72

The Essentials of Biostatistics for Physicians, Nurses, and Clinicians,
First Edition. Michael R. Chernick.
© 2011 John Wiley & Sons, Inc. Published 2011 by John Wiley & Sons, Inc.

CHAPTER 6

6. Hypothesis Testing

The classic approach to hypothesis testing is the approach of Neyman

and Pearson, initially developed in the 1930s. It differed from the
approach of signifi cance testing that was proposed by R. A. Fisher but
was clear and methodical, whereas some of Fisher ’ s ideas were obtuse
and poorly explained. The differing opinions of the giants in the fi eld
of statistics led to many controversial exchanges. However, although
Fisher was probably the greatest contributor to the rigorous develop-
ment of mathematical statistics, his fi ducial theory was not convincing
and was largely discredited.
The Neyman and Pearson approach starts out with the notion of a
null and alternative hypothesis. The null hypothesis represents an unin-
teresting result that the experimenter wants to refute on the basis of the
data from an experiment. It is called the null hypothesis because it
usually represents no difference, as, for instance, there is no difference
in the primary endpoint of a clinical trial when comparing a new treat-
ment with a control treatment (or placebo).
The approach fi xes the probability of falsely rejecting the
null hypothesis and then determines a fi xed sample size that will
likely result in correctly rejecting the null hypothesis, when the differ-
ence is at least a specifi ed amount, say δ. When we reject the null