I have a data set in which I recorded cricket calling songs of four
populations that are divided into two zones, allopatric zone and
sympatric zone. In addition to calling song characters, I also noted
age and temperature to adjust for calling song characters. My goal of
these analyses is to see whether calling song characters differ
between two zones after adjusting age and temperature.
The followings are the structure of my data set:
dependent variables: pro, cro, pdo, cfo, pno, ...
fixed factors: zone(1,2) and pop(1,2,3,4) (pop is nested within zone)
covariates: age and temperature
So I ran the following syntax:
pro cro pdo cfo pno BY pop zone WITH age temperature
/METHOD = SSTYPE(3)
/INTERCEPT = INCLUDE
/CRITERIA = ALPHA(.05)
/DESIGN = age temperature pop zone pop(zone).
My questions to everybody is:
1. Is this syntax right?
2. How do I test interactions between covariates and factors? Should I
include "age*zone" and "age*temperature" in the design? Or any other
3. The results of the above GLM showed that age did not affect any of
calling song characters, but temperature affected several calling song
characters. Then can I exclude age from the analysis?
4. How do I test assumptions of analysis of covariance including
homogeniety of slopes?
Thank you for your helps!