10 Skills of Effective Data Scientists

Icon by George Montana Harkin via the Noun Project.

Icon by George Montana Harkin via the Noun Project.

The field of data science is a relatively new one, with a growing handful of academic programs making early attempts at setting a rubric. As a result, its definition and the roles it encompasses remains up for debate. There are plenty of broad definitions, but the suite of skills at the disposal of successful practitioners is more oblique. Sure, data science involves statistical modeling and computer engineering, but what precisely does that entail? A recent post at Data Science Central by Mitchell A. Sanders presents an in-depth survey of the skills shared by effective data scientists in the industry.

Sanders segments the activities of a data scientist into three discrete categories:

  • Capture
    • Programming and Database Skills
    • Business Domain Expertise and Knowledge
    • Data Modeling, Warehouse, and Unstructured Data Skills
  • Analyze
    • Statistical Tool Skills
    • Math skills
  • Present
    • Visualization Tool Skills
    • Storytelling Skills

Sanders goes further in his breakdown, acknowledging the diversity of opinions in the field by citing other takes on the topic. Some of the other lists recognize personality traits that are just as important as technical skill and expertise, such as curiosity, innovative thinking, and the indefinable yet useful quality of intuition.

At Data Community DC, NASA Research Scientist Oscar Olmedo diagrams the process of data science in a pyramid which shares much with Sanders’ outline, though it differs in the segmentation of tasks:

Olmedo echoes others in the field with his emphasis on data scientists who can pose the right question. This is an ability that draws upon many of the skills Sanders cites in his post: Domain Expertise and Knowledge, Data Modeling Skills, as well as innovative thinking and intuition. Olmedo writes:

Probably the most difficult part of beginning a data science project is knowing what questions to ask of data. For example, let us say we had a set of data with X number of parameters. We are not just blindly going to start analyzing the data, we must first ask a question of data. Furthermore, we may not even necessary have the data or any data at hand at the time of asking. But once we know the question we can move to the bottom most tier of the pyramid.

Though there are plenty of perspectives about the skill sets of effective data scientists, many of them have much in common and differ largely on which skills are prioritized and what ancillary traits are desirable.

What do you think are the most important skills and traits for data scientists?


Tags: , , ,

3 comments on “10 Skills of Effective Data Scientists

  1. Randy Bartlett on said:

    Could use more meat on data analysis. IT and Business Analytics are different animals.

  2. Keep in mind that business in ONE application of data science. At its core, a business acumen is not a requirement for a Data Scientist; however, it MAY be of value when attempting to align Analytics with business processes. But it should be noted that thinking in terms of the business is not always the best way forward and can even be detrimental. Many times the “numbers” tell a story that is counterintuitive to business intuition yet reveals something the “business-type” individuals never could have seen. This is the power of Data Science and we need to remember that it is a true science and affects virtually every field of study and industry. Its power comes from existing in a truly scientific state, independent of the business it is attempting to assist. Only then can it offer truly objective results on the state of the data at hand.

  3. Very useful outline of data scientist skills. Curiosity and creativity are core to data scientists based on our 2012 research. Companies should never hire exclusively for skills, that’s the number one mistake analytics hiring managers make.

Leave a Reply

Your email address will not be published. Required fields are marked *


You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>