3
$\begingroup$

Is there a standard, or most accepted, way to present a value for sexual dimorphism? For example, one could use

$$\frac{\text{Female value}}{\text{Male value}} -1$$

This would give negative values when the female value is smaller than the males, and positive when male is smaller. My issue with this though is that when a female is twice the size of a male its value is -0.5, while when a male is twice the size of a female its value is 1.0, so they do not spread symmetrically around 0.

Edit:

Just playing around with some dummy data in R I came up with a possible solution

$$0.5 - \frac{\text{Female}}{\text{Female + Male}}$$

This gives a zero centred value with equal distribution around the mean. For example, when Females are twice the size of males the value is -0.166, and when males are twice the size of females the value is 0.166, when they are the same size it is 0.

enter image description here

$\endgroup$
  • 1
    $\begingroup$ I think you should consider something parametric, that takes the entire trait distribution into account, since there can probably be quite large intrasex variation (at least in some species). For instance, looking at distribution overlap between sexes. $\endgroup$ – fileunderwater Feb 3 '15 at 11:25
  • 1
    $\begingroup$ Oh, and have you seen the Wikipedia page Sexual dimorphism measures? It should give you a couple of ideas. $\endgroup$ – fileunderwater Feb 3 '15 at 11:27

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.