Enterprise IT Watch Blog

Aug 31 2015   9:30AM GMT

Top-notch data scientists are grown, not born

Michael Tidmarsh Michael Tidmarsh Profile: Michael Tidmarsh

Big Data
Data Science
Data scientist

Big data image via Shutterstock

By James Kobielus (@jameskobielus)

You can anoint yourself with whatever lofty job title you wish. That doesn’t mean that others need to buy into your conceit. Or, more to the point, it doesn’t mean others will feel a compulsion to hire you into that role or pay you at a rate commensurate with your inflated self-regard.

Data scientist is a job title that many people are adopting, if for no other reason than that it’s supposedly the “sexiest” job in the 21st century. Given the fact that many people see it as the ticket to a rewarding career, no one can deny that some of these nouveau data scientists are well-intentioned incompetents, while others may be outright charlatans trying to cash in on the latest trend. There are confidence artists in every profession.

Considering how important data science is to the big-data analytics revolution, we can’t afford to let underperforming data scientists take root in our development shops. But, given the shortage of data science skills deficit, we also can’t afford to turn away promising individuals who, though they may not be “naturals,” might become exceptional data scientists if given an opportunity to grow their skills in the right environment.

How do you know a good data scientist, or as least someone with real potential, when you see one? When hiring and promoting, how can you distinguish a high-quality data scientist from someone who talks a good talk but just can’t produce? Or, if you’re grooming fresh data-science talent in your organization, how can you tell who’s got true potential from those who may try hard but just don’t have the skills or aptitude to succeed?

There is no shortage of commentary on what makes a good data scientist. If you’re aspiring to this career or grooming others for it, it’s best to group the advice into two broad categories: “here’s why someone will never become a top-notch data scientist” vs. “here are the milestones that someone must hurdle if they want to become one.”

Some commentators lean far more in one direction than the other with their advice. A few years ago, for example, Vincent Granville offered his 13-point checklist and linked to an online diagnostic test for distinguishing so-called “fake data scientists” from those who ostensibly have the right stuff. Though I disagree with this polarizing approach to talent management, one can’t deny that some people may not have the innate aptitudes or work ethic needed to excel in this field.

Some commentators recognize that “data scientist” is an aspiration as much as a vocation. This past month, Bernard Marr presented his 5-point list of signs that someone isn’t quite ready to become a peak-performing data scientist, but may get there some day through focus, education, and perseverance. However, Marr’s list isn’t very different from Granville’s in its reliance on stigmatizing characterizations—e.g., “you aren’t creative”—that feel more like insults than constructive advice. And his call to readers to “add to my list of signs that someone is NOT a data scientist” feels like he’s building a detailed profile to be used for excluding people from this field.

My feeling is that good data scientists are grown—through education, experience, and incentives—rather than born. For example, creativity is in the eye of the beholder. The best data scientists can demonstrate that when they’re given a challenge that calls it forth. In this blog from three years ago, I discussed the attributes that you should look for when evaluating a data scientist’s performance on real-world projects. These traits include curiosity, intellectual agility, statistical fluency, research stamina, scientific rigor, and a skeptical nature. Some of these may be innate, but some may be entirely the result of a determined individual’s efforts to grow and stretch in their chosen career.

Education is the very heart of professional growth, and most data scientists seek it out wherever it’s available. In a separate blog three years ago, I discussed the core curriculum that every good data scientist needs to master in order to excel in the job. Many of the new generation of data scientists are largely self-taught, though it’s usually best to have some formal education under your belt.

My perspective on this issue is intensely personal. Ten years ago, I didn’t know much about data science, data management, big data, or analytics. However, I was offered a high-visibility job that required that I master all of these topics, without the benefit of classroom instruction or a degree, practically overnight.

Many of you read my work. I’m in my mid-50s, and now I’m someone that many regard as an expert in this field. I’m not a data scientist, per se, but I obviously know the subject matter. I wasn’t born to do this work. I’ve grown myself into it.

And I’m not faking it. I author every last word that goes out under my byline.

 Comment on this Post

There was an error processing your information. Please try again later.
Thanks. We'll let you know when a new response is added.
Send me notifications when other members comment.

Forgot Password

No problem! Submit your e-mail address below. We'll send you an e-mail containing your password.

Your password has been sent to:

Share this item with your network: