Ive been involved in analytics since 1977, and, until recently, had never heard of this creature we call a "data scientist." Yet, the hybridized skillsets associated with the data scientist job title aren't new -- it's just that they've traditionally linked to a specific profession or industry.
For example, we've long had natural, social, and behavioral scientists who possess skills in statistics, survey design, and research methods. We've long had epidemiologists, biostatisticians, and demographers. We've long worked with data abstractors, medical and surgical claim coders, and fraud and abuse detection programmers. And computer whiz kids, quants, geeks, dorks, dweebs, and nerds have long lived among us.
So, quantitative skills, data-processing skills, visualization, and reporting skills are nothing new. What is new is that cheaper access to huge stores of structured and unstructured data (affectionately known as big-data) is creating a demand for data generalists in the labor market. The advent of big-data is creating a need for people who can engage a big glob of data, methodically and systematically explore and distill the contents, and find something that will give an organization a competitive advantage -- and justify the investment made in creating a big-data shop.
You're probably ready to jump on the data scientist bandwagon! But before you make the leap, first consider this question: What's the likelihood that the labor market demand for this job will be sustainable for the next 10 to 20 years?
It's a really important question for today, what with so many analytic professionals making important decisions concerning post-baccalaureate professional development.
The lack of a reliable crystal ball notwithstanding, lets consider two basic labor market elements -- supply and demand -- and how they relate to the outlook for data scientists.
First, the long-term supply of data scientists will be based on the prevailing hours and wage levels. In general, a career choice is based on the assumption that the average weekly hours of employment, and the average hourly earnings, are enough to live on comfortably -- at the least. Someone performing data scientist tasks, as a part of a broader job responsibility, should be fine at the moment, especially employable as a consultant for a competitive salary.
However, it's too soon to tell if the hours and earnings are promising enough to encourage more people to follow this path. It's too early to establish wage projections for full-time data scientists.
Second, to the extent that big-data analytics creates widespread revenue growth and profitability for firms and industries, demand for data scientists will continue. Even though training and software vendors are promoting the need for data scientists, and some companies are early entrants in the market, any long-term hopes have to be grounded in what happens in private industry.
For example, as much as we criticize lawyers, we need them because people living in community will always have contractual and non-contractual disputes. In like manner, constant big-data analyses will have to become an indispensable component of doing business for the extension of the data scientist profession.
In short, the data scientist buzz is great for analytics; many are excited over this new area of analytic skill deployment. If you can expand your current duties into a data scientist role, then that's ideal. However, before committing to making this your long-term career path commitment, I encourage the curious to pay close attention to the foundational dynamics of labor markets.
The laws of supply and demand preceded big-data, and will be here long after the next big thing replaces it.