[R] How to solve this problem ?

[Daren Tan]

Small progress, I am relying on levene test to check for equality of variances. Is my understanding correct, the larger the p-value, the more likely the variances are the same ?

> trt
 [1] 1 1 1 1 1 2 2 2 2 2 3 3 3 3 3 4 4 4 4 4
Levels: 1 2 3 4
> levene.test(rep(rnorm(5), 4), trt, option="median")

        Modified Robust Brown-Forsythe Levene-type test based on the absolute
        deviations from the median

data:  rep(rnorm(5), 4)
Test Statistic = 0, p-value = 1



> From: daren76 at hotmail.com
> To: r-help at stat.math.ethz.ch
> Subject: How to solve this problem ?
> Date: Wed, 20 Aug 2008 13:33:23 +0000
>
>
> I have disabled html text editing mode, thanks to Prof. Ripley for the kind reminder.
>
> Given three geological devices that takes 5 readings at 4 environmental conditions (A to D). What will be the proper approach to select the most reliable device ?
>
> m1 <- c(73,52,49,53,83,43,58,94,53,62,75,66,41,72,70,75,57,59,85,84)
> m2 <- c(31,38,30,35,36,26,27,38,22,31,24,35,36,31,38,33,32,28,33,30)
> m3 <- c(65,57,36,40,36,30,40,34,37,40,33,33,37,29,37,37,30,33,40,35)
>
> names(m1) <- rep(LETTERS[1:4], each=5)
> names(m2) <- rep(LETTERS[1:4], each=5)
> names(m3) <- rep(LETTERS[1:4], each=5)
>
> Before writing this email, I have tried to compare the sd for each device at each condition, but ran into obstacle on how to formulate the null hypothesis. Alternative solution tried was ANOVA, I am unsure whether it can help, as it compares the differences in means of each group.
>
> Thanks
>
> _________________________________________________________________
> Easily edit your photos like a pro with Photo Gallery.
> http://get.live.com/photogallery/overview