Nadia Chung, a biostatistician working on the California Teachers Study, uses her mouse to drag a new variable into the graph displayed on her computer screen. Immediately, the visualization changes to include the new data.
Data visualization refers to representing information in the form of charts, graphs, diagrams or pictures. Research teams across population sciences and epidemiology are using data visualizations more and more to explore information. Whereas reading a table of numbers to discern key findings takes a specific set of skills, when data is visualized the salient data points are immediately apparent.
“Using Tableau has transformed what we’re able to see, and the speed at which we can see it,” Ms. Chung tells me. “People ask us all the time, ‘Can I just see the data?” Before Tableau, ‘seeing the data’ meant reviewing rows of data. But now when someone asks to see the data, there’s a dynamic visualization for them to view.”
Below, a visualization built using Tableau displays the most common cancers diagnosed in the California Teachers Study cohort.
Using Tableau has helped remove some of the bottlenecks that used to delay data exploration. As Emma Spielfogel, a Research Analyst on the team, explains: “You don’t have to be an expert statistician or data analyst to explore the data. Since Tableau is a point-and-click program, meaning it does not require any coding or technical knowledge, researchers who are not as familiar with writing code can still have the opportunity to explore our data. Anyone can create insightful—and pretty—graphs with the click of a button.”
As the National Cancer Institute's Working Group on Data Science recently recommended, data should be made more broadly available to the research community for sharing and analyzing. Today, the CTS is using data visualizations to help researchers and study participants explore study data. Visit our Study Population page to explore additional CTS visualizations! To request access to analyze or explore CTS data, please visit our For Researchers page.