comparing 2 groups (4 repetitions) - correct stat?

#1
Dear all,

I'm currently having a lot of trouble finding the adequate statistical analysis for my current experimental data.

The experiment:
I work a chick embryonic spinal cords. One side of the spinal cord is treated while the other side serves as a negative control. After treatment I dissect the spinal cord, separate both sides (the control and treated) and then cut each side into small explants that I culture for 2 days. After that I measure neurite outgrowth.

I obtain 1 neurite outgrowth value per explant. That gives me about 5 values per spinal cord side and I want to compare the treated side to the control side. As such I suppose the stat I should use is simply a T-test. However this experiment was repeated 4 times so expect this might actually be a 2 way-ANOVA (my variable being the measure of neurite outgrowth and my factors being the (treated and control) sides and the repetitions. Does that seem correct to you guys?

Any help would be greatly appreciated!

Cheers!

Marc
 
#2
Hi, I do not want to over simplify things but why not just do 4 separate paired t-tests (or paired mann whitney u tests because of your low sample size per experiment).

Doing a 2-way ANOVA will still force you to finally end up with groups of 5 measurements per condition to compare anyway.

You will then have 4 times confirmation of a significant difference between treated and control.

Mixing dependent (5) and independent (4) measurements is usually not an easy road to travel.