The Golden Age of Statistical Graphics

arXiv:0906.3979v1 [stat.ME] 22 Jun 2009 Statistical Science 2008, Vol. 23, No. 4, 502–535 DOI: 10.1214/08-STS268 c ⃝Institute of Mathematical Statistics, 2008 The Golden Age of Statistical Graphics Michael Friendly Abstract. Statistical graphics and data visualization have long histo- ries, but their modern forms began only in the early 1800s. Between roughly 1850 and 1900 (±10), an explosive growth occurred in both the general use of graphic methods and the range of topics to which they were applied. Innovations were prodigious and some of the most exquisite graphics ever produced appeared, resulting in what may be called the “Golden Age of Statistical Graphics.” In this article I trace the origins of this period in terms of the infras- tructure required to produce this explosive growth: recognition of the importance of systematic data collection by the state; the rise of sta- tistical theory and statistical thinking; enabling developments of tech- nology; and inventions of novel methods to portray statistical data. To illustrate, I describe some specific contributions that give rise to the appellation “Golden Age.” Key words and phrases: Data visualization, history of statistics, smooth- ing, thematic cartography, Francis Galton, Charles Joseph Minard, Flo- rence Nightingale, Francis Walker. 1. INTRODUCTION Data and information visualization is concerned with showing quantitative and qualitative informa- tion, so that a viewer can see patterns, trends or anomalies, constancy or variation, in ways that other forms—text and tables—do not allow. Today, statis- tical graphs and maps are commonplace, and include time-series graphs of economic indicators, pie- and bar-charts of government spending, maps of elec- tion results (e.g., the “red” and “blue” maps of U.S. Presidential races), maps of disease incidence or out- break (perhaps related visually to potential causes) and so forth. “New” graphical methods are frequently proposed to help convey to the eyes an increasingly complex Michael Friendly is Professor, Psychology Department, York University, Toronto, ON, M3J 1P3 Canada e-mail: friendly@yorku.ca. This is an electronic reprint of the original article published by the Institute of Mathematical Statistics in Statistical Science, 2008, Vol. 23, No. 4, 502–535. This reprint differs from the original in pagination and typographic detail. range, size and scope of the data of modern science and statistical analysis. However, this is often done without an appreciation or even understanding of their antecedents. As I hope will become apparent here, many of our “modern” methods of statistical graphics have their roots in the past and came to fruition in a particular period of time. The graphic display of data has a very long his- tory (Friendly and Denis, 2000; Friendly, 2008), but the age of modern statistical graphs and maps only began around the beginning of the 19th century. In statistical graphics, William Playfair [1759–1823] in- vented the line graph and bar chart (Playfair, 1786), followed by the pie chart and circle graph (Playfair, 1801). Statistical maps have their modern origin in the use of isolines, showing curves of constant value (wind directions and magnetic declination), by Ed- mund Halley [1656–1742] (Halley, 1701). Another map technique, the use of continuous shading (from light to dark) to show the geographic distribution of regional values (literacy in France), was first used by Baron Charles Dupin [1784–1873] (Dupin, 1826). With these innovations in design and technique, the first half of the 19th century was an age of en- thusiasm for graphical display (Funkhouser, 1937; 1 2 M. FRIENDLY Palsky, 1996) and witnessed explosive growth in sta- tistical graphics and thematic mapping, at a rate which would not be equaled until recent times. The period from about 1840–1850 until 1900–1910 saw this rapid growth continue, but did something more. In the latter half of the 19th century, youthful en- thusiasm matured, and a variety of developments in statistics, data collection and technology combined to produce a “perfect storm” for data graphics. The result was a qualitatively distinct period which pro- duced works of unparalleled beauty and scope, the likes of which would be hard to duplicate today. I argue that this period deserves to be recognized and named; I call it the “Golden Age of Statisti- cal Graphics.” By the end of this period, statisti- cal graphics had become mainstream; yet, paradox- ically, the enthusiasm, innovation and beauty of the Golden Age would soon die out. To give some initial (and graphic) sense of these periods in the history of data visualization and the growth and decline in the innovations referred to above, I consider Figure 1, showing the time distri- bution of 260 items from the database of the Mile- stones Project (Friendly, 2005). This is a compre- hensive catalog of statistical and graphical develop- ments which are considered to be significant events (“milestones”)

…(Full text truncated)…

This content is AI-processed based on ArXiv data.