1.By convention (but not always), the type II error rate (known as beta level) is set to 0.20. Another common type II error rate is 0.10.

2.For 0.20, this means we want a 20% chance or less of erroneously accepting the null hypothesis of no association when the alternative hypothesis is in fact true.

3.Note that “statistical power” is 1 minus the type II error rate.