The widespread belief that women are more emotionally intelligent than men has some scientific basis, but the real picture is much more complex. One of the largest analyses on the subject shows that women, on average, achieve higher overall scores for emotional intelligence, but the advantage depends heavily on the specific emotional skill, how it is measured, and the social environment.
The meta-analysis, published in September 2026 in the scientific journal Emotion of the American Psychological Association, combines 1,279 scientific publications, 1,428 separate samples, and data on 609,278 people from 76 countries. Published and unpublished research spanning nearly three decades to the end of 2023 was included.
The study was led by Dr Mary Hausfeld from the University of Limerick, with participation from Vera Hampel and Leon Bach from the University of Zurich, and Prof Jochen Menges from Cambridge Judge Business School and the University of Zurich. Rather than treating emotional intelligence as a single, general quality, the researchers looked at the abilities of recognizing, understanding and regulating emotions separately.
The results show that women and girls, on average, are better at recognizing and understanding emotions. However, when it comes to emotional regulation, there is no such simple relationship. Men and boys show a moderate advantage in tasks that involve managing their own emotions, while women perform better when it comes to regulating the emotions of others. Overall,women's advantage is more pronounced in skills that focus on others than on their own emotions.
The way in which emotional intelligence is measured is particularly important. In tests in which participants actually have to solve tasks and their responses are scored on performance, the average advantage for women is significantly greater. In questionnaires in which people rate how well they understand and manage emotions, the differences are significantly smaller.
This difference has practical implications. Self-reported high emotional competence does not necessarily equate to demonstrated ability. The authors point out that when selecting and promoting employees, organizations should be careful not to take a candidate’s confidence as evidence of real emotional skills.
Culture and work environment also matter
The differences between women and men are not the same across all societies and occupational groups. The analysis finds that leadership position, type of work environment, geographic region, age, and the year in which the data were collected can change both the size and sometimes the direction of the observed differences.
For example, the gender gap in recognizing emotions is larger among people in management positions. Contrary to the researchers’ initial expectations, some differences turn out to be smaller in industries dominated by men. Geographic analysis also shows more pronounced differences in parts of Europe, North and Central America than in other regions.
The data also suggest a change over time. The gender gap in overall emotional intelligence has been narrowing in more recent studies, with the trend particularly pronounced among children. One possible explanation is a change in the way girls and boys are socialized and educated about emotions. However, the authors caution that the analysis does not allow for the full separation of the effects of age, generation and historical period and therefore does not prove the cause of the trend.
The study mainly questions the idea that emotional intelligence can be defined as a typically "feminine" or "masculine" quality. The results show average differences between groups, but do not mean that a particular woman necessarily has higher emotional intelligence than a particular man.
According to the authors, a more useful approach is to view emotional intelligence as a set of different abilities, rather than as a single personality characteristic. This shifts the question from “Which gender is more emotionally intelligent?“ to the more precise: Which emotional skills do we measure, how do we measure them, and in what context?