In the current issue of JSE, William Peterson's column "Topics for discussion from current newspapers and journals" includes links to articles about 2001 being "The year of the shark":
http://www.amstat.org/publications/jse/v9n3/peterson.html Peterson includes a summary of Paulos's NY Times commentary, "How to spot a trend when none exists." There is also a link to a graphic from Florida museum of natural history to back the argument that the apparent increase in shark attacks is merely due to the increase in Florida's population: http://www.flmnh.ufl.edu/fish/Sharks/Statistics/flpop.htm I would give this graphic a Tufte award. One is supposed to infer that the risk of shark attack is due simply to the increase in Florida's population. I picked the numbers off the Florida population graphic and plotted the odds of shark attack each decade relative to the odds of shark attack during 1900-1910: http://www.es.umb.edu/edg/ECOS601/sharkrelodds.jpg I don't think the Florida Dept. of tourism will be using this graphic in their "Visit Florida" campaign. While Paulos is right that the odds of shark attack are incredibly small, there is an increasing trend in the relative odds of shark attack. The dip in the relative odds of attack during the 1940-1950 era may be explicable in terms of the war. I can't really come up with a reason for the drop during the 1970-1980 decade other than the Summer 1975 release of Jaws. ================================================================= Instructions for joining and leaving this list and remarks about the problem of INAPPROPRIATE MESSAGES are available at http://jse.stat.ncsu.edu/ =================================================================
