Last Updated: January 13, 2020
"Data science" may describe a diverse array of fields and positions, but generally, data scientists work in teams or independently to analyze data and address targeted problems. After collecting and cleaning data, these professionals use programming languages and software tools such as Tableau to visualize data, identify meaningful patterns, and generate algorithms and experiments. Data scientists also present their findings and propose organizational solutions.
If you have an appetite for problem-solving and aptitude in quantitative fields such as computer science, information technology (IT), and advanced math (e.g. statistics), consider a career as a data scientist. Read on for information on data science education, salary, job outlook, career options, and professional resources.
What Does a Data Scientist Do
Data science ranks among the nation's most lucrative careers, as most companies must use data to inform their real-world decisions. As a result, this profession continues to grow rapidly, and the U.S. Bureau of Labor Statistics (BLS) projects data scientist and statistician jobs to grow at above-average rates over the next decade.
Data scientists help companies make data-driven decisions by creating mathematical models to address real-world problems. They provide companies with a basis for more sound, data-based decision-making, as opposed to relying on gut feelings or the whims of executives.
Data science is experiencing a boom, and online data science master's degree programs become more accessible every year. Read on for a comprehensive guide to the profession, its salary potential, the skills it requires, and resources to help you decide whether to pursue a data science master's degree.
Key Hard Skills
Data science professionals must possess concrete, learnable abilities, or "hard skills." Depending on the nature of their work, data scientists usually have skills in several areas of expertise, including computer programming languages and software tools.
- Big Data Analytics: Big data analytics involves the use and analysis of large data sets to help companies use information about consumer trends and market patterns. The term "big data” refers to data sets so complex that normal data processing software could not work with them.
- Java: Programmers use Java, a programming language, for a wide variety of purposes. One of the simplest and most common computer languages, Java is essential knowledge for any computer science-related professional, including data scientists.
- Machine Learning: People who study machine learning examine the models, algorithms, and mathematics that computers implement to improve performance. Machine-learning algorithms that use data fall under the umbrella of data science.
- Hadoop: Created by the Apache Software Foundation, Hadoop is a group of open-source software tools used by data scientists to work with big data. Data scientists who master the Hadoop suite of tools often stand out against competing candidates.
- Python: Python is a general-use programming language created in 1991. Many programmers consider mastery of Python as basic knowledge, necessary for any computer science professional. Python includes interpreters, which allow it to work with a variety of operating systems.
- Data Mining/Data Warehouse: Data mining is the process of looking within a large set of data for previously unrecognized patterns or insights. A data warehouse is a database system created specifically for data analytics. Data scientists should be proficient in both data mining and data warehouse creation.
- SAS: SAS is a suite of software products created specifically for data management and analysis for business insights. Use and mastery of SAS is one of the absolute foundational skills for data science professionals, helping them create value for their employers by addressing real-world problems.
- SQL: Short for Structured Query Language, the programming language SQL works specifically with database management. All prospective data scientists should maintain a firm grasp on the latest trends and advances in SQL.
- R: R, a programming language, works with statistics and graphics. Many computer science professionals consider R to be foundational knowledge. Given its importance in statistics, R is particularly relevant to data scientists.
Key Soft Skills
Unlike hard skills, soft skills refer to the general characteristics of a successful worker in a specific career or across careers. For example, adept data science professionals have soft skills in communication, organization, and problem-solving. Soft skills for data science professionals also include analytics talent and business acumen.
- Analytics: Understanding and interpreting data constitutes the heart of data science, so business and data analytics skills prove essential to data scientists. Some employers seek skills in machine learning, a branch of analytics which creates performance prediction systems.
- Business-Focused: Many employers seek data scientists who can interpret data and use it to inform business strategies for improving efficiency, productivity, and sales. Data scientists can specialize in the data science field of business analytics, in which old performance data guides present and future business moves.
- Communication: Most data scientists must communicate their findings, interpretations, and ideas to employers and colleagues. Consequently, clear writing and public speaking skills make data scientists more effective in collaborating with team members and superiors.
- Problem-solving: Most companies, governments, and nonprofits turn to data scientists for their ability to solve problems through data-driven insights. Data scientists can help organizations clarify which aspects of a problem can be solved using data.
- Organization: Data scientists need good organizational skills to address projects involving complex parts and data sets. Professionals who understand how to organize data lakes (large repositories of raw data), for example, may know how to improve company outcomes by sourcing more representative amounts of data.
Data scientists' tasks depend on their position and industry. Database administrators, who may work in large data set industries such as education or healthcare, spend their days organizing, storing, and ensuring the accessibility and security of an organization's data. These administrators also modify, test, and maintain databases, updating them when necessary.
Approved users, such as data analysts and other data scientists, access this data and interpret it by conducting and analyzing data studies. Some database administrators design and create data processes to improve data accuracy and usefulness.
