Marc Mulwray is the new marketing research director for AT&T Wireless. Marc was hired to help AT&T address new challenges from Verizon, Sprint, and T-Mobile in the highly competitive environment for wireless customers. New deals and pricing plans, new claims about network speed and new device offerings seem to emerge daily.

One of the crucial challenges for AT&T is that of current customer retention, as other players try to pull customers away with deals, including paying their cost to break their contracts with AT&T. Current customer retention is especially important as growth in the total number of wireless customers has waned given that nearly everyone has a wireless phone or tablet. 

Marc has been charged with determining how many current AT&T customers will switch to another wireless provider over the next six months given the current deals offered by competitors. To address this question, Marc and his team designed and fielded a national survey among its current customers. The survey covered a number of areas, including customer demographics, psychographics, wireless usage, and so on; but the key questions relate to their likelihood to switch to various other carriers in response to the deals, pricing and phones they are offering. Surveys were administered online based on e-mail invitations with links to the survey. The survey was completed by 1,200 customers who gave complete answers to all questions.

Initial results indicate that 14 percent of customers are likely to switch. The margin of error is 2.8 percent, which means that (at the 95 percent confidence level) the actual percentage of customers switching could be as low as 11.2 percent or as high as 16.8 percent. Given that AT&T has more than 100 million customers, this is a difference in customers lost of 16.8 million minus 11.2 million or 5.6 million. AT&T senior management is concerned about this error range of ±2.8 percent, which means that error spans a total of 5.6 percentage points. Further customer retention efforts must be budgeted now, and AT&T senior managers want firmer numbers on which to base strategies and budgets.

Questions 

1. How could the error range be reduced without collecting more data? Would you recommend taking this approach? Why or why not?

2. Do you think AT&T senior management would find this approach to reducing the error range satisfactory?

3. If 1,000 more respondents were surveyed and 20 percent of them indicated that they would switch, what would the error range become?