Subscribe to DSC Newsletter

What type are you? Six Job Categories for Data Scientists

Once dubbed as the sexiest job of the 21st century by The Harvard Business Review, data scientists take pride in having adept technical skills in providing solutions to problems through data visualization, pattern recognition, text analytics, and data preparation among many other skills.

Given the various industries that utilize data and draw valuable insights from it to enhance their businesses and services, data scientists play a huge role in the progress of any business.

In fact, data scientists can be found in at least eight fields: statistics, mathematics, data engineering, machine learning, business, software engineering, visualization, and spatial data.

If you’re a student of data science and analytics, you won’t have to worry about employment opportunities after you graduate. Listed below are six types of jobs for data scientists:

 

1.    Data analyst

Perhaps the most common job, data analysts are similar to data scientists in such a way that they both use data for solving problems. While data analyst may be unable to develop new systems or algorithms, they focus their job on collecting, processing, and performing statistical data analyses through statistical tools and techniques.

Once they have relevant information, they begin data interpretation by spotting trends, correlations, and patterns in complex data sets to provide for improvements.

 

2. Business Analyst

If data scientists concern themselves in improving business processes such output, distribution, and productivity, then they can be fit for business analyst positions.

They are responsible for devising data-driven solutions to organizational problems, and inventing new systems to improve efficiency in personnel, product offerings, and service processes.

 

3. Marketing Analyst

Data scientists who find themselves interested in market conditions, consumer behavior, and product studies may appreciate landing jobs as market research analysts. They often guide companies in identifying which products or services will sell, to which specific market type, and at what price.

Market research analysts are responsible for gathering relevant data on consumer demographics, their buying behavior, and competitors’ profiles. They are also in charge of assessing the company’s marketing strategies to determine whether it works for them or not. Often, their job outputs impact on how a company designs, sells, and distribute their products and services.

 

4. Data Architect

Data scientists can be architects that develop and maintain data management systems.

They work on creating plans for integration, centralization, protection, and maintenance of a company’s internal and external data sources. Data architects’ jobs are significant by providing easy access to pertinent and significant information for everyone in the company.

Their responsibilities cover database structures, inventories, data acquisition, security, and recovery.

5. Quantitative Analyst

With the objective of reducing risk and increasing profit generation for a business, quantitative analysts make use of financial data to guide management decisions in investments, pricing, and risk management.

“Quants,” as they are also called, examines market trends and dynamics, trading strategies and system performance, and investment management to enable them in testing new models, products, and programs to ensure their company won’t be at risk of financial loss.

 

6. Statistician

When data scientists get themselves involved in the application of statistical theories and methods for data gathering, analysis, and interpretation, they likely fall under the job category of statisticians.

Oftentimes, statisticians tackle market research for quality control and product development for businesses, while they can also work in the academe for heavier statistical applications.

Statisticians are responsible for data extraction through existing models or new ones, whichever is required by the situation. They also employ statistical tools and algorithms such as SAS and SPSS when analyzing and interpreting data, while simultaneously recognizing trends, relationships, and patterns among their data sets.

Now that you’ve read through them, which of these types do you see yourself taking?

Views: 6706

Tags: Job, data, scientists, type

Comment

You need to be a member of Data Science Central to add comments!

Join Data Science Central

Comment by Mike Morgan on November 30, 2017 at 3:46pm

I like this article a lot, because it is reaching out to open people's eyes about the value of the full range of skill domains in the field of data science.  I myself have posted a less informative but similarly targeted essay on my LinkedIn site.

All I would suggest is that network architecture - the provision, configuration, network protocols, network platforms like Hadoop, software running on those platforms like Spark, seem to have become an obsession among people in the IT-managed flavors of data science.  Other than funding hardware and server rental, these things are pretty easy to learn if you know about things like IP addressing, etc. but setting up the platforms to begin with, in an unfamiliar IT environment, (please excuse the language), can be a total bitch.  Which is why this capability surfaces in so many data science job descriptions.  No one likes to go through all that sorrow more than once!

Once data science moves onto a multi-processor-server network platform, in principle, one can program how the data are scraped from the web, how data tables can be imported, and how weblogs and page tags are managed in a way that is more friendly to analytical programs.  (Expensive) commercial packages do and have existed for these functions, but to what extent, and how flexibly, I hesitate to say, and I wouldn't want to offend anyone.  The freeware like Java core, Hadoop, Apache Spark (+ Python, R, SQL) are more easily available and, very recently even more easily install-able if you already have Hadoop running on Linux/Unix.

The more advanced, highly iterative data science algorithms probably still cannot run on any single network slave node (server), unless the mathematics are hardwired into a narrowly-specified input device (e.g. digital facial recognition).  So, the vast legion of new data science algorithms, such as NLP and deep learning, must still be run on the master server after pulling (or while continuously pulling) data out of the cluster node servers.  Especially text database builds with huge volumes of mostly useless words and phrases.  My opinion, not strongly supported by evidence, is that companies are more and more often automating the big text database builds and running automated text mining and reporting when various measures can be output.  This is too bad, since it will remove humans from the process and thus stunt the growth of text mining science for those companies to become more competitive.  I hope it's not the case, but it seems likely.  Bigger and bigger haystacks within which to find the same needle.

But the really strong quantitative analytics that data science is most recently known for will, I believe, get out there onto the cluster nodes, so the master server doesn't have to spend so much time processing cluster-collected data (One can also push big data out onto the cluster nodes to run multiprocessing, but I've never seen that work as fast as one might hope, having expended all that time and money to export data out to the cluster.)

There is hope, however, if only because our wonderful military continues to pour funds into robot soldiers that can be organized and controlled by the master server but still have data/image analysis/recognition and responsiveness at the individual slave node level - perhaps, where possible, hardwired with very specific functionalities. 

Again, a very worthwhile opening of conversations around the various facets, capabilities and hopes for data science to serve business (and maybe humanity).  Thanks so much, Laura, for offering your post to help get people aware of what data science actually consists of!

Videos

  • Add Videos
  • View All

© 2019   Data Science Central ®   Powered by

Badges  |  Report an Issue  |  Privacy Policy  |  Terms of Service