Working with zoo closed population - observation study - Which test?

LizeMaeve

New Member
Our title: Investigating the anticipatory and aggressive behaviour of a captive group of Sulawesi Crested Macaques (Macaca nigra) during different feeding times and enclosure location.

Confused as to what stats test to use as
> we have 3 independent variables (explanatory as observation)
1) Feeding event: feed, pre-feed, post-feed and control
2) Individual: 8 identified macaques
3) Inside/Outside: where they observed inside or outside

Basically, we want to look at all 3 factors including individuals and have played around with doing a combo of 3 separate 2-way anovas, a mixed model Anova, but a lot of our data is non-normal and quite a small sample size.

As its a small behaviour study and the zoo is looking for recommendations we really want the individual responses to have focus hence why we don't think its really a nested study for example

I am just confused with the whole repeated measures, paired tests etc. what does our data fall into? When researching there isn't a lot on zoo statistics and what to do when you're just looking at the same individuals in the same enclosure so it's hard to apply other examples such as medical or psychology examples.

Another problem we have is we don't have the same number of observations in each of the factors e.g. we don't have the same number of observations for pre-feeding as we do feeding due to how the study ended up occurring

Any advice would be great, the stats platform I usually use is R but again it's confusing as there's so many variations to ANOVAs and GLMs that we just don't know which one is applicable and how to apply given our non-normal data,

Ideally we want to look at whether each of the factors appear to influence the response variable and to make sure none of the factors are having a combined effect hence why we were trying the series of two-way ANOVAs, however we could not get our mixed model ANOVA to work as when we looked at the coefficients some of our factors had been removed e.g. some of the individuals have been taken out and we do not know why.

Here is our data in our table. The response variable is the time (sec) displaying anticipatory behaviour in a 5 min period, recorded every 15mins over 4 days

Thank you so much