The Women-Are-Wonderful Effect (We Don’t Live in a “Culture of Misogyny”)

Attitudes towards women are hostile and contemptuous, according to the standard model of prejudice (as described by Glick & Fiske, 2001). This view is prevalent:

A “culture of misogyny” is “deeply-rooted in our society” (Kathleen Wynne, Premier of Ontario). “We live in a culture of casual misogyny” (Megan Leslie, former deputy leader, New Democratic Party in Canada). We live in a “patriarchal misogynistic society” (Tina Garnett, Equity Committee Coordinator, York University). “Misogyny runs so deep in this society […] all the women-hating, woman-blaming, woman-fearing instincts” (Polly Toynbee, columnist for The Guardian). “[W]e should look in the mirror for woman-hating culture” (Stephen Hume, columnist for the Vancouver Sun). “This Is How Much America Hates Women” (Anne Helen Petersen, PhD, writer for BuzzFeed News). “America really, really hates women […] there is actually nothing women can do that is right” (Megan Murphy, Feminist Current).

But the actual psychological research on gender attitudes and stereotypes paints a very different picture, one where women are viewed more positively than men. Glick and Fiske describe this phenomenon (the women-are-wonderful effect) as an “extremely robust” finding (although it is found more strongly among women than among men).

Continue reading “The Women-Are-Wonderful Effect (We Don’t Live in a “Culture of Misogyny”)”

Gender Differences in Employment Priorities and Interests

Men and women tend towards different fields and even different positions within a field. One of the possible reasons for this is that the genders differ in their preferences when seeking employment. What does the evidence say? Here I look at two large meta-analyses covering over one million people to find out.

Somewhat surprisingly, they indicate that men are only minimally more interested in earnings and leadership roles than women are (women lean a bit towards caring more about benefits). These differences in priorities thus appear to be relatively small factors in explaining differences in leadership roles or average salary.

However there are much larger differences in men’s and women’s vocational interests, with men being more thing-oriented and women being more social or people-oriented. These differences in interests thus appear to be relatively large factors in explaining men’s predominance in fields like engineering and computer science, and women’s predominance in fields like education and nursing.

The differences in vocational interests line up with with psychologist Simon Baron-Cohen’s empathizing–systemizing theory (PDF of slides). He suggests that women tend towards identifying a person’s thoughts and feelings while men tend towards analyzing or building systems (mechanical, natural, abstract, collectible) and that this is at least partially a biological difference (caused by differences in testosterone exposure).

Are there any preferences that can explain (partially) gender differences in leadership roles and average salary? Yes: mothers are much more likely than fathers to prefer to stay home with their children, which puts career development on hold.

I see nothing fundamentally wrong with encouraging men and women to consider different options and interests, but we can’t ignore when differences exist.

Read on for a closer look at the evidence.


  1. Effect Sizes
  2. Job Attribute Preferences
  3. Interest Areas
  4. Stay-at-Home Parenthood

(Length: 900 words)

1. Effect Sizes

If you take a random man and a random woman, there’s a very high (92%) chance that the man will be taller. That’s a very large gender difference in height. If this chance was 50% it would mean they both have an equal (50:50) chance of being taller, which would indicate no gender difference. As a simple rule of thumb, let’s say that 50-60% is a small difference, 60-70% is a moderate difference, and above 70% is a large difference.

(This percentage is called CL, or common language effect size. I’m calculating it from Cohen’s d statistic provided in the meta-analyses, using this tool from R Psychologist.)

2. Job Attribute Preferences

The first meta-analysis looks at men’s and women’s preferences for different qualities and outcomes in their work. It surveys 242 previous studies, covering 600,000+ Americans, 1970 to 1998. Of the 40 job attributes looked at, 33 had gender differences, but the effect sizes were generally quite small.

Surprisingly, men were only a bit more likely to care about earnings (and women a bit more likely to care about benefits). The (relatively) larger and more notable differences (as pointed out by the authors) involved women being more socially-oriented.

Here are some of the findings. Note that the 53% by earnings does not mean that men care 53% more—it means that given a random man and a random woman, there’s a 53% chance that the man cares more (and a 47% chance that the woman does).

Attribute Cares more CL Size
Power/influence/authority Men 51% Small
Earnings Men 53% Small
Benefits Women 53% Small
Leadership role Men 54% Small
Physical work conditions Women 54% Small
Feeling of accomplishment Women 54% Small
Opportunity to make friends Women 56% Small
Leisure time off job Men 57% Small
Solitude (n/s: small sample size) Men 57% Small
Easy commute (n/s: small sample size) Women 58% Small
Opportunity to help others Women 60% Moderate
Working with people Women 60% Moderate

3. Interest Areas

The second meta-analysis looks at men’s and women’s broad vocational interests. It includes 47 inventories published between 1964 and 2007 with a total of 81 samples consisting of ~500,000 (American and/or Canadian) men and women.

This study uses a few different ways to measure interest areas and they overlap with each other. It looks at two dimensions (things–people and data–ideas), six RIASEC types (widely-used vocational choice categories), and three STEM interests.

Their results both support and expand on those from the previous meta-analysis. They find only small gender differences in the interest category (enterprising) that includes leadership roles and economic objectives (in line with the minimal gender difference in desire of influence/authority and leadership roles from the previous meta-analysis).

They find quite large differences in preferring to work with things rather than people. This backs up the finding from the previous meta-analysis of women being more socially-oriented (although these effect sizes are larger and thus even more notable).

Here are all of the findings. Given a random man and a random woman, there is a 74% chance that the man will be more things-oriented than the woman. This difference is highly visible in other categories as well (realistic, social, and engineering).

Attribute Higher CL Size
Things (vs. people) Men 74% Large
Data (vs. ideas) Women 53% Small
      RIASEC types
Realistic (things, gadgets, outdoors) Men 72% Large
Investigative (science/math/biology/medicine) Men 57% Small
Artistic (creative expression) Women 60% Moderate
Social (helping people) Women 68% Moderate
Enterprising (leadership, economic objectives) Men 51% Small
Conventional (well-structured environments, e.g. business) Women 59% Small
      STEM interests
Science Men 60% Moderate
Mathematics Men 60% Moderate
Engineering Men 78% Large

4. Stay-at-Home Parenthood

According to a 2015 Gallup poll, almost 6 in 10 mothers would prefer to stay home with their children, compared to almost 3 in 10 fathers. Even for couples without children, almost 4 in 10 women prefer to stay home (compared to over 2 in 10 men).