Air Safety: End of the golden Age? Tables and Fi Table 1: Passenger Mortality risk for Various World-wide jet Services, 1990-1999 Type of service Death Risk per Flight First-World Domestic 1 in 13 million International within First World 1 in 6 million nternational between first and lin500.000 Developing world Within Developing-World lin500.000 Countries identified as First-World are listed in the text Death risk per flight is the chance of being killed on a randomly-chosen(nonstop) flight over the period 1990-99 e approximations attend the calculations, and the denominators are rounded off to the nearest half million. The mortality-risk difference between domestic and intemational flights in the first-world is not statistically significant: If major fatal crashes arise under a Poisson process at rate i per million flights, then the observed difference is consistent with short-term fluctuations around the mean of that process. However, the fact that ar observed difference might be construed as a temporary fluctuation does not mean that it must be so construeAir Safety: End of the Golden Age? Tables and Figures Table 1: Passenger Mortality Risk for Various World-wide Jet Services, 1990-1999 Type of Service Death Risk per Flight First-World Domestic 1 in 13 million International Within First World 1 in 6 million International Between First and Developing World 1 in 500,000 Within Developing-World 1 in 500,000 Notes: Countries identified as First-World are listed in the text. Death risk per flight is the chance of being killed on a randomly-chosen (nonstop) flight over the period 1990-99. Some approximations attend the calculations, and the denominators are rounded off to the nearest half million. The mortality-risk difference between domestic and international flights in the first-world is not statistically significant: If major fatal crashes arise under a Poisson process at rate
per million flights, then the observed difference is consistent with short-term fluctuations around the mean of that process. However, the fact that an observed difference might be construed as a temporary fluctuation does not mean that it must be so construed