In the introductory chapter, Healy highlights the importance of cognitive perception in the construction and reception of data visualization as the theoretical backbone for good data visualizations. Regardless of its forms, data visualization should prioritize the audience to select conscious design choices reflective of the creator’s ideology and tailored for the context of the audience because visualizations bear the responsibility of structuring a dataset and indicating a correlation. Consequently, graphs failed to contain these functions would be considered bad. Healy lays out three aspects of a bad visualization: bad taste, bad data, and bad perception. Bad tastes hinder the clarity of the data; bad data fail to serve the central message of the graph; bad perceptions present datasets in visually confusing manners. Specifically, Healy focuses on the perception of data visualization seeing its intricate yet undismissable ability to direct the human mind with lines, shapes, and colors. The designer of a data visualization has to bear in mind the visual cues these elements bring about when designing a graph.
However, one of Healy’s basic qualifications for what composes a bad visualization seems to me slightly questionable. In his definition of bad data, Healy instantiates misleading use of data with the example of the New York Times graph “How Stable Are Democracies? ‘Warning Signs Are Flashing Red’” (Taub, 2016). In terms of its physical nature as a data visualization, of course, this graph is inarguably bad as it misrepresents the raw data and over exaggerates the tendencies. Yet Healy overlooked a vital component of data visualization––the visualization of a dataset is to relay a message to its audience. There should not be any judgment over whether a message is “good” or “bad” when the focus is on the data visualization itself. Therefore the visuals should not be called “bad” if it obviously closely adheres to the message. They are leading exactly in the direction the creator intended. In this sense, the New York Times graph might be a “bad” data visualization at most but a highly successful communicational tool.