What kind of data science are you doing?Finding a place to beginOne of most common questions for budding data scientists concerns how to begin.
While university programs and bootcamps for data science have begun to assemble curricula for would-be data scientists, there are still a number of questions that need to be addressed.
Learning a few programming languages and the basics of machine learning is a good start, but doesn’t provide much direction for additional knowledge that would help newcomers secure positions.
Defining a data scientist by the type of data they work with can help individuals to identify professional targets, narrow the set of skills to focus on developing³, and from the other side, encourage standardization of the language appearing in job descriptions.
When applying for data science jobs, it would be helpful to be able to speak to the type of data you are applying to work with.
Cross-domain transference of skillSimilarly, having descriptions of the kind of data a data scientist works with as part of their title would give greater meaning to an otherwise generic Data Scientist label.
With this information, businesses that share data types but live in different industries would be able to more quickly assess the fit of individuals looking for positions.
Similarly, it would help give data scientists awareness of areas outside of their own experience in which their skills could be effectively applied.
Data scientists are valued if anything by the impact they make, and the greater their knowledge or intuition of a domain, the greater this can be.
Neither new nor completeThe proposed way of thinking about data science is not original.
Previous writers have discussed the comparison of data science to science⁴, while others have suggested defining what “data science is by its usage — e.
g.
, what data scientists get paid to do,”⁵ or by the output they create⁶.
These all very much resonate with the idea of borrowing from the established taxonomy of science and finding lines of similarity along the axes of observation.
And along these axes, additional delineating characteristics are still going to be needed, which could be according to the type of work performed⁷ or duties required.
As the field of data science and the areas in which data scientists find themselves occupied evolve, the language we use to describe what it is that data scientists do will also need to evolve.
The suggestion presented here is to begin including terms that reflect the type of data, observations, or purpose, that are most relevant to the type of work being done.
[1] http://drewconway.
com/zia/2013/3/26/the-data-science-venn-diagram[2] https://www.
kdnuggets.
com/2016/10/battle-data-science-venn-diagrams.
html/2[3] https://towardsdatascience.
com/why-you-shouldnt-be-a-data-science-generalist-f69ea37cdd2c[3] https://towardsdatascience.
com/is-data-science-a-real-science-2920bb2529aa[5] Schutt, Rachel and O’Neil, Cathy (2014).
Doing Data Science.
O’Reilly Media Inc.
, Sebastopol, CA[6] Garten, Yael (2018) The Kinds of Data Science.
Harvard Business Review https://hbr.
org/2018/11/the-kinds-of-data-scientist[7] https://www.
linkedin.
com/pulse/one-data-science-job-doesnt-fit-all-elena-grewal/.