Effect Size with meta() for non-normal data

[ross.co...@gmail.com]

Hi Chris,

Home range size in my dataset of males, females, and juveniles are quite skewed and their medians are a more meaningful measure of central tendency. Since effect size with the meta() call compares means, I wanted to ask if there is still an appropriate way for me compare groups and calculate effect size with meta(). Much appreciated.

Ross

[Christen Fleming]

Hi Ross,

Is the distribution skewed because you are grouping males, females, adults, and juveniles together, or because there are some dispersers?

I haven't derived any robust statistics that propagate uncertainty, but you might consider something simple like a trimmed mean.

Best,

Chris