MAKE BIG SENSE OUT OF BIG DATA IN ONE OF THE MOST

IMPORTANT NEW FIELDS OF THE 21st CENTURY

Earnings

Entry Level annual earnings: $85,000

Median annual earnings: $115,000
High Level annual earnings: $170,000
Top Level annual earnings: $250,000

Traits Required

Technical mastery
Analytical skills
Business acumen

The Work You Will Do

Analyze data for actionable insights

Organize massive data sets
 Create algorithms
Present findings

Where You Will Work

Offices in big cities

Tech companies

Education Required

PhD
Certification

Pluses

High Job Satisfaction

Lucrative
High demand

Minuses

Can be overwhelming
Competition is increasing
Requires a high degree of education
Introduction
Data scientists have a fascinating job. They are the invaluable
professionals that know what to do with the mountains of
information that is being created and gathered from every corner
of the world every single minute of the day. In this highly digitized
world today, everyone seems to have an urgent need to do
something with all the “Big Data” for fear of missing out on
something. Businesses, from banks to car manufacturers, depend
on data scientists to help them make important decisions that will
bolster their bottom line and edge out the competition.
Government agencies at all levels use their findings to make
important policy changes. Even nonprofit organizations depend
on data scientists to help them improve the lives of countless
people.

It is a challenging job to take an enormous pile of information and

distill it into something useful. There is simply too much
information for the average person to process and use. It is the
data scientist’s mission to analyze data for actionable insights, but
theirs is no easy task. Using a combination of math, statistics, and
programming skills, data scientists are able to compile, organize,
clean, and interpret any number of data points. They then use
their expert analytic skills to uncover hidden solutions that can be
applied to a wide range of purposes.

It takes some rigorous education to become an effective data

scientist. They need to become expert programmers, using tools
like Python, SQL, R, and Scala.They need cutting-edge skills in
machine learning, data visualization, artificial intelligence,
modeling, and optimization. They have to be expert in using Excel.
Most data science professionals have graduate degrees, either a
master’s or PhD, in data science, statistics, computer science, or
mathematics. Data scientists come from a wide range of
educational backgrounds, though some technical training is
always necessary. For example, those educated in business or
human behavior can offer unique perspectives that allow them to
arrive at more accurate and relevant conclusions.

There is a wide array of opportunities for professionals with the

advanced skills and know-how to work in data science roles. The
demand is high now – well above the national average – and the
trend is zooming upward. Over the next few years, the analytics
market is expected to grow by leaps and bounds until it is at least
one-third of the global IT market. Organizations of all kinds are
scrambling to integrate data science into their business
processes. Although the demand is rising steadily, there is a huge
deficit on the supply side of qualified employees, creating
excellent prospects for future data scientists.

Those in the field can look forward to a promising job market and
longevity, and there is more good news as well. Data scientists
enjoy excellent compensation for their work, with a median
income of almost $100,000 and the potential to earn $250,000 a
year. The pay really stands out as an attractive feature, plus there
are also reports of high job satisfaction. No wonder data scientist
has been voted the Best Job in America four years in a row!
What You Can Do Now
Your primary focus in high school should be doing well and
getting ready for college. Data science is as much a profession as
lawyer, banker, or software engineer. That means there is plenty
of competition from other smart kids who want to get into a good
college. The most important class to ace is calculus. Although not
often available at the high school level, statistics is another good
choice. Take AP level classes if possible. For the rest, pick classes
that will set you up for a STEM (Science, Technology, Engineering
and Mathematics) major, such as chemistry and computer
science.

Make some room for flexibility in case you change your mind
about data science. Classes like history, a foreign language, and
public speaking and essay writing bolster your communications
skills – useful in just about any field.

Since data science has become a hot field, it has started to show
up in high schools. If your school is offering this class, you are in
luck. Do not hesitate to sign up. Not only will it offer some
preparation for future studies, but you also will learn more about
the multitude of opportunities that await future data scientists.

Learn to code and start building a portfolio of work. Learn Python

and do a project or two. Bolster your coding chops by competing
in hackathons like Kaggle.

