We all know visualizations are an excellent way to summarize information, but some visualizations are better than others at condensing certain data. Sometimes dimensions don't map, data density doesn't represent well in the , and a bar chart is better than that line chart you wanted to make. This is because visualizations don’t just represent data, they also represent information’s context. And making sure the context comes across to the viewer is hard to do.
One part of context is the data’s continuity. There are three main types of continuity that data can have.
Discrete – These are values like Number of Bees. They have a set amount of accuracy and each value type is numbered, not named.
Classified– These are values like Gender or Nationality. These are also discrete, but the values are different enough have a name.
Continuous – These are values like Pressure or Time. They can be subdivided into infinitely small increments.
Now that we know we have these types and we know what they are, how do we coherently visualize them?
Let’s first look at bar charts and line charts. The height of each column in a bar chart can vary indefinitely. This means that the y-axis can be continuous, because there are no set ratios that the height any bar must be, charts have individual separated columns. This means that each bar must represent individual separate data points, making the horizontal axis good for discrete and categorical data.
Unlike bar charts, line charts have a line connecting individual data points and it has points with indefinitely varying vertical positions. The line implies continuity, so the x-axis and y-axis are good for continuous data.
Choose your visualizations carefully, it’s the difference between conveying understanding and being underwhelming.