Science1 publisher2 min readPublished
Women lead on reading emotions and trail slightly on managing their own in 1,279 studies
Mary Hausfeld's team pooled 1,279 papers and data on 609,278 people; women's advantage is large on graded tasks and almost disappears on self-reports. Assessment programs built on self-ratings are measuring self-perception.
The Scientist · Science desk
What happened
- A meta-analysis published in the journal Emotion combined 1,279 academic papers and data from 609,278 people in 76 countries, covering research from the past three decades.
- Women scored higher on tasks that involve managing other people's feelings while men and boys scored higher on managing their own, so the direction of the gap depends on whose emotions are at stake.
- The gap also depends on the instrument: women's advantage was larger on performance-based tests scored for accuracy and smaller on self-report surveys where people rate their own ability.
- Over the nearly 30 years of data, overall gender differences in emotional intelligence narrowed, and the change was most pronounced among children.
Compiled by The ScientistSomething wrong?How this is made
Why it matters
- decision An employer who screens for emotional intelligence with a self-rating questionnaire is ranking self-perception, and Menges said self-perception and demonstrated ability are not the same thing.
- constraint Age, generation and time period cannot be separated in this dataset, so the narrowing cannot be offered as evidence that changing gender norms caused it.
- exposure The authors' worry is about allocation at work: an unexamined belief that women are naturally better at this lets employers push a disproportionate share of emotional labor onto them.
- contradiction Emotion-recognition differences ran larger among leaders than nonleaders yet smaller in male-dominated industries than elsewhere, which the researchers had not expected; no single selection or socialization account predicts both.
Two kinds of test carry the name emotional intelligence. One gives people a task and scores the answers for accuracy. The other asks people to rate their own ability. Women's advantage is larger on the first and smaller on the second [5]. "Women are emotionally intelligent. Men think they are," said Jochen Menges, one of the authors [6]. He said the team could not determine the reason: "Men may overestimate their abilities, women may underestimate theirs, or both may be happening. Either way, confidence and competence are not the same, and organizations should not confuse them" [8].
The researchers say emotional intelligence should not be treated as one quality, and they split it into recognizing, understanding and regulating emotions, in oneself and in others [9]. Women scored higher on recognizing and understanding [3]. "For managing their own emotions, the pattern reverses slightly, and men score higher. That matters, because being expected to manage other people's feelings is not the same as being better equipped to manage your own," said Menges [10].
The published account describes women's advantage on performance-based tests as substantial and does not attach a numerical effect size to it [7][22]. Size is what decides whether a group average has any bearing on a single candidate's assessment, and it decides how much weight the geographic result can hold: differences were largest in North and Central America and Europe compared with the rest of the world [14]. Divide the participants by the papers and the average study contributed about 476 people [21].
"Common assumptions about women's and men's emotional abilities are often too broad," said Mary Hausfeld, who led the work at the University of Limerick. "The evidence points to a picture in which differences depend on the skill being examined and the context in which it is measured" [18]. The authors do not treat small differences as trivial; they say such differences can accumulate through everyday decisions and produce gaps in outcomes such as promotion [20]. Studies included in the analysis have linked high emotional intelligence to job performance, stress management and workplace relationships through communication and conflict management [15].
Menges said organizations should stop treating emotional intelligence as a feminine trait: "Treat it as a set of skills. Use validated performance-based measures where they exist, rather than relying on self-reports. Develop these abilities in everyone" [19]. I would not run a hiring or promotion screen on self-report emotional intelligence scores after reading this. The two instrument families disagree about the thing they both claim to measure, and only one of them is scored against an answer key [5].
What to watch
- The effect sizes in the Emotion paper itself, and how many of the 1,279 studies sit behind each of the four skill domains.
- Whether commercial emotional-intelligence assessments used in hiring publish performance-based validation rather than self-report norms.
- Whether the leader versus nonleader difference survives controls for who gets selected into leadership in the first place.