Mathematics was always my worst subject in school, but the more I learn about the working world, the more I appreciate it. I would advise any young person to study as much math as you can stand, and maybe even a little more. If you can’t stomach calculus, then take statistics. A lot of the really exciting work being done now involves new applications of math. And an ever-increasing number of careers are driven by data, so an understanding of data analysis keeps growing in value.
I remember that in the 1980s Brazil was thirsty for petroleum, so to avoid a gross trade imbalance that nation relied heavily on sugarcane-derived alcohol to power its cars. In 1980 Brazil produced only 182,000 barrels of petroleum. In 2011, however, production from offshore wells had boosted that figure to 2,105,000 barrels. Recently I learned that Brazil has been aware of the existence of offshore oil deposits for decades, but the petroleum is covered by a mile-thick layer of salt, and there used to be no way to determine the right places to drill. What made drilling feasible was the development of new algorithms that could make sense of seismic, magnetic, and gravitational data. In other words, math solved this problem.
Math also plays a role in the development of green energy resources. For example, wind turbines keep getting bigger in order to capture more wind energy, but blades as big as airplane wings create new problems, such as great differences between the wind conditions at the top and bottom of the blades’ sweep. General Electric has found ways to adjust the rotors to these conditions by using sophisticated algorithms that respond to input from various sensors.
“Big data” is another field with a growing need for people with math skills. IBM estimates that every day our economy generates 2.5 quintillion bytes of data, and that 90 percent of the world’s data has been generated in just the past two years. We refer to data analysis casually as “number crunching,” but what does this involve? The most obvious task is summary of large quantities of data. But even this task is multidimensional; you need to find ways to tease out not just the central tendencies, but also measures of variability at any one time and over time. And you have to find ways to ensure the accuracy of the data you’re getting, perhaps by drawing on redundant or divergent sources and finding ways to deal with the contradictions you uncover.
A few weeks ago I blogged about careers in analyzing medical data. The McKinsey Global Institute estimates the potential value of health-care data at more than $300 billion per year, two-thirds of which would result from reducing the nation’s health-care expenditures by about 8 percent. And that is just one field with potential for creating jobs for is big-data analysis. However, McKinsey projects a shortage of 140,000 to 190,000 workers to fill the big-data-crunching jobs that the U.S. economy could generate by 2018.
Over the course of my career, I have learned how to use Excel to do the computational heavy lifting, so it no longer matters that I had a hard time mastering the times tables in elementary school. The math skills I’m learning now are not so much computational as conceptual, such as how and why to use weighted samples. To a young person, I would suggest trying to see beyond computation and focusing instead on the underlying principles of data manipulation and analysis. Those skills can lead to a lifetime career of creative and productive work.