Mariners had to rely on error-prone charts and faulty compasses they made celestial observations while standing on the decks of rocking boats, and-if all else failed-threw rope overboard in an attempt to work out how far from the seabed they were. Such measurements were particularly important at sea, where accurate navigation presented a considerable challenge. This was well into the Age of Discovery, and Europeans were concerned with the measurement of time, distance, and location. The book begins with what might be the first statistical graph in history, devised by the Dutch cartographer Michael Florent van Langren in the sixteen-twenties. In “ A History of Data Visualization and Graphic Communication” (Harvard), Michael Friendly and Howard Wainer, a psychologist and a statistician, argue that visual thinking, by revealing what would otherwise remain invisible, has had a profound effect on the way we approach problems. The right graph, he pointed out, would have shown the truth at a glance. A decade later, Edward Tufte, the great maven of data visualization, used the Challenger teleconference as a potent example of the wrong way to display quantitative evidence. Soon after takeoff, the rubber O-rings leaked, a joint in the solid rocket boosters failed, and the space shuttle broke apart, killing all seven crew members.
This is why the managers made the tragic decision to go ahead despite the weather. The chart implicitly defined the scope of relevance-and nobody seems to have asked for additional data points, the ones they couldn’t see. But most of the experts were unconvinced. Some engineers used the chart to argue that the shuttle’s O-rings had malfunctioned in the cold before, and might again. As Diane Vaughn relates in her account of the tragedy, “ The Challenger Launch Decision” (1996), the data were presented at an emergency NASA teleconference, scribbled by hand in a simple table format and hurriedly faxed to the Kennedy Space Center. The first graph contains data compiled the evening before the disastrous launch of the space shuttle Challenger, in 1986.
One more twist: the points on the graph are real but have nothing to do with auto racing.