|
6 | 6 | - Evaluate existing charts for accessibility and readability |
7 | 7 | - Make improvements to charts to increase comprehension and accessibility |
8 | 8 |
|
9 | | -A chart is good if it allows the user to draw useful conclusions that are supported by data. Obviously, this definition depends on the purpose of the chart - a simple EDA chart is going to have a different purpose than a chart showing e.g. the predicted path of a hurricane, which people will use to make decisions about whether or not to evacuate. |
| 9 | +A chart is **good** if it allows the user to draw *useful conclusions that are supported by data*. |
| 10 | +Obviously, this definition depends on the purpose of the chart. |
| 11 | +A simple and disposable chart created during an exploratory data analysis process may be useful even if it is not nicely formatted and publication ready, because its purpose is to guide an interactive process. |
| 12 | +This is very different than a chart created for communicating with the public -- for instance, a forecast map showing possible paths and intensities of a hurricane that would inform resident decisions about storm preparation and/or evacuation. |
10 | 13 |
|
11 | | -Unfortunately, while our visual system is *amazing*, it is not always as accurate as the computers we use to render graphics. We have physical limits in the number of colors we can perceive, our short term memory, attention, and our ability to accurately read information off of charts in different forms. |
| 14 | +Comprehensive advice on creating good charts is difficult, too, because what works for one dataset may not work for another, even if the variable types are similar, because we have some established conventions that should usually be followed (for instance, time usually is placed on the x-axis, with a dependent variable on the y-axis), but there are usually situations where it is reasonable to break those conventions. |
12 | 15 |
|
13 | | -## Perceptual and Cognitive Factors |
| 16 | +Finally, what makes a chart "good" requires some additional knowledge beyond statistics and programming. |
| 17 | +To make good charts, we have to understand how those charts will be interpreted, which means we need at least some basic information about human perception and cognition. |
| 18 | +The human visual system is incredibly powerful - it has a bandwidth that would make even modern computers jealous, and many computations are performed instantaneously and without requiring any process management (e.g. the calculations happen so fast and so automatically that you aren't really aware that they're happening). |
| 19 | +This comes with some tradeoffs, though - evolutionary optimizations that ensure that you can spot predators quickly weren't as concerned with your ability to accurately determine the height of a two-dimensional drawing of a three-dimensional object. |
| 20 | +So, while the visual system has some amazing strengths and is a very useful medium to communicate about data, it is important to understand the limitations of the visual system's sensors, software, short-term and long term memory, and attention. |
| 21 | + |
| 22 | +We'll start with a short exploration of some foundational concepts related to perception in @sec-cognition-perception. |
| 23 | +@sec-design-process will discuss the design process and how to leverage the grammar of graphics to provide both the full data and visual summaries that highlight key features. |
| 24 | +Where available, this section incorporates conclusions from empirical studies of charts and graphs to guide design decisions. |
| 25 | +@sec-annotation expands on this discussion, demonstrating effective use of annotations to provide contextual information that can assist viewers with the interpretation of the data. |
| 26 | +Finally, @sec-checklist provides a guide to evaluating graphics for clarity, effectiveness, accessibility, and common design pitfalls. |
| 27 | + |
| 28 | + |
| 29 | +## Cognitive and Perceptual Foundations {#sec-cognition-perception} |
| 30 | + |
| 31 | +Before we discuss how to create good charts, there is a certain amount of background information that must be considered. |
| 32 | +Charts make use of the visual system within the human brain, which means we need to understand some basic attributes of human perception and cognition in order to make the best use of this "wetware" [@wikipediacontributorsWetwareBrain2025] processing power. |
| 33 | + |
| 34 | +<!-- Basic overview of components of the visual system - eye, optic nerve, processing centers of the brain, visual cortex (hardware) and software --> |
14 | 35 |
|
15 | 36 | ### Preattentive Features |
16 | 37 |
|
@@ -422,7 +443,10 @@ The aesthetic mappings and choices you make when creating plots have a huge impa |
422 | 443 |
|
423 | 444 | [^10-graphics-4]: See [this paper](https://doi.org/10.1080/10618600.2016.1209116) for more details. This is the last chapter of my dissertation, for what it's worth. It was a lot of fun. (no sarcasm, seriously, it was fun!) |
424 | 445 |
|
425 | | -## Representing Data Accurately |
| 446 | + |
| 447 | +## Designing Good Graphics Using the Grammar {#sec-design-process} |
| 448 | + |
| 449 | +### Representing Data Accurately |
426 | 450 |
|
427 | 451 | In order to read data off of a chart correctly, several things must happen in sequence: |
428 | 452 |
|
|
537 | 561 | Extra dimensions and other annotations are sometimes called "chartjunk" and should only be used if they contribute to the overall numerical accuracy of the chart (e.g. they should not just be for decoration). |
538 | 562 |
|
539 | 563 |
|
| 564 | +## Evaluating Graphics: Effectiveness, Clarity, and Comprehension {#sec-checklist} |
| 565 | + |
540 | 566 | ## References {#sec-good-graphics-refs} |
0 commit comments