Do men and women differ in personality?

Do men and women differ in personality?

On average, yes, by a small amount, and the average is a poor guide to any individual. Secondary analyses of Revised NEO Personality Inventory data from 26 cultures, covering 23,031 people, found women describing themselves as higher in Neuroticism, Agreeableness, warmth and openness to feelings, and men describing themselves as higher in assertiveness and openness to ideas. Those differences were small relative to the variation within each gender, and they replicated in both college-age and adult samples (Costa et al., 2001).

A later study gave the Big Five Inventory to 17,637 people in 55 nations and found women reporting higher Neuroticism, Extraversion, Agreeableness and Conscientiousness than men across most of those nations (Schmitt et al., 2008). Different instruments, different samples, many languages, one consistent direction.

What follows is less dramatic than the headline. A small average gap between two very large groups says almost nothing about the person in front of you, and this page is mostly about why.

What did the two large cross-cultural studies find?

One thing first, because it is what makes the comparison meaningful. The five dimensions being compared are not an artefact of one language: the same structure held across 17,837 respondents in 56 nations and 28 languages (Schmitt et al., 2007).

The 26-culture analysis used a long-established questionnaire in each culture's own language. Its central result combined two things: the direction of the differences was broadly what people expect, and the magnitude was small next to how much individuals differ. The authors also found, to their surprise, that the size of the differences was not constant across cultures (Costa et al., 2001).

The 55-nation study tested that variation directly. It replicated the finding that differences are larger in prosperous, healthy and egalitarian cultures, and reported that higher human development, meaning a long and healthy life, equal access to knowledge and education, and economic wealth, was the main nation-level predictor of a larger gap. Its authors reported that the variation across cultures appeared to be driven mainly by changes in men's traits (Schmitt et al., 2008).

The word small in that column comes from Costa et al. (2001), which reports the differences as small relative to individual variation within genders. The 2008 report gives no numeric effect sizes, so none appear here.

Two rows are mixed because the dimension hides the answer. Extraversion contains both warmth and assertiveness, and those facets moved in opposite directions; Openness contains both feelings and ideas, and the same thing happened. Reported as one score, opposing facet-level differences partly cancel.

Why does a small average difference matter so little for you?

Because two groups with different averages still occupy almost the same range. Picture the Agreeableness scores of thousands of men and thousands of women laid over each other: the two shapes are nearly on top of one another, with the women's shape a little to the right. Many men score above the average woman, and many women below the average man.

• An average is not a description of a person. Knowing someone's gender tells you very little about where they fall on any of these dimensions.
• Within-group variation is the large effect. People of the same gender differ far more from each other than the two group averages differ (Costa et al., 2001).
• Facets can point opposite ways. Warmth and assertiveness both belong to Extraversion, and carried differences in opposite directions.
• Self-report is what was measured. These studies recorded how people described themselves, not how they performed under observation.

So the findings do not support statements of the form men are like this and women are like that. They support something narrower: average every man and every woman in these samples, and the two averages differ slightly. Your own score is your own, and reading it against your gender adds nothing useful.

Why are the differences larger where sex roles are less traditional?

This is the part that most surprises people, and it is well replicated. The 26-culture analysis found the differences most pronounced in European and American cultures in which traditional sex roles are minimised, against the simple prediction that equal roles would produce similar personalities (Costa et al., 2001). The 55-nation study replicated that pattern (Schmitt et al., 2008). Observer ratings agree: when students in 50 cultures rated someone they knew well, the differences replicated the self-report results and were most pronounced in Western cultures (McCrae & Terracciano, 2005).

Why that happens is genuinely unsettled, and this page will not pick a side. Costa et al. (2001) discuss the possibility that in traditional cultures people attribute masculine and feminine behaviour to a person's role rather than to their traits, which would change how people describe themselves. Schmitt et al. (2008) propose that in developed nations the traits of men and women are less constrained and so diverge more freely. Others raise the possibility that people compare themselves with those around them when answering, so the same questionnaire means something slightly different in different places.

All of those are proposals rather than findings. The two studies established the pattern; neither established the cause, and honest reporting stops at that line.

What does this mean for reading your own result?

Most questionnaires place your score against a reference sample of adults and report it as a percentile. If that sample holds men and women in roughly population proportions, your percentile answers how do I compare with adults in general, not how do I compare with people of my gender. For the dimensions above, those two answers can differ slightly.

That difference rarely moves a score across a band boundary, but it is worth knowing for the same reason age differences are: a percentile depends on who is in the comparison group. Where a result sits close to a boundary, do not read much into which side of the line it falls.

The more useful move is to read the facets. A single Extraversion percentile can hide a high warmth score and a low assertiveness score, exactly the pattern the group-level findings show. How to read your personality test scores covers percentiles and bands, and what emotional stability means covers the dimension most discussed here.

How does Wellington measure this?

Wellington is a science-based personality assessment from Therabot Labs LLC. It measures personality as continuous traits, built on the Big Five and HEXACO models, and writes the results back as a personal report rather than a type.

The free Snapshot asks 60 questions and scores 6 dimensions: Extraversion, Agreeableness, Conscientiousness, Emotional Stability, Openness to Experience and Honesty-Humility. Each trait is the average of several statements on a five-point accuracy scale, reported as a percentile against the reference sample, in one of three bands, Quiet below the 30th percentile, Balanced to the 70th, High above. The bands and the Portrait names exist to make the result easier to talk about; the percentile is the measurement, and it is always shown alongside. Our norms are provisional while the norming sample is completed, and they are general rather than separate for men and women, so read your percentile as a comparison with adults at large. The first ten questions are answered before anything is asked of you; after question ten you sign in with an email address so your answers are saved and your results can be shown. No card is involved.

Nothing in a Wellington report is written differently depending on your gender. In the written report, every trait has three pages, one per band, prepared in advance by the writing and psychology team, and your answers decide which you read. The Wellington Membership adds 336 further questions to reach 90 traits, including every Big Five facet, which is where a mixed Extraversion or Openness score becomes legible, and continues toward all 255 traits across eight domains. Wellington is a wellness tool for reflection, not a medical or psychological diagnosis, and it does not replace care from a qualified professional. It is not designed for hiring, and you can export or delete your data at any time.

Questions people ask

Are women more emotional than men?

Women report higher average Neuroticism, a tendency toward worry and negative feeling, and higher openness to feelings (Costa et al., 2001). Both gaps are small next to how much people of the same gender differ from each other, and the finding describes self-reported tendencies in large samples rather than how any individual feels.

Do these differences show up when other people do the rating?

Yes. When students in 50 cultures rated an adult they knew well, using an observer version of the same questionnaire, the differences between men and women replicated what self-reports had shown, and were most pronounced in Western cultures (McCrae & Terracciano, 2005). So the pattern is not only an artefact of how people describe themselves.

Are the differences big enough to predict anything about one person?

Not usefully. The average gaps are small next to the spread within each group (Costa et al., 2001), so the two distributions overlap almost completely. Gender is a weak predictor of any individual trait score, and a score is a far better guide than group membership.

Should a personality test use separate norms for men and women?

It is a real choice, defensible either way. Separate norms answer how do I compare with people of my gender; combined norms answer how do I compare with adults in general. Wellington uses general norms, provisional while the norming sample is completed, so a percentile compares you with adults at large.