How to Become a Data Scientist

“Organizations are increasingly realizing that in order to maximize their benefit from data, they require dedicated leadership with the relevant skills.” ~ DJ Patil, U.S. Chief Data Scientist, White House Office of Science and Technology Policy

With data science proliferating every market segment and with new ways of exploiting massive data assets being discovered daily, the public and private sector alike are competing for top talent. According to executive recruiting firm Burtch Works, the hiring market for analytics professionals and data scientists has “gone into overdrive.”

But as you talk to hiring managers and survey job descriptions in all the clearinghouses for job vacancies, a funny thing can be observed: Even amid an unprecedented level of demand and a huge shortfall of qualified applicants, the expectations for experience, breadth of knowledge and advanced education aren’t being adjusted downward to attract a larger candidate pool- employer expectations are increasing.

Generally speaking, employers look for job candidates to fill data scientist positions that have:

  • An advanced degree – According to executive recruiting firm Burtch Works, about 88 percent of data scientists hold at least a master’s degree. The most common fields of study include mathematics, statistics, computer science, engineering and, more recently, data science as field-specific programs have become widely available.
  • Experience with programming languages like Python, Perl, C/C++, and Java.
  • Proficiency in big data software platforms like Hadoop, Hive, and Pig.
  • In-depth knowledge of at least a few analytical tools like SAS and R.
  • Familiarity with cloud tools like Amazon S3.
  • SQL Database/coding capabilities

Steps to Become a Data Scientist

So, how does someone with a background in computer science, math, statistics or engineering become an expert in managing, modeling, and visualizing big data?

Make Sure You Have a Qualifying Undergraduate Degree
Supplement Your Undergraduate Education and Prepare for Graduate School
Be Prepared for the Graduate Admissions Process: Passing Exams and Gaining Experience
Earn a Master’s Degree in Data Science
Begin a Career in Data Science
Keep Learning and Specialize – Certification Options

 


 

Step 1. Make Sure You Have a Qualifying Undergraduate Degree

Data science master’s degrees are generally available through selective graduate programs reserved for the most qualified students. Making sure you have the qualifications grad programs expect starts at the undergraduate level.

In general, data science graduate programs prefer candidates with an undergraduate major in:

  • Applied mathematics
  • Computers science
  • Applied statistics
  • Engineering
  • Physics

Some programs accept students with other undergraduate majors, provided they have completed undergraduate courses in:

  • Computer programming
  • Statistics
  • Calculus I & II
  • Linear algebra

But the right major alone isn’t always enough. Most programs prefer candidates with a 3.0 GPA and above.

 


 

Step 2. Supplement Your Undergraduate Education and Prepare for Graduate School

Often times even the most qualified candidates must complete a number of undergraduate courses before beginning graduate studies. Fortunately, there are a host of avenues to get you there:

Bridge Courses – Some graduate schools provide well-qualified students that have met all standard admissions requirements but that lack proficiencies in certain areas with the opportunity to complete bridge courses. Bridge courses give you a chance to fulfill undergraduate prerequisites and gain the skills required to begin graduate-level coursework:

  • Mathematics Bridge – courses including algorithm analysis, linear algebra and data structures
  • Computer Science Bridge – courses related to database management and administration
  • Programming Bridge – courses in programming languages like Java, C++, R and Python

Not all colleges and universities with data science master’s degree programs provide bridge programs, and those that do often require students to earn a minimum grade point average in bridge courses before entering the graduate program.

Massive Open Online Courses (MOOCs) – Massive open online courses (MOOCs) provide you with another option for honing the skills you need before studying data science at the graduate level, all through open access online courses.

The advantages of a MOOC includes:

  • Open access – Allows access to top-level professors at respected schools across the country.
  • A more diverse student body – Open courses allow all interested students to attend the course, regardless of location.
  • Data collection measures success – Programs capture the success and/or failure of students, providing professors with precise information used to improve courses.
  • Flexible – MOOCs aren’t fixed into traditional term and semester models, allowing a greater level of flexibility when it comes to scheduling.

