Complex relationships

Data caused a shift

Today, every manager, analyst, engineer, or even an educator needs to have the basic skill of visual communication. This shift is because of the ever-rising transfer and translation of data occurring offline, online, and in real-time.

Previously, you constructed relationships and plotted them on charts. But, as an analyst, you will have to get your hands dirty and go all exploratory when it comes to surveying much more complex relationships.

Learning to communicate complex relationships effectively is key to data analytics. Effective data visualisation is imperative to understanding different data types and to communicate the presented complex relationships between the data.

Let us explore ways to convey the most complex relationships in the most simplest forms.

Complex relationships

The term complex relationship often refers to the study of complex systems. This approach is used in data visualisation to identify how more than one set of relationships between the values or data points work.

For example:

A scatter plot can easily illustrate a correlation between two variables (one variable per axis). However, what about:

• relationships that are complex in nature and have three, four, or more variables?
• relationships between physical space (such as a real-world location) and a value at that location?

Some other set of relationships that can be complex to visualise are:

• displaying the relative population density, GDP, air quality, temperature
• mapping the voting preference, car ownership, unemployment rates, or other data that can be grouped by a political region on a map
• illustrating bond yield rates that change year on year, and also on the length of the bond.

How to visualise complex relationships

Before we move on to the details, let us observe the 3-D chart here from The New York Times. This is an interactive chart, so please feel free to play with the features and reflect on the several relationships that the analysts have tried to establish.

Observe: The Yield Curve: A 3-D chart [1]

This is an excellent example of how a relationship as complex as the economic future can be explained (with the help of a yield curve) to an audience who are not necessarily an expert in the field.

How can we do that?

Typically, the tools data analysts use to display such complex relationships vary and depend on the organisation. However, to visualise such relationships, you can choose to display your data in several types of maps and graphs. Some of them are as follows:

• 3-D plots
• spatial data
• heatmaps
• other chart types

We will be exploring them one after the other subsequently. If you are like many others who get anxious when adding animations to your charts and prefer to have an alternative, you can even combine these techniques and use multiple charts to create a trend display.

Why do you think?

Tech giants and electric vehicle companies such as Apple and Tesla are moving towards a more data-visualisation-driven reality for their products.

What kind of complex relationships do you think would be required to visualise and in which products?

References

1. Aisch G, Cox A. A 3-D View of a Chart That Predicts The Economic Future: The Yield Curve [Article]. New York Times; 2015 Mar 18. Available from: https://www.nytimes.com/interactive/2015/03/19/upshot/3d-yield-curve-economic-growth.html