Beruflich Dokumente
Kultur Dokumente
Variance Known
Independent Samples
Reject if:
Reject if μ0 not contained in:
x z
Reject if:
Reject if:
Variance Unknown
Independent Samples
Reject if: Reject if μ0 not contained in:
s s
x t , x t
n
Reject if P-Value < α
2 n 2
for ν=n-1
Use Table II to look up Reject if:
values*
for ν=n-1
Reject if:
for ν=n-1
If
Reject if P-Value < α
Reject if:
Reject if:
Variance Unknown
Dependent Samples
Reject if: Reject if 0 not contained in:
s s
d t d , d t d
n
dj d
1 n Reject if:
2
sd2
n 1 j 1 for ν=n-1
Tests on Variances
Reject if not contained in:
or
Hypothesis Test Statistic Critical Region P-Value CI
contained in:
or
* Recall that zα or tα represent the value of z (or t) such the area in the tail to the right of zα or tα is α.
Finding P-Values:
Z-Table: Find z0 using rows and columns of table. Determine area based on type of hypothesis test (left-, right-
or two-tailed). Remember for two-tailed tests you have multiply by a factor of 2 the probability in one of the
tails. Since the Z-table reports areas from the left, you may have to do 1 minus the area for it is a right-tailed
test.
T-Table: Go to degree of freedom row closest the correct value of ν. Find the values of α that the test statistic t0
falls in between; so the we can write - . If , where is the level of significance, then
we reject the null hypothesis. The above discussion is for the one-tailed test, the two-tailed test requires us to
multiply the P-value bounds by 2, i.e. 2 - , and then check against the level of significance.
F-Table: Typically better to use software.