Making Sense of the Scribbles

Working through data is just like the beginnings of a doodle. The journey starts with an absent-minded scribble. Moving randomly with no particular direction, chaotic and abstract at the same time. Moving up and down and right and left. You have no idea what you’re doing, and yet the pen keeps scribbling.

Here is the point where we usually start. Upon viewing that first spreadsheet, it doesn’t make much sense. You’re handed a pile of numbers bunched up into tables that resemble a huge mess, an absent-minded doodle.

Then comes the ‘a-ha!’ moment. Scribbles turn to squiggles. Patterns and repetitive waves in your data emerge, and the puzzle becomes clearer piece by piece. As you move the pen in all manner of directions, the hope is that you find one that makes the most sense.

Now you can turn your numbers into charts and graphs, pivot tables and summary statistics. Then you take a step back to breathe in the colors that present before you. It’s beautiful, but it remains incomplete.

Of course, by now you’ll have the urge to lift the pen, but another part of you is curious enough to see what comes next. You look back into your scribbles to find that you’ve been drawing in specific patterns all along. Then you take a leap, choose a direction, and start drawing that straight line.

When everything starts to make sense, you’ve gained a sense of understanding of what came from raw digits and letters. That’s what you call insight.

Most people are unaware of the power of data-driven decision making, especially in politics. We live in a world of data. A Twitter rant about one pothole on Michigan Ave. can be just as insightful as hearing a complaint directly from a local. The only difference is on a social platform; you can get thousands of those rants in a short amount of time, identify the patterns from the data, illustrate it using meaningful visualization, and decide what course of action to take from the knowledge that you’ve acquired.

While data is an essential tool, it can also be deceiving. Darell Huff wrote in his book, “How to Lie with Statistics,” elaborating on this topic. He mentions, “a well-wrapped statistic is better than Hitler’s “big lie” it misleads, yet it cannot be pinned on you.”

It doesn’t take much skill to lie with statistics and get away with it. Take the recent Presidential election; remember how many polls predicted Hillary Clinton’s victory? Although it’s a big accusation to claim that pollsters deliberately lie about their data. In most cases, they were just misled by the data they interpreted. They scribbled, decided on a direction, and were confident enough in their direction to finally draw a straight line.

“It ain’t so much the things we don’t know that get us in trouble. It’s the things we know that ain’t so” – Artemus Ward.

That is why it’s important to supplement quantitative data with qualitative data. Combining years of insights from the ground with numbers that we obtain from a 1000-feet view of our target paints a robust, and final picture. Only when we complement first-hand knowledge and experience with significant numbers that we can truly create a work of art from meaningless scribbles and lines.

 

About the Author | Yusri Jamaluddin is a Data Analyst for Grassroots Midwest. He is currently pursuing a Master of Public Policy degree from Michigan State University. He holds an MS in Business Analytics and a BS in Information Technology and Web Science from Rensselaer Polytechnic Institute, New York. He is highly passionate about data analytics and visualization.

Get updates when we post.