There are a large number of MOOC providers, although the three leaders include:

Boot Camps – Data science boot camps, building off the success of coding boot camps, are popular offerings for data scientists at nearly every career stage. If you want to brush up on the fundamentals or solve real world data science problems through a beginner course of study before entering a grad program in data science, boot camps may be a good fit to get your skills up to speed. A data science boot camp may also be the ideal way to decide if a grad degree in data science is right for you.

Boot camps may run anywhere from 5 days to 4 weeks in duration, and requirements vary according to the technologies being taught. Even beginner boot camps require some knowledge of programming and/or scripting language. Expect to find data science boot camps offered in larger cities.

Some boot camps are designed as intensive fellowships for data science graduates with advanced degrees who want to find work in the data science field. These programs generally offer free tuition, along with the opportunity to work alongside a professional mentor.

Boot camps often conclude with a job fair and placement program were graduates are introduced to companies scouting data scientists. This gives graduates a shot at some highly sought after jobs with companies that run the gamut from burgeoning startups to major players in the field of data science.

 


 

Step 3. Be Prepared for the Graduate Admissions Process: Passing Exams and Gaining Experience

In addition to a bachelor’s degree in a related field, data science grad program candidates are also expected to provide the institution with letters of recommendation and an admissions essay. Admission also relies heavily on:

  • Minimum scores on the quantitative section of either the GRE or GMAT (sometimes waived with an undergraduate GPA of 3.0 or higher)
  • Related practical experience

Prepare to Take the GRE or GMAT

The admission process for a data science master’s degree program often includes submitting GRE or GMAT scores. In general, institutions look for candidates who have scored in the 85th percentile in the quantitative section of the exam.

You can increase your chances of success on these exams by taking advantage of available study resources and tips:

  • GMAT – The Graduate Admission Management Council (GMAC) provides GMAT test takers with preparation tips, study guides, and prep tools to prepare them for success on the quantitative section of the exam.
  • GRE – The Educational Testing Service (ETS) provides GRE test takers with a number of resources, including the POWERPREP II software, to prepare them to take the quantitative reasoning section of the exam.

Gain Valuable Experience

Institutions often look to a candidate’s practical experience to determine eligibility and preparedness for master’s-level courses. Those likely to receive consideration for admission often possess experience in areas like:

  • Data analysis
  • Business intelligence
  • Data mining
  • Pattern recognition
  • Machine learning
  • Software engineering
  • Business analytics

You can (and should) begin building your resume while satisfying the requirements for admissions by gaining experience in entry-level data scientist jobs that provide plenty of opportunities to increase your proficiency in areas like coding, programming, and big data software.

Business acumen is nearly as important as programming and statistical experience for a data scientist, so a strong background in business and strategy showcasing your leadership and communication skills and analytical mindset should be part of your resume.

 


 

Step 4. Earn a Master’s Degree in Data Science

Depending on the institution, a master’s degree in data science may go by a number of titles:

  • Master of Information and Data Science (MIDS)
  • Master of Science in Data Science
  • Master of Science in Statistics: Data Science
  • Master of Computational Data Science

These programs encompass about 30 credits of coursework and between 18 and 24 months of full-time study. Coursework includes:

  • Applied Machine Learning
  • Data and Network Security
  • Data Mining
  • Experimental Statistics
  • Exploring and Analyzing Data
  • File Organization and Data Management
  • Research Design and Application for Data and Analyses
  • Statistical Sampling

A master’s degree in data science, through its multidisciplinary curriculum, prepares students to:

  • Retrieve, organize, clean, and store data
  • Imagine innovative uses for large datasets
  • Apply statistical analysis and machine learning techniques
  • Understand ethical and legal considerations associated with data privacy and security
  • Identify patterns and make predictions

Program Delivery Options

In addition to full-time, campus-based data science programs, a number of institutions offer students alternate delivery options:

  • Online Programs: Designed for working professionals, today’s online data science master degree programs allow you to complete your coursework through a distance-based delivery method, offering both flexibility and convenience. Online programs involve the same level of rigor as their campus-based counterparts with features like videos, interactive case studies, and self-paced lectures.

