The Telecommunications Act of 1996 has brought about many interesting statistical problems related to determining when an Incumbent Local Exchange Carrier (ILEC) is providing service to a Competing Local Exchange Carrier's (CLEC's) customers that is at least as good as the service that it provides to its own customers. One such topic is statistical testing for parity in a self-effectuating enforcement mechanism. Such a mechanism is one that will automatically test for parity and determine an ILEC's compliance using a production process with little-to-no manual intervention, on a monthly basis. Performance comparisons of ILEC customer service versus CLEC customer service in a self-effectuating system are based on observational data, not a designed experiment. The sample sizes one encounters vary considerably, and therefore the power of a hypothesis test varies from test to test. In order to address the concerns of both the ILEC and the CLEC, it has been suggested that a test's critical value be chosen so that the Type II Error probability at a "meaningful" alternative is equal to the Type I Error probability. We will discuss a technique for balancing these probabilities, and argue that the methodology inherently defines the materiality factor.