Scatty.com

5 Fundamentals Of Data Visualization

Big data and analytics are here. In fact, they govern the world, and ignoring that is no longer an option.

In order to keep up with the huge piles of data that you can find in every industry and profession, we all need powerful and reliable tools. More often than not, those tools are found in the realm of data visualization.

If you are going to work closely with data visualization resources, then you will want to remember the fundamentals. They can help you avoid many common mistakes and extract as much value as possible from your efforts.

Top 5 Ways to Make Money on Facebook
Top 5 Ways to Make Money on Facebook

Data Cognition and Perception

The fundamental purpose of data visualization is to make it easy to understand abstract data at a glance. There are countless tools available, all built on the principles of cognition and perception. Can people understand the representation, and is it leading to meaningful conclusions?

To master cognition and perception, any visualization can be analyzed by asking a few key questions:

  • Is a clear relationship on display?
  • Is the data representation accurate?
  • Can you easily compare quantities?
  • Is it obvious how the information can and should be used?

If your visualization scores an easy “yes” to each of these questions, then it is successful in terms of cognition and perception.

Design Evaluation

The next fundamental utilizes an evaluation to ensure that the visualization is accomplishing its goals while remaining an accurate and reliable display of information. With so many visual representations available, it’s difficult to distill evaluations into a single checklist, but a general set of criteria can help you build a more specific evaluation for your project: 

  • Scale: Inconsistent scaling on a graph can distort the data representation.
  • Accuracy: Always triple-check that the data is accurate in the final visual.
  • Convention: Follow visual conventions to avoid confusion.
  • Cherry picking: Never exclude data to alter the conclusions.
  • Bias: Look for signs of bias, whether intentional or not.

Depending on the visual representations, you may also need to evaluate how easy it is to read the presented information and whether or not the visualization is leading to reasonable conclusions (as the saying goes, correlation does not equal causation).

User-Centered Design

The third fundamental that we are exploring is one of the most popular philosophical orientations for data visualization.

The concept is simple. How does the end user interact with the information represented?

As an example, a visualization of user statistics for a smartphone app might be used to help developers make decisions about the next set of updates. If you’re preparing this visualization, then it’s important to remember that the developers are not the ones who ultimately use this information. The data is informing development for the sake of the end user, and data representations should remember that.

Explanation vs Exploration

When dealing with data, there are two different things that happen. During analysis, you explore data to see what it can tell you. When you present the data, you explain the findings to others.

Visualization is involved in both of these aspects. When exploring data, you can use whatever representation tools make sense to you until you find something worth sharing. For explanation, visualizations have less freedom, as they need to present clear information to your audience.

It’s easy to get stuck on explanation and forget about exploration. Manage your time and explore the data as deeply as you can before honing in on the best solutions for explanation.

Communication

The fundamentals so far provide ways to approach visualization and enhance its value. The final pillar looks at data visualization from the other direction. There are two purposes to data visualization: making sense of the data and communicating.

When focusing on communication you need only remember two things. Make clear points, and aim for your audience. With a large data set, it’s easy to draw many conclusions, but when you present information (especially to non-experts), it’s essential that you distill everything into simple and clear points.

Catering to your audience is also vital. When one nuclear physicist presents findings to a room full of other nuclear physicists, the visualizations will look very different than if they are presenting that information to a board of investors (who presumably pay for the research). Always rethink the visualization from the perspective of your audience. It’s a simple but essential step in the process.

Leave a Reply

Your email address will not be published.