1
$\begingroup$

Wikipedia tells me that men have 7 to 8 time higher levels of testosterone than women, on average.

Since the hormone is linked with performance in certain athletic events, my hypothesis, which I presented to a disagreeing friend, is that female top athletes in certain fields will have levels above the average male.

I found Why do endocrine profiles in elite athletes differ between sports? (Clinical Diabetes and Endocrinology 2018) which in Figure 10 presented testosterone levels for men and women in a number of olympic athletes: enter image description here

As you can see in the plot it seems that even in the cohort of olympians there is an overlap, even if small, between men and women. Am I misinterpreting these result? I noticed that the units (nmol/L) are different from those used on Wikipedia (ng/dL), is this somehow important?

So:

  1. Are there information about the distribution of testosterone levels among the general population to help answer the question of the share of women exceeding the average male level?

  2. Am I interpreting the above results correct, are there indeed female olympians with higher testosterone levels than male olympians?

(The mapping between numbers and sport)

  1. Power Lifting (18 M and 1 W)
  2. Basketball (27 M and 14 W)
  3. Football (Soccer; 37 M)
  4. Swimming (100 M and 91 W)
  5. Marathon (1 W)
  6. Canoeing (7 M and 1 W)
  7. Rowing (36 M and 25 W)
  8. Cross Country Skiing (8 m and 9 W)
  9. Alpine Skiing (11 M and 12 W)
  10. Weight Lifting (10 M and 7 W)
  11. Judo (26 M)
  12. Bandy (19 M)
  13. Ice Hockey (38 M)
  14. Handball (23 M and 29 W)
  15. Track and Field (95 M and 49 W)
Improve this question
$\endgroup$
4
  • 1
    $\begingroup$ Did you generate these plots yourself? If not it's good to cite the source. Anyway, if you have the data you can perform a statistical test to see if the differences are significant. Did you try any? Try Mann-Whitney test or student t-test and then you can conclude if the overlap you see is indeed significant. $\endgroup$
    – WYSIWYG
    Commented May 13, 2019 at 11:42
  • $\begingroup$ @WYSIWYG: I attributed the sources, but perhaps not clearly enough (see paragraph before the plot). Thank you for your suggestion, but I think it would be better done with a larger dataset of the general population, this is only for olympic level athletes. $\endgroup$
    – Filip Nilsson
    Commented May 13, 2019 at 12:11
  • 1
    $\begingroup$ Did the authors of the paper discuss the possibility that some of the athletes misused testosterone for doping and thus the results might not reflect the "natural" testosterone levels? $\endgroup$
    – Arsak
    Commented May 13, 2019 at 12:20
  • 1
    $\begingroup$ As you can clearly see from the data, some women athletes have higher testosterone levels compared to some men even in the same sport. However, this does not negate the statement that a random (or average) female athlete is unlikely to have higher testosterone than a random (or average) male athlete (you can perform a statistical test to check). $\endgroup$
    – WYSIWYG
    Commented May 13, 2019 at 14:11

1 Answer 1

Reset to default
0
$\begingroup$

To the first question we can look at a paper such as "Estimating age-specific trends in circulating testosterone and sex hormone-binding globulin in males and females across the lifespan" from Handelsman et al. The figures they provide are:

Female Male

The top image is of females, the bottom of males. We can see throughout a lifetime that the top 2.5% of females and bottom 2.5% of males seem to have approximately the same amounts of testosterone. If we went to the top and bottom 1% there would likely be consistent overlap between the male and female distributions.

I believe you are interpreting the figures correct, although it is important to keep in mind these figures do not include any information on age and are a single point measurement. To facts that may limit the depth of inference you can make about testosterone levels among the sexes.

Improve this answer
$\endgroup$

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .