Men And Women: Similarities Or Differences?

Note to my readers: Sorry for the long hiatus. My life has been very hectic as of recent, jumping from one job to another, getting a promotion at the latter, quitting due to a lack of any freedom or respect, then picking up yet another job which currently occupies my time. I originally had something much bigger planned for this post, but time restrictions prevent it from coming to fruition. I believe that the following post is nonetheless informative, and I hope you all enjoy it.

It's a question that many people struggle with and has great implications for the study of our species: are men and women more alike than different or more different than alike, and what differences exist between the sexes?

It's a tough question, really, and is influenced by many different factors. I've reviewed these types of dilemmas several times. Despite the suggestions of recent studies, the gender gap in spatial abilities cannot be solidly attributed to an evolutionary perspective because, in short, there are many clear ways in which culture and society influence the manifestation of our spatial abilities, regardless of gender. Aside from that, it is shown that in more egalitarian societies, the spatial ability gap decreases. It's a perfect example of how culture and environment can influence psychological traits in a way that may automatically be assumed to be a product of sexual selection.

Briefly, in what I dubbed the "ROK Complex," I reviewed definitions of attractiveness by gender, and ultimately (and most relevantly) how well human sexes adhere to sex roles commonly found in nature in accordance with Bateman's Principle. As it turns out, what little success there has been in analyzing this has brought some questions to the idea that humans follow Bateman's Principle, although the study cited only examined 18 human populations. This is possibly the most straightforward answer to whether or not sex roles and major gender differences manifest greatly in humans.

Nonetheless, stereotypes of males and females still find popularity even within the scientific community. Sexual attitudes are a big topic in the sciences, as is the psychology behind men and women's preferences in sexual partners. While refuted in that post, the answer was (again) not conclusively drawn, and still did not get to the heart of the question at hand. Whatever piecemeal debates we can examine and draw conclusions about, there is a looming question that overbears it all: overall, are men and women more similar than alike? Overall, what differences exist? My posts so far haven't examined this, and it's because it's a question I really didn't have the answer to.

I was curious what the literature had to say on the matter. I found a research report on the APA's website that referred me to some great pieces. Namely I was introduced to researcher Janet Shibley Hyde, a leading scholar in her field, and her gender similarities hypothesis. The gender similarities hypothesis states that men and women will be similar in most (but not necessarily all) psychological variables, which is to be contrasted to the differences model which states the opposite. This, while more socially favorable, has not found much favor among psychologists, simply because it seems intuitively ridiculous. Hyde recognizes this and that there are many pervasive myths surrounding the idea of gender differences in various behavioral and cognitive faculties, and so set out to see what the body of scientific knowledge we had available at the time (2005) actually revealed.

Thus came the famous study "The Gender Similarities Hypothesis" where Hyde reviewed 46 meta-analyses on gender comparisons in various psychological/physiological faculties. This study revealed quite an array of surprising results, which I'll briefly review below:

• On the vast majority of the examined variables, the effect sizes of the gender differences examined were small or close to zero (d ≤ 0.35).
• Of these, 30% were close to zero (d ≤ 0.10) and 48% were small (0.11 < d < 0.35).
• Those faculties that show larger differences are throwing distance/velocity (explained by large muscle mass and bone density in males), some sexual attitudes, and aggression.

But the results aren't so simplistic either. Social context was taken into account in many of these studies as well. Surprisingly for aggression was just how much it may rely on societal expectations -- when gender identities of test subjects were hidden, women actually displayed more aggressive tendencies than men. In addition to this, many of the gender gaps fluctuated over time, showing that a consistent, biological explanation doesn't conform well to some of the data. All in all, the big takeaway from this groundbreaking review was that, contrary to popular belief, men and women are much more similar to each other than they are different.

When this study was published, researchers went nuts, and who's to blame them? It's not surprising that so many people would be rushing to the lab setting after hearing that those sex differences they thought were just common sense turned out to be unsupported by the data, or at least fairly ambigious. A slew of meta-analyses were published in the years following, further increasing the pile of research available to accept the plausibility of the gender similarities hypothesis versus the gender differences hypothesis. When it seemed like the debate may have been coming to a close, instead it raged on hotter than ever.

So it was no surprise when in 2014, Hyde released another review "Gender Similarities and Differences" which revealed results that were about the same as before, this time calling upon a much larger pool of research with a much more detailed review, in which they not only offer explanations from their previous study, but offer new insights to address the issue of gender/sex differences now and reconcile previous study design flaws.. The article is not available to the general public, however I can provide it by request. There's also far too much to review in this post, so instead I'll provide some of the highlights that I believe people are most curious about, so long as I haven't previously covered them.

