Data visualization made easy
Learn to autochart and filter with visual analytics
By Stuart Nisbet, SAS VP of Research & Development
John Wilder Tukey, a mathematician who first coined the term "exploratory data analysis," was right when he suggested that the idea of visualization helps us see what we have not noticed before. That is especially true when you are trying to identify relationships and find meaning in huge amounts of collected data. Sure, analyzing the data can tell the story, but wouldn't seeing the results help you more easily grasp the meaning?
Analyzing data and displaying the results with graphs and charts makes patterns, trends and outliers easily visible. For example, what if you had data on cell phone use? Using basic bar chart techniques, you could likely spot some interesting correlations. You might notice that areas with certain types of networks experience more dropped calls. Another analytic visualization could show opportunities for growth in a particular region.
Analytic visualizations are critical to gaining fast insights from your data. If sophisticated analyses can be performed quickly, even immediately, and results presented in ways that showcase patterns and allow for querying and exploration, people across all levels of your organization can understand and derive value from massive amounts of data faster than ever before.
Drag and drop – it's an autochart
So, it's clear to see the value offered by data visualization. But what about creating the visuals? Especially when working with large amounts of data, it can be difficult to decide which graph is best to use. In SAS Visual Analytics, intelligent autocharting produces the most appropriate visual based on what data you drag and drop onto the visual palette.
If autocharting does not create the exact visualization you had in mind, you can select a specific visual to build. However, when you are first exploring your data, autocharts are useful because they provide a quick view of the data. This automation opens up the world of visualization to business analysts and nontechnical users, enabling them to interactively explore and drill through data and display it in many different ways to answer different questions.
"What does this mean?" pop-up boxes also make visualizing your data easier by providing explanations of complex analytic functions that have been performed, as well as identifying and explaining the relationships between the data variables that are displayed. (See figure 1.)
Filter for added focus
When working with massive amounts of data, being able to quickly and easily filter the data is important. What if you only want to view data for a certain region, product line or some other variable? Filtering makes it easy to refine the information you see. In SAS Visual Analytics, you simply add a measure to the filter pane or select one that is already there, and then select or deselect the items to filter.
But what if the filter isn't meaningful or it skews the data in undesirable ways? One way to better understand the composition of your data is to use histograms. Histograms provide a visual distribution of the data with cues for how the data will change if you filter on a particular measure. This gives you an idea of the effect a filter will have on the data before you apply it to your entire analyses. (See figure 2.) Rather than relying on trial and error or instinct, you can use the histogram to help you decide which areas to focus on.
Share what you see
Creating data visualizations is all about communicating meaning. So share your ideas. Ask questions of others. Make observations. Easy-to-use collaboration capabilities promote idea sharing while saving valuable time. You can easily annotate screen captures of your visualizations and reports, then email them to others, who can add their thoughts as well. Or capture your comments via video and audio, and share them that way. Data visualizations are great for showing and sharing information.
Visualizing your data can be both fun and challenging. If you are working with big data, it is easier to understand information in a visual instead of a large table with lots of rows and columns. However, with the many visually exciting choices available, it is possible that the visual creator may end up presenting the information using the wrong visualization. In some cases, there are specific visuals you should use for certain data. In other instances, your audience may dictate which visualization you present. In the latter scenario, showing your audience an alternative visual that conveys the data differently may provide just the information that's needed for them to truly understand what it all means.
Bio: Stuart Nisbet is a Vice President of Research & Development for SAS. He directs the development of SAS Enterprise BI and SAS Visual Analytics products, iOS mobile application development, statistical and business graphics, device drivers, reusable component libraries for all SAS solutions, the SAS Output Delivery System, and the SAS Retail Space Management suite.