greater childcare, household, and administrative duties, which may lengthen revision time. The authors further suggest that higher levels of perfectionism or risk aversion among women may contribute to longer responses. Review times were also longer for authors from low-income countries, possibly due to limited resources, language barriers, or regional bias.
3. Does a longer review affect quality?
The time gap was unrelated to readability, article length, or number of authors. Evidence from fields such as economics shows that papers by women often achieve higher impact after major revisions, implying that reviewers may apply higher thresholds to female-authored work. This does not mean lower quality; rather, women may need to meet stricter standards to gain equal recognition.
4. How can bias be reduced?
Double-blind review, which conceals the author's identity, may limit gender or nationality bias. Journals could also monitor and publish gender-related review statistics and provide implicit-bias training for reviewers. Academic institutions should recognize women’s additional social burdens, such as childcare, and offer more flexible support to lessen their impact on research progress.
yh.chang.snoopy@gmail.com
2025/01/18
Director/Professor Hsien-Ho Lin, Institute of Epidemiology and Preventive Medicine, National Taiwan University
Associate Professor Chen-I Kuan, Institute of Health Behaviors and Community Sciences, National Taiwan University
This study examines whether gender is associated with peer-review duration in biomedical publishing. It reports that manuscripts authored by women take significantly longer to pass through review. Papers from low-income countries also face longer review times, suggesting that women researchers in low-income settings may experience compounded disadvantages linked to both gender and resource constraints.
These gaps may look like a “technical” feature of the publication process, but they can accumulate into career and knowledge-production effects. When certain groups’ work is accepted and published later, their findings may gain visibility and citations more slowly and may be disadvantaged in competitive evaluations for promotion, grants, and international collaboration. In this sense, review delays can function as an invisible barrier to building measurable outputs and reputational capital.
The pattern also reflects broader organizational structures. Prior research suggests women in biomedicine are less likely to enter influential decision-making and professional networks and less likely to advance into leadership roles. Such barriers can limit opportunities and voice, and may feed back into publication trajectories and impact, creating a subtle but persistent cycle. Underrepresentation in leadership may also weaken gender sensitivity in health policy and practice. During COVID-19, for example, constraints on schooling, childcare, and long-term care intensified women’s work–care burdens, emotional strain, and risks of domestic violence—issues that institutions may not have identified quickly enough to respond with adequate support.
A key contribution of this study is that it uses review time as a measurable indicator to strengthen empirical discussion of harder-to-observe processes such as network exclusion and promotion barriers. Improving fairness in biomedical knowledge production may therefore require not only greater transparency and reform in peer review, but also structural efforts to expand women’s access to decision-making networks and senior leadership.
The study has limitations. Although it finds an association between women authorship and longer review time, it cannot disentangle whether manuscripts require more revision on average or whether women face higher standards at comparable quality; causal claims about gender bias remain indirect. The authors also do not test how regional variation might affect gender identification (e.g., East Asian names may be harder for foreign editors or reviewers to gender). Despite these constraints, the study raises important questions, while explanations of the underlying mechanisms require further evidence.
Professor Hsien-Ho Lin hsienho@gmail.com
Associate Professor Chen-I Kuan chenikuan@ntu.edu.tw
|
|
|
한국과학기술미디어센터(SMCK) 소개
한국과학기술미디어센터는 근거 기반의 과학 정보를 언론에 제공하는, 과학계와 미디어 사이의 다리 역할을 하는 독립 비영리 조직입니다. 잘못된 정보와 가짜 뉴스가 넘쳐나는 세상에서, 제대로 된 전문가의 해설과 의견을 빠르고 다양하게 기자들에게 제시하고 이를 체계적으로 아카이빙하는 역할을 합니다.
2025년 7월 이사회를 구성하고(이사장 노정혜 전 한국연구재단 이사장) 센터장(이근영 전 한겨레 과학전문기자)을 선임했으며, 같은해 9월 개소식을 열며 활동을 시작했습니다.
*참고 기사:
SMCK 역할
SMCK는 세 분야 전문가인 과학자, 기관 커뮤니케이터(홍보팀), 기자에게 구체적인 도움을 드리고자 설립됐습니다. 각각 아래와 같습니다.
- 과학자, 연구자에게는 의견과 해설이 온전한 맥락과 함께 제공되는 안전한 발언 공간이 돼줍니다. 선의를 위해 한 논평이 기사화 과정에서 왜곡되거나 부정확하게 변질될 우려를 줄이는 완충 작용을 합니다.
- 기관 홍보 담당자에게는 기관의 성과를 기자들에게 보다 객관적이고 정교하게 알리고, SMC 글로벌 네트워크를 통해 영향력을 높일 기회를 제공합니다.
- 기자에게는 사안을 해석하는 데 도움이 될 치우침 없는 종합적인 정보를 빠르고 풍성하게 제공하고, 이를 통해 기사에서 과학과 기술을 보다 자유롭고 편리하게 활용하도록 돕습니다.
SMCK는 이를 통해, 궁극적으로 근거에 기반해 사안을 합리적으로 판단하고 이것이 정책에까지 반영되는 사회를 만드는 데 기여하고자 합니다.
해외 협력
사이언스미디어센터(SMC)는 2002년 영국에서 최초로 설립됐고 현재 호주와 뉴질랜드, 독일, 스페인, 대만 등으로 확장됐습니다. 글로벌 네트워크에 포함된 6개 조직은 엄격한 독립성과 신뢰성이라는 가치를 공유하고 있으며 협력을 통해 주요한 국제 과학 이슈에 공동 대응하고 있습니다.
한국은 신생 조직으로서 글로벌 네트워크와 긴밀히 협력하고 있습니다.
|
|
|
* 지난 의견은 '지난 의견 다시 보기'를 선택해주세요. |
|
|
내용문의: 윤신영 미디어국장 yoonsy@smck.or.kr
한국과학기술미디어센터(SMCK)
|
|
|
|
|