Introduction To R

---

The Answer...

Prep

> seeded <- scan('seeded')
> unseeded <- scan('unseeded')

> x1 <- log(seeded)
> x2 <- log(unseeded)

> n1 <- length(x1)
> n2 <- length(x2)

> s1 <- var(x1)
> s2 <- var(x2)

Compute S squared

> ssq <- ((n1-1)*s1+(n2-1)*s2)/(n1+n2-2)

Compute SE12

> se12 <- sqrt(ssq)*sqrt((1/n1)+(1/n2))

Hence t

> t <- (mean(x1)-mean(x2))/se12

The answer is therefore...

> t
[1] 2.544369