The proliferation of ways to measure things – point of service terminals, web analytics, geographic and temporal records, even semantic information – means businesses are drowning in data. This has led to a new class of engineer, the “data scientist,” whose job it is to perform the sophisticated mathematical gymnastics required to extract actionable information from this mass of numbers. According to mathematician Cathy O’Neill, the skills of a data scientist include not only crunching numbers, but also visualizing the results.

As Carnegie Mellon statistician Cosma Shalizi points out, O’Neill’s description of the skills required of a data scientist are precisely those of a suitably well-educated statistician, even if he or she has only an undergraduate degree in the subject. Granted, Shaliz teaches at Carnegie Mellon, which is among the best engineering schools on the planet, so that’s not to say that everyone with a B.S. in statistics has mastered modern regression, advanced data analysis, data mining and statistical visualization.

This re-branding of statistical literacy as “data science” points out a larger trend – disciplines that were formerly the domain of the specialist, such as statistics, are now more important to a larger segment of the business world than ever. The fact that so few students view even a fraction of this level of mastery as necessary – and that schools often do not offer even a basic statistical education to non-math majors until the post-graduate level – suggests that in this area, perhaps even more than other areas associated with engineering, there is a yawning gap between the skills our workforce possesses and the skills employers require.

Physicists have long been drifting into Wall Street, which can use their mathematical abilities to manage hedge funds and the like. Will we see a similar drift of mathematicians into startups where business decision-making was formerly the sole domain of sales and C-level executives?