Explore this career to make sure it is a good fit. Talk to actual data
scientists to get some advice on what you should be doing to
prepare for college and beyond. If you have family or friends
working in the tech sector, they probably know someone you can
talk to. Your counselor can also help you arrange for job
shadowing, internships, and field trips.
History of the Profession
Compared to most careers, this one is very new. The trailblazers
and pioneers have been around for a while – even before the
explosion of big data and the ability to make sense of it – but
nearly eight out of 10 data scientists on the job today have been in
the field for five years or less.

It all started with statistics and a mathematician named John

Tukey. In 1962, Tukey predicted a tectonic shift in the world of
statistics toward data analysis. Acknowledging the growing power
of modern electronic computing, and the merging of statistics
and computers, he foresaw a time when statistical results could
be presented in hours rather than days or weeks. He wrote in The
Future of Data Analysis that a new field concerned with learning
from data would emerge as an empirical science. Though not yet
named as such, he was the first to describe the field of data
science.

In 1960, the International Frederal for Information Processing

(IFIP) was established under the auspices of UNESCO. The
organization was recognized by the United Nations and joined
together over half a million professionals from 50 different
international societies and academies of science around the
world. The global organization conducted research, developed
standards for processing data, and promoted information
sharing. In 1968, the IFIP Congress presented a course plan titled
“Datalogy, the science of data and of data processes and its place
in education.” For the first time, the term “data science” appeared
in professional and academic publications.
Later, in 1977, the International Association for Statistical
Computing (IASC) was established. The organization’s mission
was to link traditional statistical methodology, modern computer
technology, and the knowledge of domain experts in order to
convert raw data into meaningful information and knowledge.

IBM and Apple released the first personal computers in 1981 and
1983 respectively. From that point on, computing evolved quickly,
providing the ability to collect data more easily. It would still be
nearly two decades before companies would start to convert that
data into actionable information. In fact, it was in 1999 that Jacob
Zahavi started reporting on the need for new tools to handle the
massive amounts of data that were available to businesses. He
pointed out that conventional statistical methods only work well
with small data sets and not with millions of rows of data. Zahavi
called for the special development of data mining tools to
overcome the huge technical problem of scalability.

Data Science in the 21st Century

Throughout the 2000s, various academic journals began to
recognize data science as an emerging discipline. Along the way,
the field was propelled forward by the introduction of tools,
methodologies, publications, and training programs. One of the
most important tools was SaaS (Software-as-a-Service). Created in
2001, this was the precursor to cloud-based applications. One
year later, the International Council for Science began publishing
the Data Science Journal. In 2005, the National Science Board of the
United States recognized the need for more experts in data
science. The organization advocated for a data science career
path to encourage more people to become data science
professionals.
By 2008, data scientist was a buzzword. The Harvard Business
Review dubbed it the “hottest job of the 21st century,” succinctly
declaring the purpose of a data scientist is to create value from
data through analytics. The publication was correct. Data science
had arrived. Technology had taken a giant leap and processing
speeds were blazing fast. Consumers were embracing
technology, especially mobile technology, that provided
companies like Apple and Amazon with the ability to collect
unprecedented amounts of personal data.

There was now a clear path for those who wanted to conquer the
mountain of big data and get paid well to do it. Job listings for
data scientists increased by 15 percent in 2011. Seminars and
conferences devoted to data science and big data sprang up and
multiplied like dandelions. In 2013, IBM proved (with statistics)
that 90 percent of the data in the world had been created within
the previous two years alone.

Data science became known as a source of profit for business

and was integrated into the corporate culture within the IT
industry. Data science degree programs emerged to train new
data scientists. Technical advances including machine translation,
robotics, and speech recognition moved it into other areas. It
gradually became an important part of academic research into
the biological sciences, healthcare, medical informatics, the
humanities, and social sciences. Data science now influences
governments, businesses and organizations of all kinds
worldwide.

Today, data scientists are essential to successful businesses, and

their value is reflected in the size of their paychecks. The demand
for data scientists is continuing to grow as more applications of
their skills are discovered. The amount of data being collected
continues to grow, too, along with the need for skilled
professionals to analyze it. Data science will continue to evolve as
new methods of data collection and dating mining enter the
picture. For example, in just the last couple of years, wearable
health trackers have become popular. Today, almost 80 percent
of healthcare consumers are willing to share, through this
technology, their personal data that tracks vital signs, location and
movements, and other lifestyle information. The only questions
for the future are, where will big data come from next and what
analysis techniques will be developed to provide even greater
insights?
Where You Will Work
When wondering about where the jobs might be, think big.
Generally speaking, data scientists work for big companies in big
cities. There are several industries with high demand for data
scientists, but Amazon is the leading employer overall, by far.

