Protocol Online logo
Top : New Forum Archives (2009-): : Bioinformatics and Biostatistics

help with basic statistics to test reproducibility - (Jun/18/2015 )

Hey,

I've been going really crazy :mad: with these statistical tests, been trying to understand this since two months with no luck.:confused:
 
What i have and what i want:
so, I have two independent experiments with 7 biological replicates each and 4 technical replicates for each of the biological replicates. Now i've been trying to get a suitable method to show that both the experiments have similar results and that the experiment is reproducible. 
 
What i did:
# Found the average of the four replicates
# Used shapiro wilks test to test the average from both the experiments for    normality.
#Used Welchs two sample t-test to test for statistical difference.
 
Did I do it right?? (btw Im a total noob at statistics :()
is there something else I can do here??
Thank you,

-avipasup-

I think you have what’s called a “randomised complete block” with 7 treatments and 2 blocks.

A straight forward ANOVA will hopefully show no significant block effects.

-DRT-