Links are provided for all studies. Scores in the positive range mean that boys scored higher than girls, and scores in the negative range mean girls scored higher than boys for the tested factor. Any questions about sampling or study design can be directed to me in the comments or via email.

Lastly, I should remind everyone that these studies are all meta-analyses, meaning they each draw upon a large pool of research. Even where a factor is only examined by one study, the power of the study is found in its meta-analytic review.

Mathematical Performance

4 studies were reviewed here: Hyde et al. (1990), Hyde et al. (2008), Lindberg et al. (2010) and Else-Quest et al. (2010). The first meta-analysis from 1990 revealed that although boys and girls did not differ significantly in mathematical performance (d = -0.05), there was still a substantial gap in complex problem solving in high school (d = 0.29). So they reviewed again in 2008 and found that not only did the gap in mathematical performance not get wider, but the gap in complex problem solving appeared to have closed as well (d = 0.06). Lindberg et al. found again that the gap in mathematical performance was almost negligible, and there was a small but significant difference in complex problem solving (d = 0.16). While the results of the final meta-analysis are not given much detail, it is noted that the gender gap fluctuates both in magnitude and directionality across nations. In short, the gender difference in general mathematical performance has reached parity, and while there still may be a gap in complex problem solving, it seems to be declining (or perhaps has even closed entirely) and could be caused by sociocultural elements.

It should also be noted, bearing these data in mind, that Else-Quest et al. revealed that the gender difference in mathematical self-confidence is 0.27, and the difference in anxiety is -0.23. As Hyde notes, this means that the difference in how males and females see their own mathematical performance is higher than the difference in their actual performance. What is especially interesting about this information is that, at least concerning mathematical performance, it effectively eliminates an explanation via stereotype threat. This isn't intuitive, and it may actually surprise some people.

Temperament

This section covers three factors of temperament: effortful control (i.e. inhibitory control, attention), negative affect (i.e. emotionality, fear) and surgency (i.e. activity, impulsivity). Data includes a sample of 236,102 temperament ratings for children 3 months to 13 years of age (Else-Quest et al. 2006).

For effortful control, girls scored higher for both inhibitory control and attention (d = -0.41; -0.23, respectively). Hyde notes that "These average gender differences are in the small-to-
moderate range and contrast to larger gender differences at the tail of the distribution, where
boys with attention deficit–hyperactivity disorder (ADHD) outnumber girls by ratios of 2:1 to
9:1 across studies" (p. 383). That being said, one should not take the extremes at either tail of the distribution and assume those to be the norm. There is much more substantial overlap between genders in this category.

For surgency, boys scored higher in both areas. Depending on the measure, gender differences in activity ranged from 0.15 to 0.33, and 0.18 for impulsivity. Again, a small-to-moderate disparity, but nonetheless significant.

Negative affectivity is probably the area of greatest interest, and has the most surprising results. No significant gender difference was detected in any factor of negative affect (d = -0.06): -0.1 for sadness and 0.01 for emotionality. While it's rather intuitive to believe that women are more emotional than men, the data doesn't show this to be the case, even though there is a rather robust difference in depression between the genders.

What's also important to note is that gender differences in these factors varied with age (typically increasing with age), but also varied depending on context. While the gender difference in emotional internalization were essentially nonexistent when children were by themselves, it increased to -0.16 when adults were present.

So while temperament shows significant gender differences, these differences are much weaker than expected, and the factor commonly believed to exhibit the largest disparity in fact had the smallest. Considering that studies which controlled for social expectations had larger effect sizes, this is completely consistent with the conclusions drawn in Hyde et al. (2005). Finally, once again context can completely alter the magnitude and directionality of gender differences in these variables.

Emotions

This can definitely be a controversial discussion. Are women more emotional than men by nature, or even just a matter of statistical trends? Emotional disparity in particular motivates many gender stereotypes and social expectations, where it tends to be the case that men can more easily and acceptably express anger, while women can express most other emotions.

Only two studies are reviewed here despite the interest of the subject, however I believe this is the case because studies on emotional expression suffer from an inherent setback of the Hawthorne effect. This is where a test subject modifies their behavior while being observed because they believe it is more acceptable or desirable. However, what's interesting is that one of the studies directly examines this effect and what it does for the gender disparity in emotional expression. The results are quite interesting.

