Are gender differences really power differences? Large-scale analysis finds strong behavioral parallels

New research suggests that many psychological and behavioral differences between men and women might actually stem from differences in social power. The findings indicate that holding a high-power role produces behavioral effects strikingly similar to traits often associated with men, while low-power roles produce effects resembling traits often associated with women. This research was published in PNAS Nexus.

The origins of behavioral differences between men and women remain a topic of intense debate in the social sciences. Evolutionary theorists tend to trace these differences back to natural selection and biology. In this view, men and women developed distinct psychological attributes to adapt to different reproductive and survival roles in early human history.

Sociocultural theorists approach the issue differently, emphasizing how societal roles and social power shape behavior. They argue that hierarchical structures and unequal resource allocation reinforce gendered behaviors beyond any biological roots. A 2019 review indicated that human brains and hormonal systems are not strictly divided into two distinct categories, which supports the idea that many gender differences are shaped by culture rather than rigid biology.

At the center of this debate is the concept of social power, defined as the actual or perceived control over valuable resources. Across many areas of society, men tend to hold more power than women, occupying more leadership roles and controlling more financial assets. Because power differences are so frequently tied to gender, some researchers suspect that power dynamics might be the true source of many presumed gender differences.

Adam D. Galinsky of Columbia University and his colleagues set out to systematically test this idea. They wanted to see if the behaviors resulting from experimental manipulations of power would match up with known differences between men and women. By comparing these two separate areas of research, the team aimed to determine if power imbalances could provide a useful explanation for gender differences.

Free daily newsletter

To conduct the analysis, the researchers first collected data from experiments that randomly assigned people to high-power or low-power roles. They performed a statistical check on 270 of these experiments to ensure the existing literature on power was reliable. From there, they identified 59 distinct psychological and behavioral outcomes that are genuinely affected by having or lacking power.

Next, the research team gathered 102 outcomes from meta-analyses of sex and gender differences. Meta-analyses are large-scale statistical reviews that combine data from many individual studies to find overall trends. The researchers then grouped the findings from both the power experiments and the gender meta-analyses into four broad categories for comparison.

These four categories included agency, communion, self-evaluation, and cognitive processes. Agency involves striving for independence, assertiveness, and control over one’s environment. Communion relates to social connection, empathy, and cooperating with others. Self-evaluation involves how people view their own abilities, and cognitive processes include attention, problem-solving, and spatial reasoning.

The comparison yielded a massive overlap. The researchers found that about 71 percent of the recorded gender differences were entirely consistent with the effects of power differences. In contrast, only about 8 percent of the gender differences showed the opposite pattern.

Sonya Xinyue Xiao, a developmental psychologist who directs the IBAY Lab and was not involved in the study, told PsyPost that her confidence in the results is “pretty high.”

“The method they used is very rigorous in testing their hypotheses that focus on whether power effects are aligned with sex/gender differences when there is a significant power effect,” she said.

Specifically, high-power individuals and men both tend to display higher agency, including traits like dominance and a willingness to take risks. They also both show more positive self-evaluations and display similar cognitive processing styles. On the other hand, low-power individuals and women both tend to display higher communion, exhibiting more interpersonal sensitivity and social cooperation.

To check if this alignment was unique to power, the researchers ran a similar comparison using the personality trait of extraversion. Extraversion is known to correlate with having power and influence, so the team wanted to see if it would also perfectly mirror gender differences. The results showed that it did not.

While extraverts and high-power individuals share some traits, they diverge in the area of communion. Extraverts tend to be highly communal and social, whereas high-power individuals and men tend to score lower on communion. This divergence provides evidence that social power has a specific and uniquely strong relationship with gender differences.

As with all research, there are a few things to keep in mind. The study relies on comparing two separate bodies of literature, meaning the link between power and gender is observational in this context. The researchers did not conduct a new experiment to see if giving women high-power roles would immediately erase the observed gender gaps in these specific traits.

“The gender differences in many constructs are very small, but power differences may be large; it is important to keep in mind that the findings are pointing at similarity in directions rather than magnitudes,” Xiao noted. “As the authors mentioned in their discussions, their findings do not actually prove causal relations between social power and sex/gender differences.”

Additionally, grouping dozens of distinct psychological concepts into broad categories requires subjective judgment. Although the researchers used independent experts to classify the traits, personal biases or cultural perspectives could have subtly influenced how the categories were organized. The authors note that their results do not mean power explains all behavioral differences between men and women, but rather that it is likely a major contributing factor.

Future studies could directly test these ideas by assigning men and women to high- and low-power roles in the same experiment. This would help clarify exactly how power shapes behavior across different genders in real time.

“As a developmental psychologist, I am curious to see research examining how the potential causal relation between social power and gender difference changes over time — perhaps the role of social power increases as we age,” Xiao said. “Because gender is highly salient during childhood, whereas one’s gendered experiences of social power may come later.”

The study, “Are many sex/gender differences really power differences?,” was authored by Adam D. Galinsky, Aurora Turek, Grusha Agarwal, Eric M. Anicich, Derek D. Rucker, Hannah R. Bowles, Nira Liberman, Chloe Levin, and Joe C. Magee.