Outline

Discuss Type I and Type II Errors

Introduce the concept of power

Discuss ways to increase power

Decision Making Outcomes

                                                                        Actual Situation

                                                            H₀ True                                 H₀ False

                        Reject H₀                     Type I error                     Correct

Decision                                                                                      Decision

                                                                                                          (Power)

                        Retain H₀                             Correct                           Type II   error

                                                            Decision

POWER

The power of a statistical test is the probability that the test will correctly reject a false null hypothesis.

Another related definition is that power is the probability of obtaining sample data in the critical region when the null hypothesis is false.

WAYS TO INCREASE POWER

1.        Increase the alpha level (e.g. from .01 to .05).

Any factor that causes more of the treatment distribution to be in the critical region will increase power.

2.        Changing from a two-tailed (non-directional) to a one-tailed test.

3.        Increase the sample size (this reduces standard error).

4.        Make sure that the treatment is strong enough (this may require pilot testing)