School & District Management

In Analytics Economy, Demand for Data Scientists Outpaces Supply, McKinsey Says

By Benjamin Herold — December 08, 2016 5 min read
  • Save to favorites
  • Print

As analytics and automation transform industry, demand for data-savvy employees is far outstripping the available supply, according to a new report from the McKinsey Global Institute, the research arm of global consulting firm McKinsey.

The implications for K-12 schools?

Teaching more statistical reasoning and quantitative skills to younger students will help them be better prepared for the tech- and data-driven jobs of the future, said Michael Chui, a partner at MGI, in an interview.

“The ability to understand what probability means is now a basic life skill,” Chui said. “It has tremendous implications for us as workers and as citizens.”

The McKinsey Global Institute report, titled “The Age of Analytics: Competing in a Data-Driven World,” makes the following case about the nation’s data-science talent pipeline: In response to employer demand, higher-education institutions have begun offering more degree programs in data science and analytics (the researchers counted 100 such master’s-level programs in business analytics alone.) And students have “flocked” to such opportunities, with the number of degrees granted in such fields growing by 7.5 percent between 2010 and 2015.

Still, demand for data scientists is growing much faster, by perhaps as much as 12 percent per year.

One indicator of the resulting gap in the labor market: average wages for data scientists rose by about 16 percent between 2012 and 2014, compared to a 2 percent increase in wages overall. In a survey conducted by McKinsey earlier this year, business leaders also said finding and retaining analytics talent was far more difficult than in other areas.

The net effect, the McKinsey Global Institute concludes, is that the U.S. economy could be short as many as 250,000 data scientists by 2024.

“Providing people with the skills to get jobs is one goal of education,” Chui said.

“At least in the medium term, if students gain data-science credentials and have the skills, they will be in demand.”

‘All the Math You Can Possibly Take’

The legitimacy of the so-called “skills gap” in STEM subjects (science, technology, engineering and math) has been much debated in recent years. But when it comes to the specific field of data science, there is little doubt that “the recent ‘big data’ trend has sparked demand for data scientists in all areas, from health care to retail,” the federal Bureau of Labor Statistics wrote in 2015.

Tech companies are certainly part of that equation. Just last week, for example, social-networking behemoth Facebook launched an initiative to help educate students, teachers, and the public about “artificial intelligence,” a kind of blend of computer science , data science,, and neuroscience in which algorithms are used to design “intelligent” machines that can recognize patterns and images, translate languages, and learn some kinds of logical reasoning. (A series of videos featuring Vann LeCun, the company’s head of AI research, further explains the field.)

In a blog post, Facebook listed a number of tips for students to prepare for such fields. Chief among them: “Take all the math you can possibly take,” including probability and statistics. (And while you’re at it, the company recommends, make sure you take some computer science, and try to squeeze in engineering, economics, philosophy of knowledge, and the latest brain research, too.)

It’s not just tech companies that are seeking such skills. The fields of transportation (think: self-driving cars), retail, advertising, and health care are among those already undergoing radical transformations. Manufacturing and the public sector (including public education) have also started dipping in their toes.

Even those jobs that aren’t replaced or rendered obsolete will be dramatically changed, McKinsey argues. The reason: as data-driven automation yields new advances in machines’ ability to process natural language, recognize patterns, and even sense human emotion, everyone from administrative assistants to lawyers to industrial engineers will see core aspects of their daily work evolve or disappear.

The Importance of Statistics

As a result, the McKinsey Global Institute researchers argued, for every data scientist that employers hire, they will need even more “translators,” able to connect new information to real-world business problems.

“These are the people who know enough about data science to be good consumers of data, but who also have enough domain knowledge to bring the resulting insights into organizations and effect change at scale,” Chui said.

While Facebook makes the case for more math of all kinds, Chui is particularly keen on making sure students are taught the building blocks of data science at an ever-younger age (one instructional strategy: games involving dice for elementary students).

Three years ago, he outlined his argument in a piece published by Business Insider.

US students are shuttled along a familiar path in mathematics: first stop algebra, then geometry and trigonometry, and finally, the ultimate destination, calculus. This time-honored curriculum seems increasingly out of touch in a world that is flooded with noisy and voluminous data. The majority of students need to be immersed in the more practical discipline of statistics, which has greater relevance for the jobs being generated by a digital economy. The key to making sense of all the data now at our disposal is statistics. At leading companies, decisions once driven by HiPPOs (the Highest-Paid Person’s Opinion) are increasingly made by conducting experiments that draw on the core skills of statisticians. Rather than relying on gut instinct, businesses now find ways to test hypotheses and use statistical methods to analyze the results, applying the classic scientific method to decision-making.

A quick look at the total number of students taking related Advanced Placement exams highlights the ways in which K-12 schools have been relatively slow to respond to such changes:

  • More than twice as many students still take the AP Calculus exam than the AP Statistics exam each year. In 2016, for example, 433,146 students took the former, compared to 206,563 who took the latter.
  • The number of students taking AP Statistics has increased by more than 34 percent over the past five years. While that’s faster than the growth in the number of students taking AP Calculus (20 percent), it pales in comparison to the 122 percent increase in the number of students taking the AP Computer Science exam over the same period.

Chui would like to see those numbers change. Human beings are naturally pretty bad at understanding things like probability, he said. But in a rapidly changing labor market, in which machines will perform not only rote labor and calculations, but many more advanced tasks, humans are going to need such higher-order conceptual skills more than ever.

“The two questions I ask people are, ‘When was the last time you had to take an integral, and when was the last time you had to make a decision based on a large, incomplete, and inconsistent set of data?’” he said.

“Usually, the answers are ‘a long time ago’ and ‘since breakfast.’”

See also:

A version of this news article first appeared in the Digital Education blog.