In: Statistics and Probability
Data were collected on number of credit hours earned from a random sample of 40 JMU students. The mean was 57.2 and the standard deviation was 8.9. Conduct a test to determine whether the true mean number of credit hours earned differs from 55.0, allowing a Type I error rate of 0.05. Assume that the distribution of the number of credit hours earned is normal. Show your full work. You may not use your calculator’s built-in function.
Solution:
The null and alternative hypothesis are
H0:
= 55.0
H1:
55.0
n = 40
= 57.2
s = 8.9
= 0.05
Since population SD is unknown,we use t test.
The test statistics t is given by ..
t =
= (57.2 - 55.0)/(8.9/40)
t = 1.563
n = 40
So , df = n - 1 = 40 - 1 = 39
= 0.05
So ,
/2 = 0.025
Now,
sign in H1 indicates that the two tailed test.
So, the critical values are
The critical region : t <
or t >
=
0.025,39
= 2.023 (using t table)
t = 1.563 < 2.023
Fail to reject the null hypothesis.
There is no sufficient evidence to support the claim that the true mean number of credit hours earned differs from 55.0.