45 pages • 1 hour read
Darrell HuffHow to Lie with Statistics
Nonfiction | Reference/Text Book | Adult | Published in 1954A modern alternative to SparkNotes and CliffsNotes, SuperSummary offers high-quality Study Guides with detailed chapter summaries and analysis of major themes, characters, and more.
Chapter 5Chapter Summaries & Analyses
Chapter 5 Summary: “The Gee-Whiz Graph”
Chapter 5 covers the issue of using statistical graphs to exaggerate or distort data. Huff notes the usefulness of charts: Tables can be challenging for a general audience to understand, and words don’t often work well when covering statistical data. However, graphs present many ways to mislead audiences.
Huff focuses on the simple line graph. He first gives a straight example of the chart with both axes properly labeled and covering the entire possible range of numbers. He notes that it looks accurate to what it represents, but it doesn’t look dramatic and takes up space on the page. He demonstrates how the impression that graph makes on the reader can be changed by cutting off the bottom portion of it. The data remains the same, but the resulting graph looks more dramatic because it lacks context. He further exaggerates the results by altering the ordinate and abscissa, changing the shape of the chart as a result. Huff notes the utility of a graph over words when creating misleading results by saying that a diagram “contains no adjectives or adverbs to spoil the illusion of objectivity” (65). Because graphs contain raw numbers, rather than a written interpretation, it is harder for the audience to see the issue with the data being presented if they are not looking for it. The chapter concludes with Huff noting the prevalence of misleading graphs throughout advertising.
Chapter 5 Analysis
Returning to the idea that Statistics Is an Art Form, this chapter examines ways to visually manipulate statistics. While Huff omits other graphs and charts in his discussions, such as the pie chart, he likely chose the line graph due to its common usage.
This chapter relies heavily on Geis’s illustrations to convey its content; they support Huff’s descriptions of ways to manipulate charts with visual examples for the reader. He presents the full line graph with all the relevant information, the same chart with the bottom portion removed, and a third in which he changes the shape by altering the proportion between the x-axis and the y-axis. As a result, this chapter focuses more on how the manipulation of the graphs is achieved than on just recognizing said manipulation. This method of presentation allows Huff to visually explain the process rather than requiring a breakdown of the math involved in obtaining some of the statistical results. This helps sustain the book’s accessibility for general readers. As he notes in the chapter, the presence of numbers and raw data tends to intimidate the unfamiliar viewer, regardless of the complication of said numbers and data.
The choice in graph changes he covers is subjective in nature. Cutting off portions of the graph, changing proportions, or other forms of manipulation are not necessary for demonstrating the soundness of the results. Rather, these are artistic and design choices made by the entity presenting the data to support the message it desires to convey.
