Marquette University Trains Students in New Data Science Program

Data scientists are in demand, and Marquette is stepping up to train students for those positions.

by Kathleen Gallagher, Milwaukee Journal Sentinel / September 13, 2016 0
Capella University plans to train data analysts who have both technical and storytelling skills. NEC Corporation of America/ Flickr CC by 2.0

(TNS) — Marquette University professor Gary Krenz was working out on a weight machine at the school's recreation center when a young math and computer science major approached him.

Sam Scheel wanted to know if the rumors were true. Is Marquette going to create an undergraduate data science major?

It was a pressing question for Scheel, who was trying to cobble together data science expertise across three different departments -- math, statistics and computer science. It's also a question being asked more and more by companies scrambling to find workers who can analyze and glean insights from the mountains of data being created by computers, cameras, sensors and other devices.

"With the proliferation of data sets from the explosion of the internet of things, demand for data scientists will grow exponentially," said John Philosophos, a business development partner at Great Oaks Venture Capital LLC who lobbied Marquette to act on that demand. "This is an acute national problem, and it's exciting to see Marquette on the forefront of the solution."

Marquette began this fall offering an undergraduate data sciences major. Scheel, who is now a junior, happily signed on.

"It really fit what I was looking for," he said.

The Bureau of Labor Statistics doesn't list "data scientist" as a job, likely because it's too new, Krenz said. But the number of market research analysts, a common job for data scientists, is projected to grow by 19% during the 10 years ending in 2024, vs. 7% for all jobs overall, according to the Bureau of Labor Statistics.

A recent Monster.com search for "data analytics" turned up more than 1,000 jobs. Data scientists on average earn an annual salary of $113,436 in the United States; $90,016 in the Milwaukee area, according to Glassdoor Inc.

"Data analytics is taking over every industry, and these are extremely high-paying jobs. This is the degree every university should create," said Andy Walker, chief executive officer of Chicago-area online retail intelligence provider Channel IQ. Although neither attended Marquette, Walker and Philosophos put it among the schools they have been prodding to create a data sciences major.

"We're in the first couple years of the development of a discipline that I think is going to be around for a long time," said Jennifer Priestley, a professor of statistics and data science at Kennesaw State University who is credited by Datanami with starting the first data science doctorate program.

Five years ago, there were very few data science programs, Priestly said. But now, whether they're housed in colleges of science, math, engineering or business, such offerings are growing rapidly, she said. And they're attracting bright, young students such as Scheel.

In kindergarten, Scheel discovered that he loved school -- and that he loved math even more. At Port Washington High School, he blew through the curriculum until his senior year, when he ran out of math classes to take.

When he entered Marquette, Scheel majored in math and minored in computer science. He embraced statistics and probability, the underpinnings of data science. He struggled, however, with some of the computer science classes. While it was interesting to learn about the inner workings of computers, and the ins and outs of operating systems, Scheel was pulled in another direction.

"I enjoyed my stats class so much I just wanted more," he said.

Now he's got it. And Marquette's would-be data scientists have a clear, defined path that includes classes such as:

* "Statistical Methods" provides students with the toolbox that helps data scientists extract insights from patterns in mountains of data. This is how a data scientist might determine, for example, that when the vibration around the machine increases to a certain level, it is more likely to fail in the next 30 days.

* "Data Structures and Algorithms" teaches students about the different types of databases, and the mathematical equations that capture multiple variables to describe a result. A graduate who completed this class might recommend his employer use a distributed database, for example, the type that spreads data redundantly over many locations and is an important reason why services like Google and Facebook don't fail. This graduate would likely know as well how to use algorithms to collect and analyze data, the way Madison-based HealthMyne, for example, uses the formulas in its tool that helps radiologists determine the most effective treatments for lung cancer.

* "Data Mining" is a process of gleaning insights and patterns from massive amounts of information. Retailers use it, for example, to develop very detailed profiles of customers. Target Corp. used data mining in a well-known instance several years ago when it was able to determine, with unnerving accuracy, which of its customers were in their second trimester of pregnancy.

The strong demand for data scientists isn't close to being filled, with predictions of a shortage of 170,000 by 2018, said Walker, who previously co-founded 42six Solutions, a start-up that was acquired by Computer Sciences Corp.

"We were feeling the shortage at my last company, and we feel it at Channel IQ," Walker said. Yet universities are generally creating master's and doctorate programs, which are lucrative from a cost standpoint, but graduate small numbers of students, Walker said.

There are about 100 graduate data science programs across the country that Kennesaw State's Priestly tracks, she said.

The number of undergraduate degrees in statistics -- a key component of data science -- has nearly doubled in the last four years, making it the fastest-growing degree among all science, technology, engineering and math degrees, according to AMSTAT News, the membership magazine of the American Statistical Association, in its July 2015 issue.

DataScience.Community, which tracks activities in the discipline, counts fewer than 50 programs that offer data science undergraduate degrees on its self-reported list.

Companies and markets aren't waiting for academics to catch up. Data science boot camps have moved to fill the gap, offering 12-week training sessions, 12-month finishing schools and other options, according to Datanami, a big data news portal.

The problem for many schools is that data science crosses three disciplines: computer science, math and statistics, Walker said. It can be a major challenge to cut through three departments' bureaucracies and politics to create a unified major, he said.

Marquette, however, which has a mission of positioning itself for the jobs of the future, moved quickly after he and Philosophos approached President Michael Lovell in early 2015, Walker said. The task was made easier because all three disciplines were already in the same department, Krenz said.

Scheel, the undergraduate who is now majoring in data science, says he's pleased to have an official way to communicate his knowledge of statistics and a better chance he will find a career that uses more math.

"I'm still deciding what I want to be," he said. "But the major allows me to do what I want to do in school -- and probably outside of school, too."

©2016 the Milwaukee Journal Sentinel, distributed by Tribune Content Agency, LLC.