Companies A,M, and S are three long-distance carriers. All claim to give quality service, but S has been advertising its network as being incredibly clear. A consumer testing agencywishes to determine if S really is any better. A complicating factor in this determination is that you dont hook directly to a long-distance company. Your call must first go through your personal phone, through local lines, and through the local switch before it even gets to the long-distance company equipment, and then the call must go through local switch, local lines, and a local phone on the receiving end. Thus while one long-distance carrier might, in fact, have clearer transmissions than the others, you might not be able to detect the difference due to noise generated by local phones, lines, and switches. Furthermore, the quality may depend on the load on the long-distance system. Load varies during the day and between days, but is fairly constant over periods up to about 15 or 20 minutes. The consumer agency performs the following experiment. All calls will originate from one of two phones, one in New York and the other in New Haven, CT. Calls will be placed by a computer which will put a very precise 2-minute sequence of tones on the line. All calls will terminate at one of three cities: Washington, DC; Los Angeles; or Ely, MN. All calls will be answered by an answering machine with a high-quality tape recorder. The quality of the transmission is judged by comparing the tape recording of the tones with the known original tones, producing a distortion score D. Calls are placed in the following way. Twenty-four time slots were chosen at random over a period of 7 days. These 24 time slots were randomly assigned to the six originating/terminating city pairs, four time slots per pair. Three calls will be made from the originating city to the terminating city during the time slot, using each of the long-distance companies in a random order. The data follow (and are completely fictitious).