Question

Suppose you wish to test the claim that the mean value of the differences for a population of matched paired data is different from 0. Given a sample of n=23 and a significance level of alpha=0.05, what criterion would be used for rejecting the null hypothesis? (hint: this is simply the critical values)

Can somebody please explain how to do this in excel? I would really appreciate it.

Answer 1

From given claim we get, the null and alternative hypotheses are,

H0 : μd = 0

Ha : μd ≠ 0

sample size (n) = 23

significance level of alpha = 0.05

Degrees of freedom = 23 - 1 = 22

Since, this is a two-tailed test, t-critical values at significance level of 0.05 with 22 degrees of freedom are, t_{​​​​​​crit} = ± 2.074

Excel Command for finding critical values as follows:

Excel Command : =TINV(0.05, 22) = 2.074

Criterion for rejecting the null hypothesis is,

Reject H0, if t < -2.074 or t > 2.074