Guideline: The representation of numbers, as physically measured on the surface of the graphic itself, should be directly proportional to the quantities represented.
Source: Edward Tufte, The Visual Display of Quantitative Information, 2nd Ed., Graphics Press, 1991.
Are there any conditions where this rule may be relaxed? If this rule could not be relaxed in any situation, metro maps would be considered as big lies.
I’m not sure that Tufte’s Lie Factor guideline is valid in all circumstances, it is certainly too general. For instance, people have difficulty in making quantitatively accurate comparisons of the areas of different circles [Mei73]. If people cannot tell what the difference is between circle sizes, then any guideline that proscribes that they have to be accurate within a given tolerance is seriously in doubt. Certainly, given that humans can discern differences in length more accurately than differences in area [CM84], having the same tolerance for all visual variables would seem to be too simplistic. Generally, Tufte’s work is very interesting and useful in terms of subjective design, but it is not based on scientific evidence, hence the guidelines and rules he states should be viewed with a degree of scepticism.
Designers of metro maps do not intend to visualize quantitative data such as distance between stations. Hence I’d argue that the lie factor, whether valid or not, does not apply to such schematic maps.
[Mei73] Meihoefer, Hans-Joachim. “The visual perception of the circle in thematic maps/experimental results.” Cartographica: The International Journal for Geographic Information and Geovisualization 10.1 (1973): 63-84.
[CM84] Cleveland, William S., and Robert McGill. “Graphical perception: Theory, experimentation, and application to the development of graphical methods.” Journal of the American statistical association 79.387 (1984): 531-554.