Most online data science programs would require you to visit the campus on just one or two occasions to complete an immersion experience. This gives you a chance to you’re your peers and professors and engage in networking activities. In fact, representatives of companies looking to hire fresh graduates sometimes survey these immersion experiences when scouting for talent.

  • Accelerated Programs: Accelerated programs provide a more intensive academic schedule, thereby allowing you to complete a master’s degree in about 12 months instead of the usual 18 to 24 months.
  • Part-Time Programs: Designed for busy working professionals, part-time programs allow you to complete your master’s program through a more relaxed academic schedule. Part-time programs often take about 32 months to complete.

 


 

Step 5. Begin a Career in Data Science

A May 2016 Harvard Business Review article reported that firms are currently spending an estimated $36 billion on storage and infrastructure—and that number is expected to double by 2020. And according to Burtch Works, this spells plenty of opportunities for master’s-educated data scientists.

Companies looking to recruit big data professionals are luring talent with impressive salaries and equally impressive bonuses. January 2015 Burtch Works statistics reveal an average salary of $120,000 ($183,000 for managers) for data scientists and a mean bonus of 14.5 percent.

Data scientists enjoy successful careers in nearly every industry and sector in the U.S., including:

  • National security
  • Business intelligence
  • Law enforcement
  • Financial analysis
  • Healthcare
  • Disaster preparedness
  • Finance
  • Government
  • Public health
  • Retail

The federal government remains a large employer of data scientists. For example, the Central Intelligence Agency (CIA) hires data scientists to organize and interpret big data that is used to inform decisions, drive operations, and shape law enforcement and intelligence technology and resource investments.

The big names in search – Google, Yahoo, and Bing – are all big employers of data scientists, as are social network companies– Facebook, Twitter, Instagram, and LinkedIn.

In engineering, telecom and software, Intel, Verizon, Boeing, Apple and Oracle remain big players in data science.

Some of the biggest employers are also the very companies that develop the most often used software in data science, including:

  • Palantir
  • Pixar
  • SAS
  • Alpine Labs
  • Pivotal
  • Tableau
  • Teradata

In finance, Wells Fargo, Citi, and Bank of America are among the top firms to retain data scientists; while in e-commerce Amazon, e-Bay and Google represent the top employers. The gaming industry is also huge, with the likes of Tencent, Activision Blizzard, Sony and Microsoft leading the way

Combined, the top 20 private sector employers of data scientists in the US supported a full ten percent of all data scientist jobs in 2016:

  1. Microsoft
  2. IBM
  3. Amazon
  4. SAS
  5. Google
  6. Accenture
  7. Oracle
  8. LinkedIn
  9. FICO
  10. Bank of America
  11. Citi
  12. Tata Consultancy Services
  13. Facebook
  14. Cognizant Technology Solutions
  15. Wells Fargo
  16. Capgemini
  17. eBay
  18. Apple
  19. Hewlett-Packard
  20. EMC

 


 

Step 6. Keep Learning and Specialize – Certification Options

As a data scientist, it is important to always stay one step ahead, particularly given the speed at which Big Data technology is evolving, not to mention the rapid evolution of how the technology is being used.

MOOCs and boot camps remain popular offerings for data scientists, regardless of skill and level of education. There are boot camps for entry-level data scientists, doctoral-prepared data scientists and nearly everyone in between.

The pursuit of professional certification almost always pays off because it assures employers you have the right skillset for the job.

Some of the most coveted certifications in data science include:

Back to Top