How Do I Become a Data Scientist?
Aspiring data scientists usually begin by earning a bachelor's degree in data science or a related field such as IT, computer science, or math. Many programs include an internship component, which provides field experience useful for entry-level data science jobs. Data scientists seeking advanced positions often earn master's degrees in a data-related field.
What Is the Difference Between a Data Scientist and a Data Analyst?
Data analysts may only interpret data, but data scientists also employ business acumen, translating their analyses into useful terms for decision-makers.
How Long Does it Take to Become a Data Scientist?
Since most data scientists hold at least a bachelor's degree, becoming an entry-level data scientist typically takes at least four years.
How Much Does a Data Scientist Make?
Salaries vary based on credentials and position, but data scientists usually make generous incomes. According to PayScale, data scientists make a median annual income of $95,998.
What Do Entry Level Data Scientists Do?
Most entry-level data scientists work in teams, using data to solve problems. These professionals often collect, clean, organize, store, and analyze data, but they may also create predictive models, data visualizations, and simulations.
Data Scientist Salary Information
Data scientist salaries vary widely based on location, industry, credentials, and experience; however, as part of a valuable and burgeoning field, data scientists can usually expect above-average salaries and job security.
According to the BLS, data administrators (one type of data scientist) earn a median annual salary of $90,070 and can expect a nationwide 9% job growth rate between 2018 and 2028. According to PayScale, data scientists working in IT -- the industry employing the most data scientists -- earn a median annual salary of $91,260. According to the same source, data scientists with 10-19 years of experience make median salaries exceeding $117,446.
Data scientists often earn the most in IT and management industries, but most industries pay well for upper-level data science positions. Professionals working in New Jersey, Washington, California, Connecticut, and New York tend to make higher salaries than their counterparts in other states.
Data Scientists by Job Level
|Entry Level (0-12 Months)||$78,599|
|Early Career (1-4 Years)||$89,248|
|Mid-Career (5-9 Years)||$100,434|
|Experienced (10-19 Years)||$117,446|
How to Become a Data Scientist
Earn Your Degree
Most data scientists hold at least a bachelor's degree in a quantitative field, and many hold master's degrees, as well. Some schools offer degrees in data science, but degrees in related, quantitative fields, such as computer science, prepare students for data science careers, as well. These students learn to modify and maintain database structures which collect, organize, and store data.
They also learn to create algorithms using programming languages such as Python and R. Data science coursework usually includes statistical modeling, data visualization, text mining, and machine learning, as well. Most data science degree programs require an original capstone project and/or an internship, which gives students data science field experience and mentorship opportunities.
Most data scientist positions mandate a certain level of education and work experience. Even companies hiring for entry-level positions prioritize candidates with relevant internship experience, professional portfolios, and skills with specific software tools and programming languages. Employers typically expect entry-level data scientists to hold 2-3 years' experience with programming languages, such as R. Most data science positions also require general skills in areas such as business strategy, IT trend knowledge, and communication.
Mid-level data science professionals usually need a bachelor's degree and at least five years of relevant work experience, or they may hold a master's with three or more years of experience. These professionals typically boast mastery of advanced programming languages like Python or Java, as well as visualization software such as Tableau.
Data scientists must commit to lifelong learning to stay competitive in this rapidly developing field. Employers look for data scientists with skills and credentials relevant to the available position. A related degree and work experience may suffice for some data scientist roles, but it helps for these graduates to earn additional credentials.
Schools, professional organizations, online learning companies, and technology companies offer credentialing programs in specific programming languages and skills. For example, the Data Science Council of America offers certification in big data engineering, IBM offers an applied AI credential, and the CAP provides data analytics certification. Some certifications denote both the skill and the level achieved, such as the Cloudera certified associate data analyst credential or the certified associate in Python programming credential.
Types of Careers in Data Science
Data science opportunities vary based on education, experience, and credentials. Skilled graduates with associate or bachelor's degrees in computer-related fields may qualify for entry-level positions in computer programming or systems analytics. As they obtain requisite work experience, educated data science professionals may advance into computer and IT management or network architect positions. However, data scientists who conduct advanced organizational problem-solving and leadership skills usually hold graduate degrees.
As outlined in the following sections, industry and location also influence data science salary and job availability. Generally, however, successful data scientists can anticipate high salaries. According to PayScale, data scientists in IT made a median annual salary of $91,260 as of 2018.
Careers for Data Science Graduates
Computer and Information Systems Manager
Typically employed as IT directors, IT security/project managers, chief technology officers, or chief information officers, computer and information systems (CIS) managers help to generate and implement company technology goals through installing and managing computer systems. CIS managers boast computer-related bachelor's or graduate degrees and relevant work history.
Median Annual Salary: $142,530
Computer Network Architect
These professionals create local, wide-area, and intranet data communication networks. Computer network architects usually hold a computer-related bachelor's degree and have some work experience.
