181

If a map showed every pothole, traffic sign, and store, it would be far too cluttered to be useful for planning a road trip. On the other hand, if a map included only a couple of reference points, it would be too easy to get lost and miss the destination. So a good map gives us just the right level of detail, calling our attention to special features and main roads.

182

The undigested blizzard of data we encounter in modern society can feel overwhelming, like that first type of map. But if we simply ignore data, we risk the pitfall of the second type of map. Failure to detect patterns in data in a timely manner has had serious consequences, ranging from the loss of a NASA spacecraft to large-scale misguided financial trading practices that caused billions of dollars of losses. Therefore, we need to develop good skills to "read" and appropriately summarize data so that we can navigate the terrain of information and numbers where we live and travel. Just as it helps for directions to have both numerical information (e.g., "3.2 miles on Gluckin Avenue") and visual diagrams or landmarks (e.g., "Turn right just after you pass the water tower"), it is important for data analysis to have both numerical and graphical techniques as well.

This chapter starts with an introduction to the concepts of exploring data from one quantitative variable. We will begin with a graphical technique called a histogram, which can serve as a middle-of-the-road approach in terms of the amount of detail it reveals about the data. Later, we explore graphical techniques that include more detail (stemplots and dotplots) and then less detail (boxplots). Along the way, we’ll throw in some calculations that can help us judge the center and spread of our data. The chapter’s concluding sections highlight distributions of data of a particular shape, known as normal distributions. This class of distributions is special because of its pervasiveness in statistics and its presence in the real world. Happy travels through statistics!