xml - How to get a mean from data.frame? -

i have data.frame, got this: data <- ldply(xmltolist("http://www.nbp.pl/kursy/xml/a025z100205.xml"),data.frame)

i create list :

list <- data[[6]] 

then deleted na values

list <- list[!is.na(list)] 

and got this

[1] 0,0900 2,9915 2,5851 0,3850 2,7805 2,0566 2,1043 4,0921 1,4918 2,7837 [11] 4,7009 0,3723 3,3450 0,1561 0,5496 0,2615 2,3211 0,4987 0,4005 0,5589 [21] 0,9894 2,0923 1,9688 1,1851 5,7733 0,0643 0,2271 0,3884 1,5965 0,8687 [31] 0,0981 3,1673 0,2557 0,4384 4,5657 35 levels: 0,0900 2,9915 2,5851 0,3850 2,7805 2,0566 2,1043 4,0921 ... 4,5657 

but can't mean list. because of levels? how can this?

the class of 'list' factor due non-numeric component (,). when there non-numeric element in column, creating data.frame, default option stringsasfactors=true. can include argument stringsasfactors=false inside data.frame character column, still , should replaced. assuming meant decimal, replace . using sub, convert numeric as.numeric , mean.

mean(as.numeric(sub(',', '.', list))) 

edit

after reading data using op's code, understand 4th , 6th columns numeric (but columns 'factor' class). if interested in 6th column, above, if need mean of 4th , 6th, use sapply

 sapply(data[c(4,6)], function(x)         mean(as.numeric(sub(',', '.', x)), na.rm=true))  #przelicznik kurs_sredni   #298.00000     1.66298  

or

m1 <-  `dim<-`(as.numeric(sub(',', '.', as.matrix(data[c(4,6)]))),                  dim(data[c(4,6)])) colmeans(m1, na.rm=true) #[1] 298.00000   1.66298 

note: better not name objects function names.