Gender Stereotypes Have Weakened Over the Past Decade

This post was authored in collaboration with Chiara Terzo, a research fellow at the Italian Institute of Technology (Center for Translational Neurophysiology of Speech and Communication).

In 1922, the journalist Walter Lippmann coined the modern usage of the word “stereotype” as it refers to the beliefs and representations that society holds about social groups, such as the stereotype that “men are good at math” or “women are good at caregiving.” In using this term, Lippmann was intentionally referring to the “stereotype” plates used in the printing trade.

That is, before the advent of stereotype plates, printing required placing individual letters into a block to create the “forme” where ink was then applied. Stereotypes, by locking the entire page of a newspaper in a forme and by molding a matrix, allowed newspapers to reproduce and rapidly distribute many copies of a message. According to Lippmann, the name “stereotype” communicated that our stereotypes about social groups were similarly widely spread and unchangeable.

Gender stereotypes are widely held mental associations that ascribe a man or woman with specific attributes or roles, such as career vs. family and science vs. the arts. Such stereotypes seem to be persistent over time, in part due to the fact that gender social categories are considered stable and essential divisions in society. Additionally, at a broader societal level, gender stereotypes are embedded in cultural products such as language—creating the perception that stereotypes are consensually held by everyone in society and therefore perceived as true and unnecessary to change (Charlesworth, Yang, Mann, Kurdi & Banaji, 2021).

But what do the data actually say? Have stereotypes remained stable over time, or is there a silver lining to the potential for long-term change?

In a new study published in Social Psychological and Personality Science, Ph.D. candidate Tessa Charlesworth and Professor Mahzarin R. Banaji at Harvard University tested the patterns of implicit and explicit gender stereotype change over a decade (2007-2018) from 1.4 million volunteer respondents across the U.S. and other countries (Charlesworth and Banaji, 2021). The implicit gender stereotype was investigated via the Implicit Association Test (IAT), a computer-based time-reaction task that assesses the strength of mental associations stored in memory. Specifically, participants completed an IAT measuring the association strength of the gender categories “male” and “female” with the attributes of either “science”/”liberal arts” or “career”/”family” categories. The explicit gender stereotype was measured by directly asking participants the degree to which they thought an attribute (e.g., “science”) was female or male.

The results showed that both explicit and implicit gender stereotypes have weakened by 13-19 percent between 2007 and 2018. This pattern of change was largely consistent across many different demographic groups—such as men and women, liberals and conservatives, or Black and White respondents—as well as across different geographical locations. Thus, although previous studies found that implicit cognition seems to be more resistant to change compared to explicit cognitions, this study shows that implicit gender stereotypes have, in fact, changed and at similar rates to explicit measures. Thus, on the bright side, the results showed a weakening of gender bias over time.

However, by using a statistical method that allows forecasting future patterns of change (similar to predicting stock markets), the authors showed that the stereotype is “far from eradicated.” Indeed, it could take at least 134 years for the implicit association of male-career/female-family to reach neutrality (i.e., no bias), and between 37 and 74 years for the male-science/female-arts stereotype to pass neutrality.

Why did stereotypes change over time? Is there hope for further change?

The authors suggest that societal-level changes are likely responsible for these widespread, long-term changes in explicit and implicit gender stereotypes. For instance, society has witnessed many changes in the last several decades, including changes in women’s representation across careers (i.e., STEM fields; Charlesworth and Banaji, 2019) and social media movements raising awareness about gender discrimination and violence (e.g., the #MeToo movement). As society is changing, according to the authors, implicit and explicit gender stereotypes are “far from the rigid stereotypes of a printing press.” Rather, they say that “the contemporary notion of gender stereotypes may be better reflected as a dynamic representation that is responsive to the changes in society.”

Can we speed up this process of change?

“Yes!”—Charlesworth says—“we have just looked [at] recent data (2016-2020) for implicit attitudes (e.g., race, skin-tone, sexuality), although we haven’t yet analyzed the recent gender stereotype data. We find that our forecasts often underestimated the amount of change that actually occurred in that time period, suggesting that we have indeed sped up in recent years. This is a heartening statistic because it says that the continued interventions and efforts are paying off.”

Is COVID-19 affecting gender equality, and shall we expect a big impact on gender stereotype change predictions?

We are now facing a worldwide pandemic that is inevitably affecting the lives of billions of people. Women’s jobs (i.e., education, nursing jobs, social workers, etc.) are more vulnerable to this global crisis compared to men’s jobs, and women’s employment is dropping faster than usual. “I'm sure that there will be many changes as a result of COVID,” said Charlesworth, assuring that she will be soon looking into this topic. Evidently, as Charlesworth and Banaji emphasize, continued focus and intervention will be needed to smooth the path of change for the future.