Data Scientist – What’s in a Name?

The rate at which many businesses are amassing huge quantities of so-called “big data” has created an urgent need for experts capable of managing, extracting, analysing and interpreting these enormous datasets. This, in turn, has led to the dramatic growth of new roles such as “data scientist” and “informatician” that were all but unheard of just a few years ago.  However, what distinguishes the role of a data scientist from that of a statistician, informatician or analyst?  Although there have been a number of interesting blogs and discussion posts on the subject, it is difficult to find any clear agreement.

Data scientists, statisticians, informaticians and analysts alike occupy the world of information or decision science and have complementary skill sets that together unlock the value in data.  They all work in an environment that is interdisciplinary in nature and they need to blend a combination of technical knowledge and programming skills with contextual understanding, teamwork and communication.

So, what’s in a name?  There is clearly a great deal of overlap, but the key to differentiating these roles is in looking at the balance of the different skill sets each possesses.  Whilst together they need to draw from all of the specialisms listed above, each has their own forte.  A statistician is likely to have more technical knowledge in statistics, often having formal academic training to PhD level, and be capable of selecting or developing the most appropriate statistical methods for drawing inferences from data.  A data scientist is likely to have more knowledge of advanced programming languages and be better equipped to deal with large data, potentially across multiple databases.  An analyst is likely to have more specialised knowledge of the relevant commercial sector and have very focussed specialist expertise in that area. Whereas an informatician is likely to be more adept at designing data interfaces to obtain and retrieve data and have greater expertise in producing data visualisations via dashboards for the end-user, for example.

Rather than focus on titles, it is more important to think about what the different groups actually do – they each use a blend of talents to help businesses to make decisions that ultimately optimise their products, services and processes.  A statistician, data scientist, informatician or analyst could likely independently carry out a project that involved retrieving data from various sources, modelling and analysing the data to make inferences and predictions, and then explaining the results with appropriate visualisations.  However, the ideal project would be delivered by a team with each member applying their own specialist expertise to find the best solution.

Related Articles

Big data

Data science