Online Engineering Programs > FAQ > How Do I Become a Data Scientist?

How to Become a Data Scientist

Find schools


Data scientists are the artists of the numeric world, shaping raw figures into something truly meaningful. In practical terms, data scientists use scientific principles and methods to glean knowledge and trends from large datasets. Their findings provide valuable, actionable business and research intelligence that can improve organizational efficiency, productivity, and earnings. They also spot trends and correlations in research data, predict sports outcomes, or project election results. The Data Science Association notes that data scientists are distinct from business analytics professionals, who tend to run the same statistical analyses on a regular basis. Data scientists, by contrast, are far more likely to conduct novel analyses.

This guide examines how to become a data scientist, including the typical knowledge, skills, education, and certification for this high-growth career.

Data Science Skills & Knowledge

Successful data scientists have a solid understanding of math, statistics, algorithms, data modeling, and machine learning and know how to communicate their findings—verbally and visually—to colleagues of various backgrounds. They are innately curious, tenacious, and willing to challenge even the most common perceptions and biases. While many of a data scientist’s abilities are seemingly innate, the right education cultivates the practical and specialized knowledge driving their work. Data scientist programs introduce students to the technical and analytical tools used in today’s job market, including a myriad of coding languages and applications:

  • Python: A long-used, general-purpose programming language ideal for beginners
  • Java: Another common, general-purpose programming language
  • R: An open source language for data analysis and visualization
  • MatLab: A language that supports mathematical modeling
  • SQL: Common language for database management
  • Scala: A Java-friendly language ideal for working with real-time data
  • Julia: A relatively new but growing open source programming language
  • SAS: SAS is both a proprietary language and a popular data analytic software suite
  • Tableau: A data visualization platform
  • Hadoop: An open source programming framework that processes and stores data for use in other platforms, including Apache Spark
  • Apache Spark: An open source programming framework created for speed

    Why Data Science?

    Data science training is a powerful investment. When LinkedIn (Oct. 2016) released an analysis of the skills employers want most in 2017, statistical analysis, data mining, and data visualization all cracked the top 10. According to Forbes (Oct. 2016), notable employer demand and high earnings also repeatedly cemented data science’s status as the “best job in America” or the “sexiest job of the 21st century” in high-profile publications. Where does one begin?

    Steps to Become a Data Scientist

    Data science is a relatively new field, so while it is incredibly popular, there is no standard or traditional path into it. That may change. As more colleges and universities add data science programs and professional organizations design more certifications, employers’ training requirements will probably tighten. Here is a detailed step-by-step breakdown of one possible pathway to joining this field.

    1. Research Data Science
    2. Most Americans probably recognize the terms “big data” or “data analytics,” but how many know what these fields entail? While areas such as “business intelligence” have existed for decades, distinctions across sectors like business intelligence, data analytics, and data science can seem muddled. Taking the time to research each field carefully lets future data professionals identify a suitable discipline and training path. Data science blogs and organizations help tremendously. When researching prospective careers, please note that not all data scientists are necessarily called data scientists. Titles to look for:

      • Data scientist
      • Biostatistician
      • Statistician
      • Data engineer
      • Financial quantitative analyst
    3. Choose a Training Path
    4. According to Forbes, data science training typically falls into three categories:

      • Self-Directed Education – An abundance of free online resources give prospective data scientists a chance to explore and become acquainted with popular data tools, programs, and languages. Self-taught learners can then certify their knowledge with professional certifications from organizations like SAS and Cloudera (see below for more information). This path might be the least expensive road to becoming a data scientist but not necessarily the most fruitful in a market increasingly inhabited by college graduates, many of whom pursue the same certifications. Data science is also more scientifically-oriented than, say, business intelligence; therefore it can be difficult to establish the higher-level methods and concepts demanded of data scientists without some degree of formal training.
      • Accelerated Nano-Programs & Boot Camps – Organizations like Metis and Udacity now offer accelerated training—sometimes called boot camps or nanodegrees—in data science. While they may seem like middle-of-the-road options for students seeking formal training without the time and financial investment of an advanced degree program, mini-programs are far less comprehensive. Ideal candidates include students who already have degrees in fields like math, statistics, or computer science who want to segue into the field of data science. Nonetheless, such programs can give formerly self-taught data scientists an edge in the job market.
      • Data Science Degrees – As demand for data scientists grows, so too does the number of data science degree programs. A 2017 report from PricewaterhouseCoopers (PwC) offered the following breakdown of minimum educational requirements for data scientists in 2015. Note that about 98 percent of positions required formal degrees:
        • Bachelor’s degrees: 59%
        • Master’s degrees: 27%
        • Ph.Ds: 12%
        • Other: 2%
      • A bachelor’s degree in data science is a solid start, but Forbes encourages students who want to get ahead of shifting market demands to aim for master’s degrees. While not the most common credential, this more advanced may become an increasingly common entry-level educational requirement. One could also pursue a master’s in a closely related field so long as it includes coursework in high-level statistics, computer programming, and other advanced analytical methods. In general, only students who want to teach at the postsecondary level, pursue leadership positions, or work in research need doctorates in data science.

