The Story Behind the Stats

140826 Statistics-01By Paige Ulevich

In graduate school, I made extra cash by tutoring other graduate students in statistics. I like to think that I was able to take a dry and reviled subject and show how it can illustrate fascinating insights into the world around us. In my humble opinion, people hate statistics because it is usually taught backwards.

The first mistake: the mathematical formulae are taught through drill exercises. Then, teachers use word problems to demonstrate how the function can be used. No wonder students struggle with statistical concepts when they are presented as esoteric calculations with no real-world applications. The best way to motivate my students was to answer the question that haunted them every moment they spent studying statistics, “What does it mean?

If you look at a number in context, statistics can tell us a story. Here are a few examples of the story behind the numbers.

Example 1

K12 Insight’s main office is in Fairfax County, Virginia. The county has approximately 7,600 families receiving Supplemental Nutrition Assistance Program (SNAP) benefits according to the NCES School District Demographics System, while the neighboring county I live in, Prince William County, has approximately 4,400 families receiving SNAP.

Which of these is the more affluent area?

If you answered Prince William County, you’re absolutely wrong. The percentage of families receiving SNAP in my home county is 7.3% compared to Fairfax County’s 5.2%. The median household income in Prince William County is $96,160 annually compared to Fairfax County’s $109,383.

If your eyes are starting to glaze over, think about it this way: how would your life change if your annual income increased by 12%? Would you move to a bigger house in a nicer neighborhood? If you lived in Prince William County, would you move to Fairfax County? What challenges would the schools in Prince William County face versus Fairfax County? How do they differ?

Example 2

Statistics help us recognize patterns. Below is a chart of parent survey responses for a district in Texas broken out by schools:

chart 1

Immediately, you see that parent concerns about alcohol and drug use at school increase as their children get older. Although this wouldn’t surprise most people, knowing how parent concerns shift while their children are in high school can help district leaders understand what messages to focus on.

Example 3

Statistics can also help us identify broken patterns. Recently, I was building a report for a district in New York and noticed that the distribution for the same question looked like this:

chart 2

Notice anything in the chart that contradicts the above pattern? If you think there must be a mistake in the percents for Elementary School B, I had the same reaction when I saw it. I double-checked the numbers, and they were correct for all the schools. My next thought was, “What is going on at this school?”

In my humble opinion, the moment when you see the story in the numbers and want to know more is when statistics become interesting!

Why do over one-third of the parents of Elementary School B believe alcohol and/or drug use is a problem at the school? What led them to this belief? After an online search, I learned that two staff members had recently been arrested for having illegal drugs on the school’s property.

Example 4

Lastly, at the risk of sounding even nerdier, my favorite thing about statistics is when it challenges the things we think we know!

I recently discovered a great blog about the visualization of data, and I’ve become fascinated with how the data differs from commonly held beliefs. This visualization shows the juxtaposition of popular herbal supplements and the evidence of their effectiveness. Although, I’m happy to see my dark chocolate and garlic intake are proven choices for reducing my blood pressure, I am dismayed to learn the benefits of glucosamine for arthritis is only slight.

Having been diagnosed with a form of arthritis in my 20’s, I have probably invested thousands of dollars in glucosamine supplements that likely had no more effectiveness than a placebo. Apparently, I need to start eating eggshell membranes although I’m not sure the best way to do that.

So how are you using data to actually make “data-driven decisions?” Are you looking beyond the sea of numbers to find the meaningful patterns, to find the anomalies and to challenge your beliefs?

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s