Spatial Thinking in Planning Practice: An Introduction to GIS

33 Figure 4.9: Unique percentage values for people who use the term “pop” by state. Source: JM Smith, Depart- ment of Geography, "e Pennsylvania State University. "e following graph (Figure 4.10) shows exactly the same data set, only grouped into 10 classes with equal 10% ranges). It’s much easier to discern patterns and outliers in the classi!ed data than in the unclassi!ed data. Notice that people in a large number of states (23) do not really prefer the term “pop” as they are distributed around 0 to 10 percent of users who favor that term. "ere are no states at the other extreme (91-100%), but a few states whose vast majority (81-90% of their population) prefer the term pop. Ignoring the many 0-10% states where pop is rarely used, the most common states are ones in which about 2/3 favor the term; looking back to `Figure 3.13, these are primarily northern states, including Pennsylvania. All of these variations in the information are obscured in the unclassi!ed data. Figure 4.10: Classed percentages of people who use the term “pop” by state. Source: JM Smith, Department of Geography, "e Pennsylvania State University. As shown above, data classi!cation is a generalization process that can make data easier to interpret. Classi!ca- tion into a small number of ranges, however, gives up some details in exchange for the clearer picture, and there are multiple choices of methods to classify data for mapping. If a classi!cation scheme is chosen and applied skillfully, it can help reveal patterns and anomalies that otherwise might be obscured (as shown above). By the same token, a poorly-chosen classi!cation scheme may hide meaningful patterns. "e appearance of a themat- ic map, and sometimes conclusions drawn from it, may vary substantially depending on the data classi!cation scheme used. "us, it is important to understand the choices that might be made, whether you are creating a Chapter 4: Mapping People with Census Data