Sometimes Old and Always Awesome Graphics

I love old graphs. Sometimes I think about shunning my computer for a day and building out some massive complex visual analysis with a pen and a ruler.  And very quickly I remember that, thank God, I don't have to.

Every few months I print out a set of old data visualizations on poster-size printed and frame them for the office. Here's a few of my favorite. I'll add more as time goes on.

William Playfair | Scotland | 1823

Linear Chronology, Exhibiting the Revenues, Expenditure, Debt, Price of Stocks & Bread, from 1770 to 1824 (William Playfair, 1823)

Playfair invented just about every basic form of statistical graph – the time-series line graph, the bar chart, and the pie chart. He never made it to the scatterplot, which took another century to come into play.

This graph was part of a volume intended to be a perpetual publication with a year added to the end each time. This print shows Playfair’s timeline extended a year beyond his death in 1823.

Charles Joseph Minard | France | 1866

Map of European Cotton Imports during the American Civil War (Minard, 1866)

A map depicting the flow of cotton into Europe before, during and after the American Civil War. A Northern blockade on Southern exports brought the flow of cotton from America (blue band) to a halt. As a result, following the war, America would have to compete with India and China (orange band) and Egypt and Syria (brown band) for a share of European cotton consumption.

A Tufte favorite, Minard is best known for the infographic depicting troop losses during Napoleon’s 1812 Russian campaign. If you don't know it, just scroll down a bit.

Eugene Pick | France | 1858

Tableau De L'Histoire Universelle depuis la Creation jusqu'a ce jour (Pick, 1858)

Stream of Time maps were popular in the 1800’s, showing growth, decay and conquests of civilizations over the millennia. Pick’s map opts for the year 4963 B.C.E. as the year of creation, and the starting point for his map.

I love it for its ambition, though the execution is a total mess.

Florence Nightingale | England | 1858

Diagram of the Causes of Mortality in the Army in the East (Nightingale, 1858)

Florence Nightingale, in addition to pioneering modern nursing, is also known as a pioneer of statistics and graphical representations of data. Specifically, Nightingale provided detailed analyses of mortality in the Crimean War, where she became prominent as a Nurse.

This is a unique diagram that Nightingale herself invented. Known as a Polar Area Diagram, Coxcomb, or the Nightingale Rose, it is a circular histogram, with this version breaking down the causes of death in the British army over 2 years.  It's pretty, but probably not as effective as simple bar chart.

Fletcher Hewes, Henry Gannett | United States | 1883

Rank of the most populous cities at each census: 1790-1880 (Hewes & Gannett, 1883)

This visual from the Statistical Atlas of the United States offers a fascinating look at population growth by city in the US, as measured by each decade’s census.

The arrow metaphor describes entries (fletchings) and exits (arrowheads) into the Top 40 list. Scribble lines show city mergers (e.g., Roxbury becomes a part of Boston between 1860 and 1870). Summary histograms in the bottom left describe the exponential growth of certain regions. Our favorite touch is the individual city population bars, which wrap around the page as the population becomes too great for the allotted space.

William Farr | England | 1852

Temperature and Mortality of London, for every week of 11 years (1840-50) (Farr, 1852)

Farr, an epidemiologist and statistician, believed that cholera was caused by miasma, or “bad air”, leading to his studies attempting to correlate outbreaks with temperature. The circular graphs show highly detailed records of temperature (inner ring) and mortality (outer rings) measured weekly for 12 years, identifying correlations and seasonality across over a decade of data.

Eventually, Farr accepted an alternative explanation of cholera offered by John Snow, who theorized that pathogens in water supplies were to blame. Still, Farr left a legacy of data-driven public health, predating the CDC by a century.

Charles Joseph Minard | France | 1869

Losses of the French Army in the Russian Campaign 1812-1813 (Minard, 1869)

Another Minard.  Probably the most famous statistical graphic of all time, the depiction of Napoleon’s march to and from Moscow is hard to avoid in the world of data visualization. It really is quite good, though.

The graphic elegantly depicts multiple data types (direction, latitude/longitude, troop numbers, distance, temperature and time) in a way that is immediately understandable. This method of flow chart is known as a Sankey diagram, though H. Riall Sankey would create his first nearly 30 years after Minard.