When salespeople, construction supervisors, managers, and other employees are away from the workplace, many of them carry mobile devices such as laptop computers and PDAs, often containing valuable, private data related to their jobs. Pointsec provides security systems to protect such data.To bring home the vulnerability of mobile devices, Pointsec decided to share information about the number of such devices left behind on taxis.8 The research involved conducting a survey of taxi drivers. Staff members at Pointsec’s public relations firm called major taxi companies in nine cities in Australia, Denmark, Finland, France, Germany, Norway, Sweden, Great Britain, and the United States. Each of the cooperating companies put these interviewers in touch with about one hundred drivers. Drivers were asked how many devices of each type—cell phones, PDAs, computers, and so on—had been left in their cab over the preceding six months. From these numbers, they came up with the rate of items left behind. Multiplying by the size of taxi fleets in each city, the researchers came up with city-by-city numbers: 3.42 cell phones per cab yielded 85,619 cell phones left behind in Chicago, for example. In London, the researchers concluded 63,135 cell phones were left in cabs, a startling increase of 71 percent compared to four years earlier.

1. Discuss why the sampling method and sample size make these results questionable, even though the numbers were reported as if they were precise.

2. The simple survey method described in the case may have been sufficient as a way to draw attention to the issue of data security. However, if the company were using data on lost mobile devices to predict demand for a product, accuracy might be more significant. Imagine that you have been asked to collect data on mobile devices left in cabs, and you wish to be able to report results with a 95 percent confidence level. How can you improve the sample design and select an appropriate sample size?