April 2002

Vol. 4, No. 4

Quality Quiz from Professor Cleary

You're right!


Cal Lesterol has missed the mark. The t statistic is used to test the hypothesis that there is no relationship between X and Y. In rejecting that hypothesis, one assumes that X is a valid predictor of Y. The formula to derive the t statistic is:

(Practical Tools for Continuous Improvement: Vol. I, Statistical Tools, PQ Systems, Inc., 2001. Page 307.)

In this case, the t value on Cal Lesterol's printout is -18.664. Since there are 24 data points, the degrees of freedom are 22:

df = n 2
= 24 2
= 22

Using an alpha value of .05 (the probability of rejecting the null when the null is true), the tabular t is 2.074. Since the calculated t is higher than the tabular t, the null is rejected, and one can assume that X is a predictor of Y.

Cal Lesterol's logical, linguistic response has nothing to do with statistics. Fortunately, his audience paid no attention to the t value, but simply let Cal take his chart back to the quality department for further study.


Copyright 2002 PQ Systems.

Please direct questions or problems regarding this web site to the Webmaster.