Design Educators Community

Scale Matters

November 11, 2014 / By Frank Armstrong

Designers, educators and students are bound by ethical considerations to convey information accurately. When designing data visualizations, it’s essential that we are concerned with the accuracy and integrity of the relationship between visual forms and the underlying data the forms represent. Our visual perception is remarkably sensitive and accurate when evaluating one-dimensional relationships, such as the lengths of lines or bars. However, comparing two-dimensional relationships is much more difficult because our internal process is first perception, then cognition.

Vertical bar charts allow users to compare the height of equal-width rectangles, accurately representing one-dimensional data (one variable – a single set of values) by extending a two-dimensional shape in one dimension. A problem arises when a designer attempts to represent one-dimensional data by proportionally scaling both width and height of two-dimensional shapes, such as circles. For example, let’s say that we need to visualize the relationship between two data values: 4 and 8. If we set the diameter of one circle to be twice the diameter of the other, the surface area of the larger circle is actually four times larger than the smaller circle – not a very accurate visualization. So, how would designers or design students easily scale two-dimensional shapes that accurately represent data relationships?

Scale Matters is a responsive mobile web app that was created for graphic design students at California State University, Chico. A user inputs two data values and the diameter of the initial base circle in a visualization. The app calculates the diameter of the second circle that would produce an accurate visual relationship between the surface areas of the two circles.

sm_1 sm_2

Access the Scale Matters app from an iPhone 5 or iPad.

AIGA encourages thoughtful, responsible discourse. Please add comments judiciously, and refrain from maligning any individual, institution or body of work. Read our policy on commenting.
  • Amy Rae Fox

    As a graduate student, I study the cognitive processes involved in comprehending data visualizations, and how these processes might inform the design process. I’m excited to see the Scale Matters app, as it directly addresses a common type of ‘inadvertent visual lie’. When a difference in data values is 1 dimensional, I tend to represent it via a 1 dimensional visual change (ie. a line gets longer, a point moves on X axis)… but this app opens a door for me to accurately calculate a multidimensional representation of that same change. (Well, I guess it is the app actually doing the calculation 😉

    In any case, I’m eager to make use of this! I wonder – does anyone have an opinion the appropriateness of using a multidimensional representation for a 1 dimensional change? Is this misleading to viewers, even if calculated accurately? Or is it a powerful tool for adding visual interest?