Median Annual Salary: $109,020
Using programming languages to compose and test computer code, computer programmers convert program designs into functioning applications and programs. They need at least an associate degree and to specialize in multiple programming languages.
Median Annual Salary: $84,280
Computer Systems Analyst
Computer systems analysts and architects analyze computer systems, fix problems, and create means for optimizing efficiency. Companies hire analysts based on their IT or programming skills, but computer or IT bachelor's degrees make candidates more competitive.
Median Annual Salary: $88,740
Software developers create computer programs, applications, and systems. Most of these professionals hold computer science bachelor's degrees and have strong programming skills.
Median Annual Salary: $105,590
Where Can I Work as a Data Scientist?
Given how many industries currently need data scientists, talented and qualified data science professionals enjoy excellent job and salary prospects in many parts of the country. As the tables below indicate, California and New York rank among the top five states for data scientist employment and pay.
Job prospects may appear rosy for data scientists in many states, but location does influence job availability and earning potential. According to the BLS, Texas and California currently employ the most data scientists, followed by New York, Florida, and Virginia. California and New York also rank among the top five states for median salaries, though New Jersey and Washington top that list.
|States With the Highest Employment Level of Data Scientists(Applications)||Number of Data Scientists (Applications) Employed|
|Top Paying States for Data Scientists||Annual Mean Wage|
Data scientists work in a diverse array of company and industry settings. Some work for large companies, which have better employee resources and training, while others thrive in flexible, growing environments, which usually exist within small companies. Some data scientists work independently as freelancers or consultants.
BLS data indicates that data scientists typically work in the computer systems design industry, which is followed in popularity by the management, education, telecommunications, and credit industries.
|Industries With the Highest Level of Employment for Data Scientists||Number of Data Scientists (Applications) Employed|
|Computer Systems Design and Related Services||16,740|
|Management of Companies and Enterprises||9,660|
|Colleges, Universities, and Professional Schools||5,700|
|Credit Intermediation and Related Activities||4,720|
Working as a Freelancer
Some data scientists forego the security and benefits of traditional employment and strike out on their own as freelancers. Freelancing offers obvious practical benefits, such as schedule and location flexibility. Meanwhile, freelance professionals can pursue only the projects that interest them most. This freedom typically appeals most to creative data scientists.
Accomplished and experienced data scientists with demonstrated talent typically perform well as freelancers and consultants, as actively sought-after freelancers who have made names for themselves as data scientists usually enjoy more choices and better pay than less-experienced data professionals.
Tim Shea has worked in data science and advertising for over 20 years at agencies like Razorfish and TBWA\Chiat\Day, and on data platforms like DataSift and Canvs.AI. He also founded three big data startups of his own and worked with clients such as Nissan, Pepsi, The Grammys, Princess Cruise Lines, Facebook, Twitter, and Reddit. Latticework is Shea's latest creation, leveraging his unique skill of bridging the divide between data scientists and marketers.
Why did you decide to pursue data science?
Having written database software for over 20 years, I slowly saw applications I was building swelling from hundreds of records to hundreds of millions of records. All of a sudden, an application would need 100 million geospatial locations or 100 million natural language conversions embedded in them just to launch. In other words, it was no longer good enough to just know SQL to be effective; you had to understand NLP, trigonometry, statistics, and machine learning. I guess you could say that data science found me, not the other way around.
What are some of the most challenging aspects of working in data science?
The hardest part of data science is that the skills are incredibly esoteric and arcane, but the output absolutely has to be understandable by mere non-data scientists. So, the trade really forces you to be able to zoom in and out of levels of complexity, like very few other trades.
At Latticework Insights, we champion our ability to be "data scientists that speak advertising” -- two skillsets that often face an enormous conversation gap.
The most rewarding aspects of the job?
More often than not, when clients come to us, they've hit some sort of wall. Maybe they've burnt themselves out spinning their wheels on a complicated data problem, or maybe they weren't even aware that data science could apply to their project. When we can jump in and add some new technique, sort of turn the problem on its side, and we can see a true "a-ha” moment on the client's face, those are super rewarding moments.
Was it challenging to find a job in the field?
Absolutely not. There are far more companies in search of data scientists than there are data scientists searching for companies. Harvard Business Review called data science "the sexiest job of the 21st century,” and in our experience, they are spot-on.
What kind of job settings have you worked in?
I've worked in every setting, from big ad agencies (breaking into private APIs to help creative directors come up with campaign ideas) to tiny, two-person startups (building machine learning apps to scan millions of restaurant menus for healthy food) and big enterprise sales organizations (wrestling with Twitter and Facebook firehoses to close sales).
Advice for newcomers to the profession?
