A consumer agency has retained an independent testing firm to examine a television manufacturer's claim that its 25-inch console model consumes just 110 watts of electricity. Based on a preliminary study, the population standard deviation has been estimated as 11.2 watts for these sets. In undertaking a larger study, and using a simple random sample, how many sets must be tested for the firm to be 95% confident that its sample mean does not differ from the actual population mean by more than 3.0 watts?