2-sided test:
We measure the probability of seeing something (again) as extreme as the observed value (or more so).
So you need to measure the P-value symmetrically both directions from the observed value--so the P value is double what it would be for a one-sided test.

#6.35, p. 333 Engine crankshafts:  We want to stop the process and fix it if the mean gets too far "off" from 224--either direction would be bad.  So two-sided.

H0 : µ = 224 mm
Ha : µ  Not = 224 mm
sigma = 0.060 mm given(the calculated sample standard deviation = .0618)
n = 16.  Std. dev. of xbar = 0.060/4 = 0.015
Calculate xbar = 224.0019375
Standardizing: z = (224.0019375 - 224)/ .015 = .0019375/.015 = 0.12917 ~ .13  (xbar is clearly close to mu)
       (If you used .0618, not the .06 you were supposed to, you would get .1254--still rounds to .13)
Farther out than .13 to the right has probability (1- .5517) = .4483.
Farther out than -.13 (symmetrical) to the left also has probability .4483.
So P-value, 2-sided, = .4483 + .4483 = .8966
  It would be very likely that we would see a result this far out again.
NO evidence for the alternative.

The results are "consistent with" the null hypothesis being true.

(This test itself doesn't quantify the evidence FOR the null hypothesis; it wasn't designed to do so.
To do inference for where the mean IS, do a Confidence Interval for the true mean, using this data.
The test only judges evidence for where it ISN'T.)


Sievers home  Math151-Sp03/Crankshafts.htm  4/18/03
This page belongs to Sally Sievers who is solely responsible for its content. Please see our statement of responsibility.