Jim: I'm sure that this is much more sophisticated than anything that the OP ever dreamed of dealing with. And irrespective of that, the issue is not about there being more than two sexes in some contexts but rather of the folly of treating categorical data as numeric.
Moreover your "neither male nor female" sexes have nothing whatever to do with missing values. Missing means that the value wasn't *observed*, not the values was weird/strange/unfamiliar.
cheers, Rolf -- Technical Editor ANZJS Department of Statistics University of Auckland Phone: +64-9-373-7599 ext. 88276 On 01/11/15 12:45, Jim Lemon wrote:
Having had to face this problem myself more than once, I sympathize with Ted's argument. First let me confess that I regard sex as a measure of the reproductive phenotype. Given the ongoing experimentation with both sex and gender, I have had to add "U" (Unstated - includes all those acronyms that can be mistaken for gamma hydroxy butyrate) to "M" and "F" in a dataset or two. Even worse is the crap shoot of sex chromosomes. While XYY is not much of a problem at all, Turner's Syndrome (XO) is neither female (although they appear to be) nor male. Given a reasonably large sample (the dream of some), nature usually provides a few permutations that, while we know what they are, don't really fit comfortably in either "M" or "F".
______________________________________________ R-help@r-project.org mailing list -- To UNSUBSCRIBE and more, see 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.
