I have 16 experiments performed under different conditions and I would like to compare them between each other. Each experiment has 5-6 samples (cells). Each cell produces hundreds of data points each (adhesion force data). When I make a histogram of the log adhesion force it seems I have a skewed distribution. I have fitted this distribution using 2 or 3 Gaussian curves using the mixtools in R. Any idea how I can determine statistically if one experiment has larger adhesion forces than another?
For example in the attached pictures I am showing four histograms that compile the information of adhesion forces with silicon. Each histogram has associated a value of sigma, mu and lambda. I would like to know, for example, if the adhesion with silicon in 1 is statistically different from the adhesion in gold.