Workshop attendees have asked what readings I recommend on graph design. If you were to read just one book on creating effective graphs, I recommend
N. Robbins. Creating more effective graphs. Chart House, 2013.
Robbins writes for a wide audience (no statistics, no computer code) and provides abundant examples—the reader can see how one design is more effective than another. The book is required reading in my data visualization course; I adapt her methods and examples in several sections of my workshops.
Despite their limitations, pie charts and bar charts appear “natural” to many audiences. Authors presenting unfamiliar graphs risk rejection because audiences resist change, becoming indifferent to a graph’s meaning while preoccupied with its form. The underlying cause of this tension is visual convention, explored in depth in a book I highly recommend:
C. Kostelnick and M. Hassett. Shaping information: The rhetoric of visual conventions. Southern Illinois Univ. Press, 2003.
Kostelnick also does a wonderful job of outlining how different discourse communities construct and perceive visual information in a short paper accessible to the non-specialist.
C. Kostelnick. Conflicting standards for designing data displays: Following, flouting, and reconciling them. Technical Communication, pages 473–482, 1998.
Warranting repeated readings are the well-known works of Edward Tufte. As Kostelnick notes, Tufte’s aesthetic is grounded in the conventions of minimalism—so is subject to minimalism’s critiques. Nevertheless, Tufte’s advocacy to connect “seeing to thinking” is, in my opinion, the essence of contemporary graph design. I’ve also attended Tufte’s workshop. If you get the chance, go!
E. Tufte. The visual display of quantitative information. Graphics Press, 1983.
E. Tufte. Visual explanations. Graphics Press, 1997.
Jean-luc Doumont covers written, oral, and visual communication in a book, like Tufte’s, that is beautifully and idiosyncratically designed. Though I refer to it less often than other sources, several graph examples have influenced my teaching and design practice. I’ve also participated in a Doumont webinar. He convinced me, by doing, that a webinar can be an effective medium.
J. Doumont. Trees, maps and theorems. Principiae, 2009.
Howard Wainer’s books are fun tramps through the graphical landscape, readable and often humorous—with only an occasional obscure technical detour. Trout in the milk includes a wonderful exposition on the life and contributions of John Tukey. Medical illuminations has implications for graph design in any discipline.
H. Wainer. Visual revelations: Graphical tales of fate and deception from Napoleon Bonaparte to Ross Perot. Erlbaum, NJ, 2000.
H. Wainer. Graphic discovery: A trout in the milk and other visual adventures. Princeton University Press, NJ, 2005. (Original work published 1997, Copernicus Books, NY).
H. Wainer. Medical illuminations: Using evidence, visualization and statistical thinking to improve healthcare. Oxford Univ. Press, 2014.
We all stand on the shoulders of giants and—as with William Playfair—sometime scoundrels. Playfair invented today’s conventional forms: pies, bars, and lines. He flouted the statistical conventions of his day by publishing graphs instead of tables, a concept roundly rejected by his contemporaries. You can read Playfair in a facsimile of the 1801 original, edited by Wainer and Spence, who add a biographical sketch and a discussion of Playfair’s legacy.
W. Playfair. The commercial and political atlas and statistical breviary. Cambridge Univ. Press, 2005.
Playfair’s arguments extolling the superiority of graphs to tables are prescient of contemporary arguments extolling the superiority of new designs, flouting convention, and imploring audiences to see a story.
Robbins acknowledges her debt to Bill Cleveland. Her stated intent is to make Cleveland’s ideas accessible to the non-technical audience. But of you want the technical details, they’re here:
W. S. Cleveland. Visualizing data. Hobart Press, NJ, 1993.
W. S. Cleveland. The elements of graphing data. Hobart Press, NJ, 1994.
Graph design serves the communication of ideas. Humans are involved; there is more to the craft than technique alone. To venture beyond the boundaries of our comfortable positivist technical framework, I recommend the following. These authors provide social context that others (except Kostelnick) tend to neglect: the human context of data graphics and the power relationship between designer and audience. Both articles are thought-provoking and repay close reading.
B. F. Barton and M. S. Barton. Modes of power in technical and professional visuals. J. Business and Technical Communication, 7(1):138–162, Jan 1993.
S. Dragga and D. Voss. Cruel pies: The inhumanity of technical illustrations. Technical Communication, 48(3):265–274, Aug 2001.
Other books on my shelf that I browse from time to time for ideas on graph design include:
S. Few. Now you see it. Analytics Press, 2009.
S. Few. Show me the numbers, 2/e. Analytics Press, 2012.
S. Kosslyn. Graph design for the eye and mind. Oxford Univ. Press, 2006.
E. Tufte. Envisioning information. Graphics Press, 1990.
And there you have it.