Thread: A statistical nightmare

I need to find a sample size that will give power <.80. I am measuring effects of chemically treated limbs vs non treated limbs...same individual...actually a ratio of the two tested values. Using lab records I am trying to compare results from the past 15 years (to eliminate age bias). Any clue where to start? My data is a ration of muscle mass treated vs untreated. 10 subjects from each year...but can get more... Pasted a ver ysmall amount of data. What I want is to be able to show that, over time, the treated sides fare WORSE in mass that do untreated. So how do I use the power requirement of .8 to determine the n?