Visualization Methods
Last updated
Last updated
Line charts are the most common way to visualize temporal data. Time is taken on the x-axis and a quantity that is measured over time is taken on the y-axis. Multiple lines can be used to represent multiple categories. However, using too many lines can make the visualization cluttered and hard to read.
In such a case, we can use grouping, filtering or highlighting.
Another solution is to use faceting (small multiples) to represent different categories in different timelines:
The advantage is that the lines become clearer and more comprehensible. However, since they are smaller and no longer in the same chart, comparing lines becomes difficult.
Lines can be replaced with areas to make it easier for our eyes to perceive shape.
Another issue with line charts is the effect of aspect ratio. Aspect ratio is the width/height ratio. A change in the aspect ratio leads to a change in the line pattern, and some patterns may be more or less visible at certain aspect ratios. One solution to this is Banking to 45: set the aspect ratio such that the average slope of the line(s) is . However, this isn't always practical. A general rule of thumb is to test different aspect ratios and choose one that best conveys the message.
This is another graph that is used to visualize temporal data. It can be used to represent proportion changes over time.
The issue with area charts is that only the width is meaningful; temporal trends are hard to read.
A Percent Area Chart can mitigate this issue.
Heat Maps and Tables can also be used to represent temporal data. Instead of representing the actual values at a specific time, heat maps use a quantitative color scale with the color intensity as a channel.
Instead of color intensity, symbol size can be used to represent the same data.
The advantage of such visualizations over line charts is that they scale better i.e. we can use them to represent multiple object categories on the same chart without causing clutter.
However, color intensity and/or shape size aren't as effective as position (the channel used by line charts) while representing quantitative information. Also, since there is one row per object (in these alternate visualizations), it is easier to view individual trends, but it becomes difficult to compare trends.
The visualization below can be used to represent event data i.e. data that describes the time at which an event occurred along with event attributes.
This chart can be used to visualize the duration of an event (with start and end timestamps).
The lines in the chart above show how long the item took to be shipped.
As discussed earlier, temporal data can be sequential and cyclic. The previously discussed techniques show how to represent sequential temporal data. Cyclic/Periodic temporal data can be visualized using simple line charts, but when we want to visualize a lot of data at once, we can use a calendar representation.
Periodic data is, by definition, periodic. This means that there is a repetitive nature. This is not well-captured by the previously discussed charts, as they do not factor in continuity between periods (shown below).
Radial layouts can reduce temporal discontinuities (shown below).
An example: visualizing global temperature change:
We must use radial layouts with caution. It is usually more difficult to comprehend and make comparisons in a radial layout, when compared to linear layouts.
The color hue channel can be used to represent different categories:
