Making data sexy

Author(s): Kim Medaris Delker

Johnny Depp and Beyoncé, watch out. There’s a new sexy out there, and it has more to do with how well you can analyze massive amounts of data than how you look in an action movie or music video.

Twenty-five years ago, the words “sexy” and “statistics” probably weren’t at all linked. But a quick Internet search of the two terms yields multiple hits in news stories and blog articles, such “Data Scientist: The Sexiest Job of the 21st Century” and “Statistics is the Sexy in Science.”

And to further spotlight the field, 2013 has been declared “The International Year of Statistics” by several statistics societies and organizations.

So just what makes statistics so hot?

“By the 1980s, statistics had become deeply entrenched in many areas of academia and industry, like medicine, psychology, marketing and chemical engineering,” says William S. Cleveland, the Shanti S. Gupta Professor of Statistics. “Today, statistics is critical to retailing and banking, and it’s how social media sites like Facebook analyze their members’ data intensively. The list goes on and on.”

Rebecca Doerge, head of the Department of Statistics and the Trent and Judith Anderson Distinguished Professor of Statistics, says that in the last couple of decades, people have started to view statistics in a new light, greatly increasing its value.

“If you collect enough data, you’re going to find trends, and what has happened is that marketing has really taken off and people are seeing the impact of that knowledge, which is resulting in data-driven decisions in business,” she says.

Decades ago, the focus was on developing new methods and technology to collect data, but with today’s rapid advances, that’s changed.

“With today’s computing tools and software, anybody can collect data, but just collecting a lot of data is useless if you don’t know which data to collect, how to design a survey or experiment and how to interpret the results,” Doerge says.

And that’s where statisticians and data scientists come in, making them a hot commodity in a world that seems to find new ways of collecting data on us by the second — via our smartphones, computers, in a store, in a car.

Cleveland, a “big data” guru, says the real change began about 1990 when larger and more complex datasets began appearing, which challenged all areas of data science and analysis.

Whereas the average person not trained in statistical analysis might quickly become overwhelmed at the amount and complexity of the data, for a data scientist, it’s a welcome — and rewarding — challenge.

Students who major in statistics can look forward to a very rosy future, no matter if they earn a bachelor’s, master’s or doctoral degree. The median salary of a statistician is $72,830 a year, and the employment of those in the field is projected to increase by 14 percent from 2010 to 2020, according to the Bureau of Labor Statistics.

Doerge says that Purdue statistics graduates can be found working at some of the country’s largest companies, from tech companies like Yahoo or Google to industries such as health care, banks, insurance, finance and many others.

“Companies fight over our students, so companies get competitive with each other when making offers to them,” Cleveland says. “And the great thing about statistics is that you can take the skills and go from one industry to another one — like health care to finance — since the core skills you need don’t change.”

George McCabe, associate dean for academic affairs in the College of Science and a professor of statistics since 1970, believes that a knowledge of statistics is important not just for those whose job it is to analyze data, but for the layperson as well. Although numbers can be intimidating to someone without a proficiency in math or statistics, “quantitative literacy” or “quantitative reasoning” is essential to understanding today’s data-driven world.

“Statistics are everywhere, from quality control — where you expect a product to work the way it’s supposed to — to medicine and studies on health care,” he says. “People need to be critical of this information, so knowledge of statistics is essential.”

Doerge, who began as a mathematician and then switched to statistical genetics (specializing in what is now known as statistical bioinformatics), says that many who choose the field — far from the green eyeshade stereotype — do so because of the impact they can have.

“It’s a very social field, and we’re doing a good job at taking the nerdiness out of statistics’ reputation by including introductory statistics courses for all undergraduate nonmajors and making the subject relevant to their daily lives,” she says.

It’s also a field that is welcoming to women — Purdue’s department is about 50 percent female.

Doerge sees a bright future for statistics as technology becomes even further advanced. For example, she envisions the ability to use statistics collected from your car (such as how fast you drive) to determine your insurance rates or linking the groceries you buy and what you eat with your medical records, resulting in truly personalized medicine.

“Statisticians will always be needed. We’ll be the last people standing,” she says with a smile. “After all the data are collected, someone still has to tell you what it all means.”