Hello Mr. Burns, Hello Mr. Dwinseminus thank you very much for your incredible quick and efficient reply! I was completely successful with the following command: pairs<-data.frame(pred=factor(unlist(input[,-c(1,ncol(input))])),ref=factor(input[,ncol(input)])) In case of the "input" example data.frame I sent with my question the above code is equivalent to: pairs<-data.frame(pred=factor(unlist(input[2:17])),ref=factor(input[,18])) Great! That is exactly what I was looking for! This simple code will save me hours! Patrick, your book looks in fact very interesting and will be my perfect reading material for the following nights :-) (probably not only the first chapter ;-). Thanks for the hint - and the free book of course. David, the "input" data.frame is the result of the reshape-command I performed. I just copied it from the R-console into the e-mail. In fact the first column "video" is not part of the data, but needed for analysis with kappam.fleiss function of the irr-package. Sorry, you are absolutely correct, I should have mentioned this in my question. I will improve when I ask my next question :-). Again I like to thank you for your help and wish you a pleasant Sunday. Greetings from Munich, Felix
Patrick Burns wrote:
If I understand properly, you want input[, -c(1, ncol(input))] rather than input[[, -c(video, 21)]] Chapter 1 of S Poetry might be of interest to you. Patrick Burns patrick at burns-stat.com +44 (0)20 8525 0696 http://www.burns-stat.com (home of S Poetry and "A Guide for the Unwilling S User") drflxms wrote:
Dear R-colleagues, another question from a newbie: I am creating a lot of simple pivot-charts from my raw data using the reshape-package. In these charts we have medical doctors judging videos in the columns and the videos they judge in the rows. Simple example of chart/data.frame "input" with two categories 1/0: video 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2 2 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 3 3 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 4 4 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 5 5 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 6 6 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 7 7 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 8 8 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 9 9 0 0 0 0 0 0 0 0 0 1 0 1 1 0 1 1 0 0 0 1 0 10 10 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 I recently learned, that I can easily create a confusion matrix out of this data using the following commands: pairs<-data.frame(pred=factor(unlist(input[2:21])),ref=factor(input[,22])) pred<-pairs$pred ref <- pairs$ref library (caret) confusionMatrix(pred, ref, positive=1) - where column 21 is the reference/goldstandard. My problem is now, that I analyse data.frames with an unknown count of columns. So to get rid of the first and last column for the "pred" variable and to select the last column for the "ref" variable, I have to look at the data.frame before doing the above commands to set the proper column numbers. It would be very comfortable, if I could address the last column not by number (where I have to count beforehand) but by a variable "last column". Probably there is a more easy solution for this problem using the names of the columns as well: the reference is always number "21" the first column is always called "video". So I tried: attach(input) pairs<-data.frame(pred=factor(unlist(input[[,-c(video,21)]])),ref=factor(input[[21]])) which does not work unfortunately :-(. I'd be very happy in case someone could help me out, cause I am really tired of counting - there are a lot of tables to analyse... Cheers and greetings from Munich, Felix
______________________________________________ R-help at 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.