Sunday, 8 September 2013

What is the difference in strength between men and women? (In terms of maximum muscular strength)


Question: If an average man competes with a hundred random women in muscle strength tests - how many from that hundred would he expect to beat? 

Answer: All of them.

- Because only about one woman per thousand is stronger than the average man.

(Subject to caveats below)


Until a few days ago, I did not realize that the difference between strength in men and women was quite so extreme, so qualitative.

I came across this fact referenced in a review-theory paper about sexual selection: "...less than 10% overlap between the male and female distributions, with 99.9% of females falling below the male mean."

WD Lassek, SJC Gaulin. Costs and benefits of fat-free muscle mass in men: relationship to mating success, dietary requirements, and native immunity. Evolution and Human Behaviour 2009; 30: 322-328.


I checked out their main references. This paper was the most striking: 

RW Bohannon. Reference values for extremity muscle strength obtained by hand-held dynamometry from adults aged 20 to 79 years  Archives of Physical Medicine and Rehabilitation  1997; 78: 26-32.

This has tables of muscle strength, with force expressed in Newtons (N) for different movements, men and women, and different age bands, dominant versus non-dominant side.

If we look at the results for dominant arm elbow flexion ('biceps' strength, more or less) - mean force in Newtons plus standard deviation in brackets - we find:

Men        age 20-29 -  285 (38) 
Men        age 70-79 -  237 (40)

Women age 20-29 -   155 (21)
Women age 70-79 -   130 (27)

And similar results are found for other muscle groups.

What is striking is that at 20-29 the difference in average strength is 285 - 155, which is 130 Newtons difference - or that, in round numbers, men are nearly twice as strong as women. (ie. 130 is nearly as big as 155)

And the standard deviation for women is just 21 - which means that there are about six standard deviations difference, which is a huge difference and means near zero chance of overlap in strengths between men and women.

Indeed, the Men aged 70-79 were much stronger than the women aged 20-29: 237 - 155 = 82 Newtons, which means old men are still roughly half as strong again as women (i.e. 82 is about half of 155) - or young women are only about two thirds as strong as old men.


Now, these numbers are presented as reference values; but like almost all real-life reference values they are not truly representative of the general population, because they are (and this is clearly acknowledged in the methods section) based on a 'convenience sample' of 106 men and 125 women selected on the basis of different ages. And these subjects are not athletes in training, body-builders, or anything of that sort.

Since subjects had to go to a lab and perform strenuous tests - this sample would include only volunteers; and exclude those too ill to come to the lab or too lazy to make the exertions.

This could be significant - for example, a high proportion of men aged 70-79 would be unable to do these tests (being too ill, demented, institutionalized, or whatever) - so the reference value at best refers to that subset of ambulant old men able and willing to do these lab tests.


Nonetheless, the take-home message is clear: the statement "men are much stronger than women" is not just true on average, but is close to being true as a generalization (at least among un-trained subjects).

I have tried this fact out on a few people, men and women, and most - like me - were surprised at the high degree of difference in strength between men and women and especially the lack of overlap.

Since the average civilized men and women do not compete physically at full strength in the course of everyday life, and seldom go head to head on the kind of objective measures of strength being used here, the scale of sex differences in strength is consistently under-estimated - probably due to the factors discussed here:



Wm Jas said...

I'm surprised to discover this, too!

(By the way, I think there's a typo in the original post. You wrote "men aged 20-79" instead of "70-79.")

Bruce Charlton said...

@WmJas - thanks, typo fixed.

dearieme said...

Compared to one girl in her twenties I was stronger, had a greater reach, had much quicker reflexes, more speed and more stamina. Yet she could give me a decent game of squash. Reader, I married her.

Anonymous said...

Regarding trained athletes, wikipedia says:

Aggregated data of absolute
strength indicates that women have 40-60% the upper body strength of men, and 70-75% the lower body strength.[56] The difference in strength relative to body mass is less pronounced in trained individuals.
A study, carried about by analysing annual world rankings from 1980–1996, found that men's running times were roughly 11% faster than women's.

The difference in average is 22%.

Mr. Sailer's work on how women get more bang for buck by using steroids would explain the discrepancy.

Some data on east german athletes.

Sylvie D. Rousseau said...

...most - like me - were surprised at the high degree of difference in strength between men and women and especially the lack of overlap.

I come from a family of taller and more muscular than average men and women and I have always observed that the overlap is only between fairly above-average women and considerably below-average men. I admit I was a bit of a tomboy but, apart from those who never wrestled with members of the other sex, logically most people should not find surprising the lack of overlap. Or is wrestling between boys and girls so rare?

Also, I would like to know the overall percentage of the overlap. Does some study mention the percentile of healthy women whose natural strength is greater than healthy men, and what percentile of men (both groups excluding athletes and body-builders)?

Bruce Charlton said...

@SDR - You must have missed the quote above saying 10 percent overlap - but I would not take this too seriously. This is the kind of data we just cannot ever know, because it is in practice impossible to do representative population studies; also the numbers might well differ over time, in different circumstances - eg related to nutrition and disease, and between races.

Sylvie D. Rousseau said...

Thanks. I missed the quote. I did not read attentively enough. An overlap of no more than 10 percent was my guess. And I was thinking of course about quite homogeneous groups.

In a totally unrelated subject, I just read David Warren's last article on not fighting big lies (in that case neo-Darwinism). You might like it:

as said...

I wonder how men and women compare in terms of intelligence.

I know IQ tests show men and women to be about the same. But in terms of using intelligence to solve real life problems, men and women seem different.

Bruce Charlton said...

@as - see comments to -

Adam G. said...

Kingsley Browne's book Women in Combat was an eye-opener.

Men and women are *different.*