    5. Choose a Data Science Degree Program (Recommended)
    6. According to the U.S. Bureau of Labor Statistics (BLS 2016), employment and earning potential increases with education. Employers often prefer candidates with formal data science degrees, but programs vary. Prospective students must decide which degree level to pursue, considering budgetary constraints and program credibility. Also, there are both traditional campus-based programs and online data science degrees available, options paying thought to program accessibility and personal learning style.

    7. Consider Graduate Certificates in Data Science
    8. Like mini-programs or nano-degrees, postsecondary certificates in data science offer targeted and advanced training in less time than most degree programs. They are particularly suitable for students with bachelor’s degrees in data science or a related field, master’s degree-holders who want an even bigger advantage in the job market, and graduates with degrees in related fields. Some data science certificates offer specialized training in the latest skills, software programs, coding languages, and more. They may also be designed to prepare students for specific professional certifications.

    9. Explore Professional Data Science Certifications
    10. Data science certifications are industry-specific credentials granted by professional organizations and vendors that denote particular skills and knowledge. They are not to be confused with “postsecondary certificates” detailed above. Though rare, a handful of colleges and universities offer their own certifications. Examples of some of the most common vendor- and college-geared data science certifications:

      Vendor Certifications

      University Certifications

    11. Keep Pace With the Latest Data Science Trends & Tools
    12. Data science is both a highly technical and fast-evolving field: new algorithm methods, applications, programming languages, and other tools emerge constantly. Data scientists who keep their skills sharp serve their employers—and their resumes—well. Some of the ways to go about it:

      • Regularly read data science industry publications; pay special attention to new products, employer surveys, and annual lists of the most in-demand skills and certifications.
      • Subscribe to blogs from well-known data scientists.
      • Review academic journals for the latest studies and research.
      • Join a professional organization or other community that promotes communication and networking among data scientists.
      • Invest in new professional certifications.
      • Consider specialized postsecondary certificates.
      • Mentor rising data scientists to keep tabs on what colleges are teaching and reinforce existing knowledge.

    Career and Salary Outlook for Data Scientists

    Data scientists are in high demand. PwC projected that the demand for data scientists will grow by nearly 40 percent between 2015 and 2020, significantly faster than most data and business analytics professionals. Employer valuation of data professionals certainly deserves some of the credit, but there are other factors at play. A 2016 report from the McKinsey Global Institute predicted that the volume of available data will at least double every three years for the next few years thanks larger and more affordable data storage coupled with a significant upswing in data captured by wireless sensors and virtual reality. Increased competition for qualified data scientists makes them especially valuable—something reflected in their earnings.

    Data Scientist Salaries

    The relative newness of data science and confusion over what it entails makes it difficult to discern how much data scientists typically earn, but all sources reviewed suggested it is much higher than the national average for all professionals. Education, professional experience, certifications, and geographic location all influence one’s earnings. As of May 2017, Glassdoor reported a mean wage of $113,436 for data scientists nationwide. A 2015 PwC analysis of more than 48,000 job postings found that the average advertised salary was $94,576, which was still the highest of any analytics job surveyed that year.

    Data Science Career Boosters

    How much data scientists earn depends on several variables, though prospects tend to improve with education and experience. Candidates with bachelor’s degrees usually fare better in today’s job market than their self-taught competition, but not as well as candidates with master’s degrees. One could say the same of certified vs. non-certified data scientists. According to PwC, the following skills also won higher salaries in 2015:

    • Data mining
    • Machine learning
    • Data warehousing
    • Extraction, transformation, and loading (ETL)
    • Mathematical modeling
    • Data optimization
    • Operating systems
    • Scripting languages
    • Software development principles
    • Product development