Week 6 Reading Post — The Truthful Art: Chapter 10 and Chapter 11
Chapter 10: Mapping Data
This chapter discusses geographical area representations as a way to map data. The main aspects of a map are scale, projection, and the symbols used to depict information. Maps can be classified as large-scale maps — a close-up with more detail — or small-scale maps, such as the Earth as a whole. Choosing scale is important because it will vary depending on the assumptions about how much your readers know. Projection is “the process of making a globe, or a portion of it, fit into a flat picture.” This can be rather difficult, and features will inevitably be distorted. There are a wide variety of projections out there based on the five properties that can be distorted: shape, area, angles, distance, and direction. Some of these maps are conformal projections which preserve continental shapes and local angles. Other maps are called equal-area projections and preserve area ratios. No one map can be both types of projection. In choosing which map is best, you must consider what size region you’re going to show and other factors as well. Maps that contain data are called thematic maps, and are meant to show attributes, statistics, spatial patterns, or relationships as well as locations. Data can be encoded onto a map using a variety of symbols. Some interesting examples include dot maps and proportional symbol maps, for which it is important to consider symbols placement and scale. Another popular type of map is the choropleth map, which assigns different colors to different areas based on the attributed data. The issue that Cairo notes about choosing colors and shades carefully is extremely important in my opinion. Human brains already assign values to various colors, and when viewing these colors in opposition, or when viewing colors together that imply different things, our understanding of the data can be distorted. For example, red appears visually striking to our eyes and often implies “this is the worst or highest value” whereas green, especially in the case of maps, might imply “this is the normal environment or landscape.” This is something I consider especially important in environmental and climate data representations because understanding patterns and spatial relationships is crucial to most science data. Other types of maps include cartograms, which I find slightly confusing visually, as our eyes are trained to read maps using geographic spatial relationships, isarithmic maps (contour maps), and Voronoi maps (which are very cool).
Chapter 11: Uncertainty and Significance
This chapter talks about data statistics that introduce margins of error and which must be addressed and understood in order to grasp the full picture. The margin of error is also called the confidence interval. Histogram curves can be used to represent distribution but can also be used to infer probability because the curve shows where most of the scores lie in the distribution. In this way, one can draw a histogram curve of sample means to obtain a distribution of sample means as well as a standard error of the mean. This can lead us to estimate how much the mean of any given sample might deviate from the average mean. One can also develop confidence levels, point estimates, critical values, and other statistical measures in order to understand data patterns and the overall story better. Cairo notes that you should take your data to experts and that understanding statistical values and significance is crucial to developing and sharing a story that is as close to the absolute truth as possible.
My thoughts after reading these chapters:
What I took away from these chapters was not only the important of statistical analyses and geographical representations as a way to communicate spatial data relationships, but also the importance of scale in data representations. After reading Chapter 10, I was thinking about ways that maps and other visualizations may distort scale and thereby influence the viewers reading and comprehension of the subject. Scale is so important when trying to understand relationships and patterns in the data, as well as the magnitude of these. I was reminded of an infographic that I’ve seen countless times in the environmental communication field, Figure 1.1 below. This infographic shows personal choices that one can make to reduce carbon emissions, and places them on a scale from left to right of smallest reduction impact to largest reduction impact. All the way on the right, as the choice that reduces one’s carbon imprint the most, is having one fewer child. This bar on the bar graph, however, is not to scale. The “have one fewer child” bar goes all the way up to 60 on the left-most axis, but this bar is distorted in a way that does not effectively display how much having one fewer child will reduce your carbon emissions. Perhaps this was done to avoid needing a taller page format, or perhaps because many people understandably cringe at the thought of being told how or how not to have children. But overall, this distortion of scale distorts the viewers understanding of the impact of carbon emissions, and, in my opinion, should not be overlooked.
