Figureing out N's

#1
Hi

I'm measuring the physiological distance that certain axons retract after injury across different injury interval groups. I have approx 5 animals per injury interval, and approximately 100 measurements per animal

Should the N that I use for my ANOVA be 5 for each group or 500. meaning should I use all of the measurments seperately as my raw data, or should I get the mean of each animal and then plug those mean values for my raw data when doing the ANOVA?
 
#2
Welcome gseif,

It seems that the overall sample size is 100*5*the number of injury interval groups. As described, it is a randomized block design with injury as the between factor and animal as the block. 100 measurements is quite a large number, are you applying any treatment to the different injury groups?
 
#3
Thank you quark.

Actually, the injury is the same across all goups. No treatment is given. the only difference between the groups is the length of time of the injury.

5 groups are as such.

1: 1 week injury interval
2: 2 week injury
3: 4 weeK injury
4: 8 week injury
5: 16 week injury

I'm trying to quantify the degeneration that occurs after progressively longer injury intervals.

thanks, I really appreciate any advice that you can give
 
Last edited:
#4
You are welcome gseif. If you have five injury group then total N=100*5*5=2500. I think the experiment is a randomized block design with injury time as the between factor and animal as the block. There are five blocks in each group, and 100 repetitions for each block.