Over the past year, there has been nothing more certain than change. The COVID-19 global pandemic has forced us apart, blurring the divisions between our workplaces and homes, and threatening our health; the Black Lives Matter movement has highlighted the needs for systemic transformations in racial justice; and the rise of populism and misinformation have challenged the tenets of democracy. Although 2020 (and 2021!) are unprecedented in many ways, the fact of change is not just a quirk of the current times. Rather, change is part of the very nature of human society and human psychology. As humans, we adapt to these changes in our world by updating our beliefs and attitudes—changing what we think and feel about the world around us.

But changing our society’s beliefs and stereotypes is hard. It can take generations (or more) for a society to change how it thinks about issues like interracial marriage or climate change, or, as we study in our newest research, gender stereotypes that associate men with careers and women with family, and men with science and women with arts. Moreover, the possibility of stereotype change faces yet another stumbling block: even if we can change what we explicitly report on typical surveys and questionnaires, the more implicit, less controllable and more automatic, associations in our minds may resist long-term change. That is, it may be hard to break the link that immediately arises between the ideas, “women” and “family” (or “arts”) or “men” and “work” (or “science”).

In our research, we set out to document whether, and if so at what speed, explicit and implicit gender stereotypes (about career/family and science/arts) have changed over the past decade. As in our previous work documenting long-term changes in other implicit biases (for example, attitudes about race, sexuality, or age), we used the massive dataset collected through the Project Implicit demonstration website through which anyone can have their attitudes or beliefs measured (and learn where they stand). Specifically, we used data from 1.4 million tests of implicit and explicit gender stereotypes for two topics—men-career/women-family and men-science/women-arts—collected continuously from volunteer respondents over twelve years (2007-2018).

Good news: over this time, explicit gender stereotypes have been decreasing in bias. Stereotypes associating men-career/women-family and men-science/women-arts have lost about 19% and 14% of their original bias, respectively. These rates of change are similar to what we have previously seen in some other slower-changing explicit biases such as about body weight.

But what about implicit gender stereotypes? Psychologists (and the general public) have long thought of these implicit, automatic biases as particularly difficult to durably change. Like habits, implicit biases are thought to be deeply ingrained in our mind and environments such that they will be resistant to updating. Despite these expectations, we found that—more good news—implicit career/family and science/arts stereotypes have dropped by 13% and 17%, respectively, over twelve years. This rate of change is therefore quite similar to what we saw in explicit gender stereotypes.

Now, the next question we inevitably have to ask ourselves is who is changing? Is this stereotype change isolated to just a few groups in society (perhaps liberals, or younger folk)? Or is it actually more widespread across people and places?

To test the spread of change, we looked at differences across groups defined by gender, race, politics, age, education, and religion—that is, we compared the rate of change in men versus women, Black Americans versus White Americans, liberals versus conservatives, and so on, to see whether the two groups were moving in parallel or not. Furthermore, because of the international nature of the Project Implicit data, we were able to test whether the rate and direction of change differed across U.S. states, different countries, and different geographic regions of the world.

Gender stereotype change was not an isolated phenomenon. Nearly every demographic group, as well as 92-98% of states (depending on the stereotype in question), 82-91% of countries, and every single UN geographic region, was moving (largely in parallel) towards less bias over time. This surprisingly widespread pattern focuses our attention to forces happening at broadest level of society, such as social movements like #metoo, widespread increases of women’s representation in science-related fields and the workforce, or changes in social norms. Change is not just from the idiosyncratic motivations or actions of a few people, but, rather, is a truly societal transformation.

These results are hopeful. They show that gender stereotype change is possible and, thus, that continued efforts to promote change in society—such as widespread interventions to get more men to be interested in caregiving roles or more women to persist in science—are worthwhile and important.

However, the results also offer cautions. Implicit (and explicit) gender stereotypes remain far from eradicated. With our statistical models we were able to estimate how long it might take for these stereotypes go away. Shockingly, if we continue at past rates, it will take at least 134 years for implicit male-career/female-family stereotypes and 37 years for implicit male-science/female-arts stereotypes. Clearly, we still have much work ahead to accelerate our individual and collective change towards equality in our gender stereotypes.

For Further Reading

Charlesworth, T. E. S., & Banaji, M. R. (2021). Patterns of implicit and explicit attitudes and stereotypes III. Long-term change in gender stereotypes. Social Psychological and Personality Science. https://doi.org/10.1177/1948550620988425

Charlesworth, T. E. S., & Banaji, M. R. (2019). Patterns of implicit and explicit attitudes: I. Long-term change and stability from 2007 to 2016. Psychological Science, 30(2), 174–192. https://doi.org/10.1177/0956797618813087

Charlesworth, T. E. S., & Banaji, M. R. (2019). Gender in science, technology, engineering, and mathematics: Issues, causes, solutions. The Journal of Neuroscience, 39(37), 7228–7243. https://doi.org/10.1523/JNEUROSCI.0475-18.2019


Tessa Charlesworth recently received her PhD from the Department of Psychology at Harvard University with Professor Mahzarin R. Banaji. She studies the patterns and sources of long-term change in implicit social cognition.