Data Science Skills Gap May Be Narrowing
Big data “hype” has attracted more new data scientists, thereby driving down salaries for new entrants for the first time in four years, according to a salary survey.
The annual survey released by the executive recruiter Burtch Works also found that the rush to take advantage of comparatively high starting salaries has reduced the percentage of data scientists with PhDs to 25 percent. “Some data scientists are also opting to skip the PhD as a faster route to the workplace, to capitalize on the numerous opportunities available,” the survey found.
The report focused on data scientists working with unstructured and streaming data, separating them from traditional predictive analytics such as business intelligence. Data scientists working with unstructured or streaming data are generally considered to possess more computer science skills. Entry-level data scientists, for example, were found to make 27 percent more than a predictive analytics professional. Annual salaries for new data science recruits are approaching $100,000, the Burtch survey found.
As salaries for data scientists have increased, so to have enrollment and degrees awarded in computer science along with mathematics and statistics. “The increased interest in all of these fields of study is undoubtedly having an effect on the market,” the report concludes.
However, the steady blending of data skills derived form online courses, boot camps and in-house training has helped traditional practitioners transition to data science roles. The upshot has been “a leveling effect” on data science salaries.
“For the first time in four years, salaries amongst individual contributor data scientists with [up to three] years’ experience decreased slightly,” the study found. Previously, “these professionals were seeing the largest percentage gains in median base salaries.”
Meanwhile, the dynamics of data-driven business also is beginning to shift from heavy investment in analytics and data science to return on that investment. The survey amplifies growing concerns that enterprises are becoming “data rich and knowledge poor.”
“As teams grow larger and investments in data science salaries increase, companies will want proof that their investment is contributing positively to the bottom line,” Burtch reported.
Those concerns have contributed to the slow down in data scientist compensation. The researcher found that while median base salaries have remained steady, entry-level salaries at the bottom of its data science pay scale fell for the first time by 2 percent to about $95,000. Previously, this category was attracting the highest annual percentage salary increase of the job categories tracked, ranging from “individual contributor” to senior data science managers.
The surge of entry-level data scientists is expected to usher in an era of specialization. “It is becoming more common to see many professionals on a large team with differing backgrounds—some in computer science, some in math and statistics, and others in engineering—all working together on the same data science project,” the report concludes.
“With many quantitative teams continuing to grow and the data science ‘unicorns’ being so difficult to hire, we expect the trend of specialization to continue.”