Dear All, I am using the code below to calculate error bars. I note that the length of the error bars can be varied by varying the constant (0.975). It does appear that any number can be substituted for 0.975, making it confusing for me to know how to quantify the error bars. I wish to quantify my error bar in terms of sigma (say 1 sigma, 2 sigma). Thanks for any idea. Regards Ogbos stderr<-function(x)sqrt(var(x)/length(x)) p<-sapply(temp,stderr) n<-sapply(temp,length) ciw<-qt(0.975,n)*p/sqrt(n) [[alternative HTML version deleted]]