Hi everybody,
I'm looking at stomach content of a damselfish, to analyze the diet in two sites. I'm going to use the numerical proportion of the prey items that I found in each stomach. In each site I captured 15 fishes. I know that is more adequate to use the weight of preys, but I haven't access to an analytical balance. The numerical proportions are the total number of a prey divided by the total number of all the preys in one stomach multiplied by one hundred. I want to know which statistical analysis could be employed to compare the sites.
My experience in fly fishing with stomach contents data is that descriptive statistics is all that is required. Best, D. Booth
One approach would be to pool the data from all 15 fish in site1, and all the fish in site 2 and get a distribution of prey type for each site;
eg site 1 prey A 15% prey B 50% prey C 35%
site 2 prey A 30% prey B 40% prey C 30%
then do a chi sq on the 2 x 3 table to see if distributions differ significantly between sites.
Dear Colleague
if your objective is just to compare sites then you can use a t test, if the data you finanally use have a normal distribution, otherwise you can use a non parametric test. if your objective is to get more info on prey features you can follow James Leigh advice.
Regards,
Redimio
- Badges
- Science topic
- Similar topics
- Medicine
- Physiology
- Biosignals
- Biomedical Signal Processing
- Statistical Analysis