In: Statistics and Probability
An Izod impact test was performed on 20 specimens of PVC pipe. The ASTM standard for this material requires that Izod impact strength must be greater than 1.0 ft-lb/in. The sample average and standard deviation obtained were x = 1.121 and s = 0.328, respectively. (a) Test H0: µ = 1.0 versus H1: µ 1.0 using α = 0.01 and draw conclusions. (b) Use the P-value approach to confirm your inference (c) Construct the 99% lower confidence bound on the mean and use it to test the hypothesis. (d) Suppose that the mean strength is as high as 1.1 ft-lb/in, the engineer would like to detect this difference with probability at least 0.90. Was the sample size n = 20 used in part (a) adequate? Use the sample standard deviation s as an estimate of σ in reaching your decision. Estimate what sample size is needed to detect this difference.
Given That:-
An Izod impact test was performed on 20 specimens of PVC pipe. The ASTM standard for this material requires that Izod impact strength must be greater than 1.0 ft-lb/in. The sample average and standard deviation obtained were x = 1.121 and s = 0.328, respectively.
(a) Test H0: µ = 1.0 versus H1: µ 1.0 using α = 0.01 and draw conclusions.
Here
vs
Given
n = 20, s = 0.328
Test statistic
= 1.65
d.f = n - 1
= 20 - 1
= 19
At , critical value is 2.54
Here cal t = 1.65 < 2.54
so do not reject H0.
(b) Use the P-value approach to confirm your inference
here p - value = 0.058 > 0.01
so do not reject H0.
(c) Construct the 99% lower confidence bound on the mean and use it to test the hypothesis.
The lower confidence bound is
= 1.121 - 0.1863
= 0.9347
(d) Suppose that the mean strength is as high as 1.1 ft-lb/in, the engineer would like to detect this difference with probability at least 0.90. Was the sample size n = 20 used in part (a) adequate? Use the sample standard deviation s as an estimate of σ in reaching your decision. Estimate what sample size is needed to detect this difference.
Here error (e) = 1.1 - 1
= 0.1 , s = 0.328
At 90%,
sample size (x) =
= 29.11 30
Here the sample size n = 20 is not adequate we need the sample size n = 30.