Nested ANOVA Experimental Design Help

#1
Hi,
This is my first time posting here so apologies if my post is not appropriate.

I am a masters student doing a project on evaluation of conditioning methods to prepare farmed abalone for the stress of live export. I am looking at whether holding time duration and the ozone levels of the tanks in which abalone are held prior to being exported effects their stress levels, which I measure through various procedures.

So I have 3 factors:

Factor A: Transport (Yes, No) 2 levels
Factor B: Duration (0, 7, 14, 21 days) 4 levels
Factor C: Ozone levels (200, 300, 400 mV) 3 levels with 2 replicates for each level

Set up:
I have access to 6 tanks, each with adjustable ozone levels and containing 1000+ abalone which I can sample at any time.

On each measuring day (0, 7, 14 and 21 days) I take 4 abalone out of each tank, two of which are measured immediately, the other two of which are put through a transport procedure and then measured.


Assuming the data obtained from my measuring procedures has a normal distribution:
Nested ANOVA- Is this an appropriate analysis for my experiment?

Transport: Factor A 1 d.f
Duration: Factor B(A) 3 d.f
Ozone: Factor C(B(A)) 5 d.f
Residuals 1425 d.f? (1*3*5(n-1)) = 15(95) = 1425????

If there is no interaction between the factors, should I pool the main effects for each factor?
So Transport: 48 obs (abalone measured) for each category
Duration: 24
Ozone: 32

Any help would be greatly appreciated.
 

Masteras

TS Contributor
#2
two levels in factor A and 4 different days of measurement.
And then factor C has 3 levels. So, for each combination you have 2 measurements or I missed it?
So you say Day 0, factor A level 1 and factor B level 1 2 measurements, factor A level 2 and factor B level 1 2 measurements and so on for day ). And then the same for day 7, day 14 and day 21?
 
#3
No, I want to treat the days of measurement as a factor, the length of time abalone are being held for.
Thanks anyway, I think the project might be too complicated with too many factors.