[EMAIL PROTECTED] (Euh) wrote in message news:<[EMAIL PROTECTED]>... > Hi all, > > I'm following the time course of a certain variable by using three > independent measurement techniques (for which I can estimate the > variance associated with the measurment) > > I can't really assume any function to describ the evolution of the > variable over time (linear, quadratic, etc). Is there a way to compare > the results and assess if all the methods give similar results ? > > ANOVA analysis at each time ? > > An example would be: > > Method 1 Method 2 Method 3 > mean stdev mean stdev mean stdev > t1 24.6 6.4 31.0 13.8 15.5 5.7 > t2 123.5 87.6 155.2 71.5 62.3 20.0 > t3 174.0 33.8 142.7 46.4 75.8 18.9 > t4 210.6 91.2 113.6 26.8 101.9 45.1 > t5 396.4 25.4 263.9 16.5 209.3 11.7 > t6 303.7 66.7 271.9 156.6 216.9 172.1 > t7 153.6 93.4 261.0 225.6 76.0 41.7 > t8 250.9 140.7 289.5 122.7 93.7 28.1
Whether or not "all the methods give similar results" is a subjective call. How close is close enough? Are some kinds of differences less tolerable than others? If you're measuring the same objects on all 8 occasions then you can treat it as a 3-way Objects x Time x Method design, with Objects random, and look at the estimated variance components. (You want all the terms involving Method to be small, but if some of them are much bigger than others then that may have diagnostic utility.) And if the numbers you gave are typical then you might try fitting a generalized linear model (or, failing that, a transformation) that deals with the apparent heteroscedasticity. Also, try plotting the data. The first thing I would try would be a separate plot for each Object, with each plot showing three curves over Time, one for each Method. . . ================================================================= Instructions for joining and leaving this list, remarks about the problem of INAPPROPRIATE MESSAGES, and archives are available at: . http://jse.stat.ncsu.edu/ . =================================================================
