You’ve got the data, where’s the story? How to find interesting patterns and do analysis

By Kathryn Sharkey

Data can be overwhelming, so when it comes to finding the story in it all, Dan Keating from the The Washington Post and Jacob Fenton from the Sunlight Foundation agree: start small.

Review the information and look for what’s interesting and jumps off the page. Fenton’s three-step guide to a story is to:

• Notice an event
• Perceive a pattern
• Explain the cause

When analyzing the data, Fenton said to pay attention to trends over time, geographic comparisons, and the key variable, which explains the differences. The best way to find the key variable can be through grouping, he said. Keating and Fenton said to pay attention to the distribution of the data rather than focus on the average. You can glean more meaning and simplify the data looking at what the changes mean for the top and bottom values. For context, Fenton said to look at what the data is compared to, even if it is zero.

Finally, Keating stressed to not let data problems sink the story. Focus on what you know and be clear about what you don’t know. Don’t claim what you can’t prove. Missing data or a lack in data can be as or more important than what you do have. Share the findings with a reasonable response time and be careful stating something as a cause, he said. Correlation alone does not establish causation. And when doing your data analysis, focus only on the data, Fenton warned, because it’s easy to miss something you’re not looking for.

Kathryn Sharkey is a journalism student at Southern Methodist University