A new study led by Molly Lewis, a research scientist in the department of psychology at Carnegie Mellon University in Pittsburgh, finds that people’s implicit gender associations are strongly predicted by gender associations encoded in the statistics of the language they speak.
Researchers examined gender associations embedded in the statistics of 25 languages and related these to data on an international dataset of psychological gender associations. Specifically, researchers examined the regularity of words such as “career” or “work” that were in close proximity to men in a large body of written work in a given language. Similarly, they examined how words such as “home,” “family,” and “children” were in close proximity to women. The researchers also examined the extent to which languages mark gender in occupation terms (for example, waiter/waitress). The researchers also gave an implicit bias test to more than 650,000 people speaking 25 languages around the world.
The results showed people’s implicit gender associations are strongly predicted by gender associations encoded in the statistics of the language they speak. By examining the percentages of women with higher education degrees in countries speaking these 25 languages, researchers found that countries, where the citizens spoke languages that had weaker associations between men and career, tended to have more women in STEM fields.
Dr. Lewis is a graduate of Reed College in Portland Oregon, where she majored in linguistics. She holds a Ph.D. in developmental psychology from Stanford University.
The full study, “Gender Stereotypes Are Reflected in the Distributional Structure of 25 Languages,” was published in the journal Nature: Human Behaviour. It may be accessed here.