Need help choosing a statistical method/model

#1
Hello,

I am currently doing a study on the defensive behavior of the European Grass Snake, and I was hoping someone could help me figure out which model to use to analyze my data.

I have 3 categories of dependent variables - 1) Pre-Capture Behavior 2) Post-Capture Behavior 3) Post-Release Behavior. Within these groups there are 7 different antipredator tactics that were measured (y/n)- these include Flee, Stay, DF, Writhe, Head Triangulation/Flattened Body, Hiss, and Strike.

**Note: At first I ranked these 7 behaviors upon 'intensity' - giving passive values a negative score and active values a positive score, and since snakes exhibited more than one behavior I added the values up for each category (Pre, Post, Release) and gave them a final behavioral score. However, I think it would be more insightful to rank each behavior as y/n (say, in a logistic regression) because there are many ways to rank the 'intensity' of behaviors in snakes and I think I could face issues of subjective influence when presenting my data in this way.**

I have a multitude of independent variables- Date, Time, Location, Temperature, Body Mass, SVL, Gender, Fed (y/n), Afflictions (y/n), Mating (y/n)... etc..

I have been reading many papers all of which seem to use a logistic regression as a common method, however none of them have the data organized into the 3 categories which they scored the behavior under...So I'm not sure if this is the best method.

What correlation matrix should I use? I was thinking Spearmans...And what type of statistical test should I use? Logistic Regression? Nested Anova? Any help would be great!

THANK YOU IN ADVANCE!!
 

Karabiner

TS Contributor
#2
Well, you could perhaps first tell us your research questions.
And your research design, and your sample size.

With kind regards

k.