i a have transformed my data to data frame named df with only column names(no rownames).each column represnt one sample with 3 observations (in deed nrow(df)=3, and ncol(df)=92).in order to check homoskedasticity of variance of my 92 samples i do: bartlett.test(df) it work and give me a result.but i'm afread of getting false result,knowen that a call of such function require a vector of data x and a factor g . g is omitted when x is a list of vector. is the call that i do true? is a data frame in my case considered like a list of vector? -- View this message in context: http://www.nabble.com/bartlett-test-tp17687757p17687757.html Sent from the R help mailing list archive at Nabble.com.