Summary: Considering sex (the biological characteristics that distinguish males, females, or intersex individuals) and gender (the psychological, social, and cultural factors shaping identity and roles) in research design can strengthen scientific rigor, reduce bias, and open new opportunities for discovery and innovation.
Source: University of Exeter
Thinking about sex and gender would help scientists improve their research, a new article argues.
In a special 150th anniversary edition of Nature, five experts make the case that sex and gender are frequently overlooked in scientific studies. They argue that integrating these factors more consistently across disciplines improves experimental design, reduces bias, and creates fertile ground for new insights and technological advances.
The authors distinguish sex as the biological attributes that make someone male, female, or intersex, and gender as the psychological, social, and cultural influences that shape how people identify and behave in society. They show how both dimensions can affect outcomes in unexpected ways and why researchers should explicitly consider them when planning, performing, and interpreting studies.
Examples cited in the article range widely: male and female shellfish may respond differently to climate stressors; social robots are frequently gendered by human users, affecting how people interact with them; and computer vision systems have been found to misclassify the sex of darker-skinned women more often than lighter-skinned men, exposing algorithmic bias with social consequences.
“It’s striking how often sex and gender are neglected in science,” said co-author Dr Robert Ellis from the University of Exeter. “We should build these considerations into every stage of research, or else give robust, evidence-based reasons for why they are not relevant. When sex or gender do matter, ignoring them can skew results and slow progress.”
The authors note progress in some areas. Early crash test dummies were modeled on an average male body, and later studies showed that female drivers were substantially more likely to suffer severe injuries in comparable crashes. That finding prompted improvements in safety testing and vehicle design with clear implications for injury prevention and saving lives.
Misunderstandings and under-consideration persist, however. The paper points out that even researchers’ own sex and gender can influence how they interpret observations, which means study teams, analysis, and reporting should reflect that possibility.
The article focuses on four domains—marine science, biomedicine, robotics, and artificial intelligence—but draws lessons that apply across disciplines. In biomedicine, for example, Dr Cara Tannenbaum highlights discoveries about sex-linked differences in immune function. Immune cells can behave differently depending on sex chromosome complement (XX versus XY) and exposure to sex hormones. These biological differences have important implications for antibody therapies and cancer immunotherapy, suggesting that treatments may one day be tailored by sex to maximize safety and efficacy.
Another striking finding discussed in the paper comes from preclinical research: mice displayed different pain responses depending on whether a male researcher was present, likely reacting to scent cues associated with men. Both male and female mice showed the effect, but female mice were more sensitive. This kind of result underscores how seemingly peripheral factors can influence experimental outcomes and should be controlled or reported.
Marine biology provides further examples. Dr Ellis notes that in some sea turtles, incubation temperature determines hatchling sex, so rising global temperatures could skew populations toward one sex and threaten species viability. Other marine species challenge binary assumptions about sex: clownfish are protandrous hermaphrodites, maturing first as males with the capacity to change to females under social triggers. In clownfish social groups, the dominant female pairs with a single large male; if the female is removed, the dominant male changes sex and subordinates move up the social hierarchy. These natural dynamics illustrate how sex and reproductive strategies vary across life forms and why generalized assumptions can blind researchers to important biological realities.
The paper’s stated goal is to “increase transparency, promote inclusion and reset the research default to carefully consider sex and gender, where appropriate.” An example from robotics shows how quickly people ascribe gender to nonhuman entities: through anthropomorphism, users assign gendered attributes to social robots, which influences how those devices are perceived and adopted. Dr Friederike Eyssel, a social psychologist and roboticist, emphasizes that gendering technologies has social and ethical implications that developers and stakeholders must weigh carefully. She calls for more field research to understand how gendered designs affect real-world use and social outcomes.
The Nature “Perspective” outlines a practical roadmap and urges coordinated action by researchers, funding bodies, journals, and universities to adopt rigorous sex and gender analysis methods. The authors write that integrating these analyses into mainstream science will enhance both scientific excellence and social responsibility in research and engineering.
The article was led by Professor Londa Schiebinger of Stanford University and co-authored by Dr Robert P. Ellis (University of Exeter), Dr Cara Tannenbaum (University of Montreal and the Canadian Institutes of Health Research), Professor Friederike Eyssel (Bielefeld University, Germany), and Dr James Zou (Stanford University).
Dr Ellis, Dr Tannenbaum, and Dr Schiebinger are members of an expert panel advising the European Commission on how to integrate sex and gender considerations into funding decisions for forthcoming EU research programs.
Source:
University of Exeter
Media contacts:
Duncan Sandes – University of Exeter
Image source:
The image is in the public domain.
Original research: Closed access. “Sex and gender analysis improves science and engineering.” Cara Tannenbaum, Robert P. Ellis, Friederike Eyssel, James Zou & Londa Schiebinger. Nature. doi: 10.1038/s41586-019-1657-6.
Abstract
Sex and gender analysis improves science and engineering
The aim of sex and gender analysis is to promote rigorous, reproducible, and responsible science. Incorporating sex and gender considerations into experimental design has advanced multiple fields, from better cardiovascular treatments to deeper understanding of algorithmic bias. This Perspective discusses how sex and gender analysis can drive discovery, increase experimental efficiency, and support social equity. It offers a roadmap for applying these methods across scientific disciplines and calls on researchers, funding agencies, journals, and universities to coordinate efforts to implement robust sex and gender analysis.