Data Visualization: Marks and Channels

June 1, 2021

Data can be organized in two main groups: categorical (or nominal) vs ordinal. Coincidentally, the ways in which the data can be represented can also be grouped in things that naturally represent order and things that naturally represent identity.

By identifying your data types, and using the proper channels to show them, you can make your data visualization clear, which is the ultimate goal.

This post focuses only on how to map raw data to visual components on a screen, but much has ben discussed about the presentation of those components, with mainly two camps: you can focus on minimalizing the amount of pixels on the screen to make the data be the king (refer to the work of Edward Tufte) vs delivering the message with the aid of visual representation (maybe David McCandless)

Data types and their graphical elements

The type of the data is its structural or mathematical interpretation, At the attribute level, what kinds of mathematical operations are meaningful for it? What kinds of comparisons do those measurements support?

The major disinction is between categorical/nominal versus ordered.

The core of the design space of visual encodings can be described as an orthogonal combination of two aspects: graphical elements called marks, and visual channels to control their appearance

Categorical

examples:

Ordered

examples:

Data types are not mutually exclusive, nor restrict you from using a numeric value as a nominal quantity, they just allow us to categorize the data and know what kind of visual representation is useful.

The graphical elements