Hello! I am working on a manuscript on sexual dimorphism in an aquatic invertebrate, where we have estimated sexual dimorphism (SD) for 7 different traits in four populations (a total of 28 SD-estimates). We have used the following formula for estimating SD: 100 * (mean male trait value - mean female trait value)/overall trait mean).
Then, we have used these SD-estimates to perform a GLM against other interesting variables, such as the intersexual genetic correlations for each of the traits. Here are my questions: 1. Is there any procedure in "R" you would recommend that takes in to account the sampling variance of the SD-estimates, rather than using the mean value of each (which is supposed to reduce error and increase Type I-error rates? 2. Is there a procedure to estimate SE for ratios such as this SD-estimate? 3. The data in these GLM:s might not be entirely statistically non-independent (i. e. intersexual genetic correlations). Can you recommend any R-procedure (package) that can deal with this problem (e. g. bootstrapping or resampling)? Many thanks in advance for input! Erik Svensson -- View this message in context: http://r.789695.n4.nabble.com/Standard-errors-of-sexual-dimorphism-tp3785770p3785770.html Sent from the R help mailing list archive at Nabble.com. ______________________________________________ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