I've given very similar advice before in previous interviews. Data scientists need to be incredible lateral thinkers, and in my experience, I've seen that the best programmers almost always come from a liberal arts background, like art history or philosophy; have a wide range of life experiences; and a gift for persuasion. Without a solid understanding of context or a very particular domain expertise, all of your data science, computer science, or statistics education will fall flat.
What are some of the best ways you gained experience outside of primary education?
In my opinion, the basic blocking and tackling of data science is just being in the trenches. Get involved with lots of different types of projects, read widely outside of your discipline, and really push yourself to understand how non-data scientists do their jobs. Find tons of speaking gigs at events, present in front of peers, and build lots of confidence in being persuasive.
What direction do you see your career path trending in?
I'm looking forward to the next 5-10 years, when data science skills become enormously indispensable, and it's going to be exciting to watch and hopefully help nurture a new, young, hungry generation of kids with wildly diverse experiences come through the system and completely change the game.
Continuing Education for Data Scientists
To keep up with this rapidly changing and growing field, data science professionals must typically continue their education through courses, trainings, and certification programs. Data science subjects include general topics such as machine learning and big data, plus specific programming languages such as Python or R and company-specific software like Microsoft Azure or Oracle SAAS.
Continuing education includes university courses from programs at institutions like Stanford, Duke, or Rice. It also includes online training and certification programs from professional organizations (such as the Institute for Operations Research and the Management Sciences) and IT companies (like Microsoft, IBM, and Oracle). More specific descriptions of these continuing education resources and professional organizations appear below.
Continuing Education Resources
- Big Data and Data Science Online Program In partnership with Simplilearn, an online training platform, Duke's four-course data science continuing education program develops professionals' expertise in areas such as Python, R, SAS, and Tableau.
- Microsoft Learn Microsoft Learn provides 80+ hours of free, task-based learning for data science professionals seeking fluency in programming languages such as Flow, Azure, and Dynamics.
- Oracle University Oracle's IT training and certification programs teach professionals to use Oracle cloud, software, and hardware products.
- Rice University Rice University's Glasscock School of Continuing Studies and other internal organizations offer data science courses, workshops, and bootcamps.
Professional Development Resources
- Metis Data science education provider Metis offers professional development courses, bootcamps, and online training for data science professionals and corporations.
- IBM Data Science Professional Certificate Advertised as accessible to anyone with computer skills, this professional certificate program entails nine data science courses and involves hands-on project and portfolio creation.
- Stanford Foundations for Data Science Certificate Stanford University's Center for Professional Development offers a three-course certificate program, which introduces students to Python, R programming, and statistics.
How Do I Find a Job in Data Science?
The data science field boasts high growth projections and salaries, but finding a data science job may still prove challenging for recent graduates. Fortunately, there are many helpful resources for data science job-seekers. Personal network connections with mentors and colleagues often prove invaluable, as do job fairs, conferences, job boards, and professional organizations. See below for descriptions of job boards and professional organizations that data scientists might find useful.
Kaggle Job Board
In addition to its data science courses, competitions, and discussions, Kaggle -- a large, diverse, Google-owned community -- features a job board that lists available positions across the globe.
icrunchdata Job Board
The icrunchdata data science job board index includes a comprehensive data science job board report based on data drawn from data science job sites. It identifies current labor information and potential directions for this field.
Data Elixir Job Board
This job board appears on Data Elixir's website, newsletter, and social media pages. Job-seekers can search the board by location and keyword.
Data Science Central Job Board
Data Science Central, a media and digital publishing company, provides this searchable job board for the company's internet community of data science, machine learning, and analytics professionals.
Professional Resources for Data Scientists
Data science professionals often benefit from the networking, continuing education, and other professional development resources available from data science professional organizations. These organizations offer exciting conferences, publications, and discussions to help data scientists stay current in a rapidly changing field.
In addition to the learning and community these organizations provide, members benefit from industry and conference discounts, exclusive job boards and directories, leadership opportunities, and awards and scholarships. See below for descriptions of key professional associations for data scientists.
- The Data Scientist Association A nonprofit working to improve data science practice and support the field's practitioners and students, the DSA establishes professional ethics and other standards, accredits schools, and provides data science resources.
- The Association for Computer Machinery's Special Interest Group on Knowledge Discovery and Data Mining Committed to advancing data science knowledge and education, SIGKDD generates standards, curricula, and publications. It also runs conferences, grants awards, and provides news and other information.
- International Institute for Analytics This international company connects data analytics experts through its research and advisory network. IAA's services include assessment, consulting, and leadership training workshops.
- Institute for Operations Research and the Management Sciences Advertising itself as the leading international operations research and analytics association, INFORMS offers conferences, publications, and an impressive variety of other resources for continuing education, career-building, and networking.
- Digital Analytics Association By providing professional development, community, and other resources, the DAA encourages the use of data to enhance the digital world. This association posts jobs, runs events, and encourages gender equity in analytics.