Despite progress in recent years, women remain underrepresented in scientific research, and this is especially true in more senior ranks. What role might unconscious biases play in holding women back?
The core of the issue isn’t a problem of inherent interest or education — female enrollment in science and engineering degrees is up, even dominating in science and medicine. But the average percentage of female researchers across all disciplines sits at only 35 percent, a fraction that only diminishes when dialing in specifically on more senior positions.
It can be difficult to tease apart the complex reasons why this happens. Past research has relied heavily on hypothetical scenarios, such as questionnaire-based ratings of candidates in mock hiring trials. In these experiments, there is nothing real at stake.
Behavioural scientists from Centre National de la Recherche Scientifique (CNRS) and the University of British Columbia instead analyzed two years of actual hiring practices across the scientific spectrum during an annual nationwide competition for elite research positions at the CNRS. Their findings were published in Nature Human Behaviour.
In line with the prestigious nature of these job opportunities, the candidates considered were all accomplished research scientists who met a high standard of scientific excellence as laid out by the CNRS.
Each selection committee was therefore tasked with making subjective decisions by judging qualitative features like an applicant’s perceived originality. These are areas where biases can creep in.
Probing implicit and explicit gender beliefs
The research team looked for both explicit and implicit gender beliefs in the committee members, split into 40 selection committees.
Explicit beliefs are conscious and deliberate. The researchers learned more about this through questionnaires. One person might chalk up any gender disparities in the sciences to women’s career choices or gender-based differences in ability. Another person might point to the existence of discrimination or demands of family life.
Next, they probed for implicit biases that committee members might not be aware of. For this they used an implicit association test, or IAT. The IAT is widely used to measure implicit associations, such as a stronger association between men and the sciences than women. These associations have the potential to bias thought and behaviour, even if these influences aren’t conscious or intentional.
The hope was that explicit beliefs and education programs might help counteract implicit biases.
Some committees don’t believe discrimination exists
All the committees had a strong “science=male” association based on their IAT scores, which were similar to the general population’s scores. This was despite working alongside both men and women during their careers as scientists.
On a committee level, there was some explicit agreement across the board that women face discrimination in science. However, about half of the committees tended to disagree that gender bias and discrimination contributes to women’s under-representation in STEM fields.
In the first year of the study, the committee members were all told that their hiring decisions would be scrutinized for gender bias. With this information at the top of their minds, they were likely more cautious with their selections.
In the second year, there was no reminder given, and therefore there was likely less conscious thought given to actively counter gender biases.
They found that committees with stronger implicit biases who don’t see external barriers or discrimination as a systemic problem had the biggest decreases in women promoted in year 2, even while accounting for the number of men and women in the applicant pool.
By contrast committees that explicitly acknowledged the possibility of bias and implicit stereotypes saw no decrease in year 2, no matter how strong their implicit biases were.
That means there is hope for explicit beliefs to counter implicit biases. To be effective, training would need to teach committee members to recognize the conditions where implicit stereotypes are influencing their judgement. It would also need to recognize external barriers that women in science face, and provide strategies likely to control the influence of implicit biases.
Diverse representation at all levels in the sciences is important, but getting there will take conscious effort. Even though most scientists would explicitly agree that the sciences are not just for men, this study points out an important direction where there is still work to do.
