This chapter traces the role of data in the initial rise of graphical methods around the early 1800s. We start with the very idea of numbers used for some wider purpose, and considered what we now call “data”. We focus attention on one important participant in this story: André-Michel Guerry [1802–1866], who used an “avalanche of data” and graphical methods to help invent modern social science.
![]() |
Figure 3.1: Alfonsine TablesA page from the Alfonsine Tables giving times of observations of
celestial events. |
![]() |
Figure 3.2: Sex ratioArbuthnot’s data on the ratio of male to female births. The average ratio, 1.07, is shown by the upper dashed line, compared to the solid line at 1.0. The curved line shows a smoothed (loess) curve through the points, with a shaded confidence interval.Source: © The Authors. Rcode: 03_2-arbuthnot-gg.R |
![]() |
Figure 3.3: First comparative shaded mapsGuerry and Balbi’s 1829 Statistique comparée de l’état de
l’instruction et du nombre des crimes. Top left: crimes against persons;
top right: crimes against property; bottom: instruction. In each map,
the departments are shaded so that darker is worse (more crime or less
education). |
![]() |
Figure 3.4: Quetelet’s (1836) maps of crimeCrimes against property (left), crimes against persons (right). |
![]() |
Figure 3.5: Suicide method and ageHistograms of the age distributions of suicide by pistol vs. by
hanging, for males. |
![]() |
Figure 3.6: Ranked listsRanking of crimes against persons in seven age groups. Connecting
lines show some noteworthy trends. |
![]() |
Figure 3.7: Reproduction of Guerry’s six mapsAs in Guerry’s originals, darker shading signifies worse on each moral variable. Numbers give the rank order of departments on each variable. Data source: André-Michel Guerry, Statistique morale de l’Angleterre comparée avec la statistique morale de la France, d’après les comptes de l’administration de la justice criminelle en Angleterre et en France, etc. Paris: J.-B. Baillière et fils, 1864.Source: © The Authors. Rcode: 03_7-guerrymap.R |
![]() |
Figure 3.8: Enhanced plotScatterplot of crimes against persons versus literacy from Guerry’s data, one point for each department. The black line shows the linear regression relation; the gray slightly curved line shows a nonparametric smoothing. Points outside a 90 percent data ellipse are identified by department.Source: © The Authors. Rcode: 03_8-guerry-bivar.R |
Copyright © 2021 Michael Friendly. All rights reserved.
friendly AT yorku DOT ca