Connecting Data and Images#

🎨 Data Visualization 🎨 Visualization Principles & Design Lesson 002

Important

✨ AI-generated content. This page was written with the assistance of an AI language model and is provided as a learning aid. Despite careful review, it may still contain mistakes, omissions, or out-of-date information. Whether you are new to the topic, a team lead, or a senior practitioner, treat it as a starting point rather than an authoritative reference: read it critically and independently verify anything you act on (code, commands, figures, and factual claims) against official documentation and primary sources before relying on it.

From numbers to visual properties#

A visualization works by mapping data to visual properties — turning a number into a bar’s height, a category into a colour, a pair of values into a point’s position. Understanding this mapping, the connection between data and images, is what lets you read visualizations critically and build them deliberately, rather than treating charts as magic.

Visual encodings#

Every visualization encodes data values as visual attributes. The common encodings:

Position — where a mark sits (a point’s place on an axis, a dot in a scatter plot). Position is the most precisely readable encoding — the eye judges position extremely well, which is why scatter plots and dot plots are so effective.
Length — how long a mark is (a bar’s height or length). Also read accurately, which is why bar charts are excellent for comparing quantities.
Angle and area — as in a pie chart’s slices. Read less accurately by the eye — people struggle to compare angles and areas precisely, which is a known weakness of pie charts.
Colour — hue (different categories) or intensity (a value’s magnitude, as in a heat map). Effective for categories and for showing a third dimension, but read less precisely for exact quantities.
Shape and size — distinguishing categories or encoding a value in a mark’s size.

The choice of encoding determines how accurately and easily the data can be read — some encodings the eye judges precisely, others only roughly.

Why the encoding matters#

Because encodings differ in how accurately they are read, the same data is more or less legible depending on how it is encoded. A quantity comparison is clear as bar lengths (accurately read) but muddy as pie slices (poorly read); a trend is clear as position over time (a line) but lost in a table. Choosing an encoding the eye reads well for the message you want is the foundation of effective visualization — a chart that encodes its key comparison in a hard-to-read attribute fights its own reader.

Reading visualizations critically#

Understanding encodings also makes you a critical reader of charts. Knowing that the eye reads position and length accurately but angle and area poorly, you can spot when a chart’s design obscures rather than reveals — a pie chart where a bar chart would be clearer, a 3D effect that distorts areas, an encoding chosen for decoration over legibility. The connection between data and image is where both good design and misleading design live.

The caveat#

The mapping from data to image is where distortion enters, deliberately or carelessly. The same encoding can mislead if its scale is manipulated (a bar chart with a truncated axis exaggerates differences), if area is used to encode a value people will read as one-dimensional (doubling a circle’s radius quadruples its area, overstating the value fourfold), or if colour implies an order that the data does not have. Every visualization is a set of encoding choices, and those choices can serve clarity or distort it — reading and making charts both require attention to whether the encoding represents the data faithfully. The next lesson turns to what makes a visualization genuinely powerful.

Hint