The two studies reviewed are Chaplin & Aldao (2013) and Else-Quest et al. (2012). In the latter study, it was found that while women were more likely to experience guilt (d = -0.27) and shame (d = -0.29), there were very minor differences for embarrassment (d = -0.08), authentic pride (d = -0.01), and hubristic pride (d = -0.09). What's interesting is that these disparities, while being small and trivial, actually show women displaying more pride contrary to stereotypes.

In the former study, gender differences in emotional expression were examined from birth to adolescence. Surprisingly, across the board gender differences in the expression of positive emotions (d = -0.08), internalizing emotions (d = -0.1) and externalizing emotions (d = 0.09) were all fairly small, although the trends seemed to be consistent with the relevant stereotypes (e.g. internalizing emotions are those such as sadness and anxiety, externalizing emotions are those such as anger). What was most interesting about this study however is that it showed that these differences vary by age, increasing gradually from early childhood to mid-childhood and into adolescence. This appears to be inconsistent with a more biological hypothesis as one would expect that if the differences were a result of biology, they would jump at puberty and into adolescence. If it were a matter of gene expression, the same might be expected. The trend here seems to be more consistent with social learning theory.

Lastly, in the same study, it was found that emotional expression varied greatly depending on whether or not the children were with adults. For example, for internalizing emotions, the disparity was almost non-existent (d = -0.03), but jumped when adults were present (d = -0.16). Again, this shows the importance of context and how social situations tend to increase, or in some cases may be completely responsible for, the gender differences we tend to observe.

Aggression

Again, another difficult factor to examine longitudinally. There were 3 studies available for examination, however two of them were from the 1980s, so only one is focused on (Archer 2004) with additional information from Bettencourt & Kernahan (1997). It should be noted that two of these studies are fairly old, but still important for the discussion.

Gender differences in physical aggression were fairly substantial (d = 0.55), and this trend appears very early -- as young as two years old among children playing together. Yet while men hold the stereotype of being more physically aggressive on average, women hold a stereotype of more "relational aggression;" that is, non-physical aggression which seeks to harm peer reputation. According to Archer, however, such a trend is not very large: -0.19 for peer reports and -0.13 for teacher reports.

But what of context? It seems to be the case that for most of these factors, context plays a key role. The same is true of aggression. According to Bettencourt & Kernahan, when violent cues are present but absent of any provocation, the gender difference in aggression is still 0.41, however when both provocation and violent cues are present, the difference is near-zero. So while men are typically more aggressive than women, and while women are slightly more indirectly aggressive that men, the difference can be mitigated by context.

Sexuality

Lastly, the big "S." How do men and women differ in terms of sexuality and sexual conduct? This includes things such as number of partners, cheating, casual sex, etc. Across a total of 14 sexual behaviors and 16 sexual attitudes examined by Petersen & Hyde (2010), almost all gender differences were small. Four were in the moderate range: Masturbation (d = 0.53), pornography (d = 0.63), number of sexual partners (d = 0.36), and favorable attitudes towards casual sex (d = 0.45). For five factors, differences were ≤ 0.10, including oral sex, attitudes about condom use, attitudes about masturbation, attitudes about extramarital sex and attitudes about lesbians. Interestingly, the difference favored women slightly for attitudes about gay men (d = -0.18). Some of these gender gaps are closing, as revealed by moderator analysis. It's also quite possibly the case that some of the larger differences were affected by reporting bias, since gender differences shrink when the respondent is given anonymity, and nearly close when they are hooked up to a fake polygraph.

Much more was covered by Hyde's review, but it would take too long to discuss them all here. Here is a list, however, of all factors examined:

- Spatial performance.
- Verbal skills.
- Attitudes.
- Personality and "the five factors" (i.e. The Big Five).
- Impulsivity.
- Interests.
- Communication.
- Helping behavior.
- Leadership.
- Depression.
- Rumination.
- Self-esteem.
- STEM.

If there is any one in particular that you would like me to summarize in the comments section, I'd be happy to take requests. Also, as stated before, I can offer a copy of the review upon request as well.

In terms of final thoughts and opinions, I'd say it's not that important what the data says when it comes to a real world understanding of gender differences in psychology and behavior. "Seeing is believing" as they say, and how people tend to act on average is much more important from a practical standpoint than what the causes of it are. Even if we were to say "context is everything" as a general rule, that doesn't say how much context exacerbates gender disparities in the listed variables. In the end, it's what we observe, not why.

Yet it stands true: We are more alike than different.

Thank you all for reading.


References:


Hyde, J. (2014). Gender Similarities and Differences Annual Review of Psychology, 65 (1), 373-398 DOI: 10.1146/annurev-psych-010213-115057