Data Disguises

With Halloween right around the corner, costumes and candy are on our mind! We’re feeling pretty nostalgic for the days of themed office gatherings and thought you might be, too. So, we thought we’d use this spooky season to share more about one of the scariest types of costumes – data disguises!

Let’s pretend it’s the week before Halloween 2019, or maybe 2023, and you’re excited for a treat-filled gathering at work. Before the office party, you do a little research and find the #1 Halloween candy nationwide is Skittles. You buy enough Skittles treat packs for your colleagues and head to work, ready to excite your whole team with a tasty rainbow sugar rush. At the end of the day, lots of Skittles bags remain and the evaluator part of your brain is left wondering how the nation’s #1 candy didn’t fly out of the bowl.

Enter: data disguises. When you return to your big data set, you realize figures are available for national and state level Halloween candy favorites. While Skittles top the national chart, the #1 candy for Maryland is actually Reese’s cups. In Virginia, #1 is Snickers, and DC’s top contender is M&Ms. Just by digging a little deeper, you uncovered regional differences in candy preferences. Although the dataset is limited to national and state averages, your organization might have even more variables to unmask, such as differences by age ranges, race, ethnicity, zip code, and the list goes on and on. It turns out, simply looking at the national average and settling on Skittles would leave many of your colleagues sorely disappointed.

Candy may seem like a trivial example given our challenging moment in history.  Still, we’re hoping it’s been a fun way to talk about digging deeper into your data, exploring differences between subgroups, and revealing what your average results may be hiding.  Pushing beyond those averages is a crucial step in discovering meaningful data stories.

This Halloween season, we challenge you to look more closely at an important data point that you collect at your organization. 

How can you break down your data to tell a more complex story? 

How can you use this information to spark data-driven conversations with your colleagues, your leadership, and your board? 

How can you be using your data to identify inequities and advance social justice?

Because ultimately, this is about so much more than candy.