Re: [R] Comparison of numeric and character vectors

2012-01-03 Thread Ista Zahn
Ah, yes I see. Thanks John and Michael. On Tue, Jan 3, 2012 at 6:06 PM, R. Michael Weylandt wrote: > There's a coercion to character implicit (i.e., the number 999 gets > converted to the string "999") and then comparison is done in lexical > (dictionary) order in which digits are lower than char

Re: [R] Comparison of numeric and character vectors

2012-01-03 Thread R. Michael Weylandt
There's a coercion to character implicit (i.e., the number 999 gets converted to the string "999") and then comparison is done in lexical (dictionary) order in which digits are lower than characters. You'll also note you get apparently strange behavior like "34" < "9" if you don't think about thin

Re: [R] Comparison of numeric and character vectors

2012-01-03 Thread John Fox
Dear Ista, This is a consequence of coercion of the numbers to character: > c("Z", "a", 999, Inf) [1] "Z" "a" "999" "Inf" > sort(c("Z", "a", 999, Inf)) [1] "999" "a" "Inf" "Z" I hope this helps, John On Tue, 3 Jan 2012 17:56:29 -0500 Ista Zahn wrote: > Hi all, > > I just discovered