Abstract
An apparent paradox occurring in statistical hypothesis testing is described and resolved. What seem to be contradictions are shown to be no more than the result of confusion on the part of the experimenter as to what conclusion should be drawn from an arbitrarily stated null hypothesis. A confidence interval approach that avoids this confusion is also described.
Additional information
Notes on contributors
T. P. Daugherty
Dr. Daugherty is an assistant professor in the Department of Industrial and Systems Engineering. He is a member of ASQC.