Hi everyone,

I have been revisiting and re-analyzing my master's data on aggression between males in multi-male groups of colobus monkeys, in the hopes of publishing it soon. A question that I had encountered during my masters analysis has re-surfaced, and I'm hoping to gain some insight. One of the results of my masters was that rank-distance between 2 males (diff in elo rating) did not influence their dyadic aggression rates. However, just from watching the monkeys for months, this didn't seem right. Some of my groups were very large (at one point, one group had 7 adult males!), and it was clear that aggression was concentrated among the top 2-3 males in that group. As such, my gut feeling is that rank-distance is not uniformly important (males ranked 6th and 7th probably have relatively little to squabble about), but is very important for males near the top. So I had been wanting to create a variable that would allow me to test this. I had proposed average elo rating of dyad members to a few people, but they didn't like the idea; the top dyad will have the highest avg rank, and the lowest will have the lowest, but in the middle things get fuzzy as avg of elo500+elo300 is the same value as elo700+elo100 etc. However, this is the best solution I can come up with to capture what I'm after. And I think overall, it could be informative. However, I've not seen it used in any of the primate literature (though I have seen a paper on social cichlids that used a similar method; adding size ranks for dyads together *), and I've not encountered any other way in the primate literature to capture what I'm after. I'm envisioning including an interaction term between rank distance and avg rank in a model, to address whether rank distance is more salient for higher-ranked dyads (it has just occurred to me; would this interaction then rescue it from the fuzziness described above? 500+300 and 700+100 would have same avg rank, but very different rank distances).

So my questions are:

-Does this seem like a valid measure of relatively where dyads/dyad members are in the hierarchy?

-If not, do you know of, or can you think of, another way to capture this?

Thank you for any insight any of you can provide!

-Luke

* Dey, C. J., Reddon, A. R., O'Connor, C. M., & Balshine, S. (2013). Network structure is related to social conflict in a cooperatively breeding fish. Animal Behaviour, 85(2), 395-402.