It should be no surprise that the biggest number of data scientists

work in the tech sector. All tech companies depend on them, but
the two top employers are Google and Microsoft. Other familiar
names with thousands of data scientists on salary include
Facebook, Quora, IBM, Apple, and AirBnB.

Outside the tech industry, major employers include financial

investment companies, professional consulting firms, banks,
pharma giants, and telecommunications companies. The rest of
the jobs are scattered across government, marketing, academia,
retail, and gaming. While most data scientists are full-time
corporate employees, independent technology consulting
services are emerging that serve an array of clients who prefer
not to hire data scientists full time. Data scientists who work
through these consultancies have the flexibility to work when and
where they want, and move around from industry to industry.

Due to the technical nature of the field, almost half of data

scientists work on the West Coast. The bulk of these are in
California, namely San Francisco, Silicon Valley, and Los Angeles.
Another large number are in Washington state, which is home to
Microsoft and Amazon. Even non-tech companies on the West
Coast appreciate the value of data science. For example,
Washington-based company, Starbucks, has several dozen data
scientists on board.

On the East Coast, most data scientists work in Massachusetts

and New York. Massachusetts, the primary tech hub of the East, is
the center for East Coast data science. In New York, there are
many data scientists working mostly for financial investment
firms and banks. There are a few other states with clusters of jobs
for data scientists. They include Colorado, Illinois, Texas, and
Georgia. In these states, jobs are concentrated in the tech hubs
and big cities like Denver, Boulder, Chicago, Dallas, Austin, and
Atlanta.

Work Environment
Data scientists work in a variety of settings with relaxed
atmospheres that are conducive to creative thinking, and in
offices designed for efficiency. In the tech sector, it can be a blend
of both. There, data scientists typically spend their days in open-
design office-like settings that allow them to work together in
teams and collaborate effectively on projects. Which environment
you will work in depends on the nature of the business, the
overall corporate culture, and the type of data science your job
calls for.

The tempo of the work environment also depends on the

company and the industry you work in. Some jobs, like those in
the financial industry, are in a high-pressure work environment
that emphasizes quick results. Some other organizations put an
emphasis on accurate results and therefore value slow,
methodical, detailed progress. Those looking to work in a tech
start-up, can expect the pace to be fast and relentless.
Data scientists often work within teams that may include
managers, IT administrators, programmers, statisticians, graphic
designers, and experts in the company’s products or services.
The Work You Will Do
A data scientist’s job, in the simplest terms, is to analyze data for
actionable insights. In practice, this is not as easy as it sounds.
Data scientists start with an enormous mass of data that may be
collected from a number of sources, such as customer
transactions, click streams, sensors, social media, log files, and
GPS plots.

The collected data, both structured and unstructured, has to be

cleaned, managed, and organized. Structured data comes
organized and formatted so it is easily searchable by basic data
mining, algorithms in spreadsheets, and relational databases.
Unstructured data is messy. It is more like human language,
unorganized, and with no predefined format. It is much harder to
collect, process, and analyze especially since most analytics tools
are designed for structured data. For this job, a data scientist is
likely going to have to create an algorithm from scratch.

Once the data has been wrangled, data scientists can get down to
the business of finding meaning in it all. Applying all their
analytical skills, they interpret the data in pursuit of a conclusion.
Sometimes they have been given questions to answer or
problems to solve. The questions may be specific or open-ended.
Other times, they have to determine the questions that need
answers. In either case, they are searching through the mountain
of numbers and stats to find something that could be useful to
their organization. They look for clues in trends, patterns,
weaknesses, surges, and other markers.
When the data scientist has arrived at a conclusion, the findings
must be presented along with recommendations to the team, to
management, and to other relevant parties, such as marketing or
IT departments. Presentations are usually in a visual format to
make it possible for nontechnical people to understand. It is
much easier to see clear patterns in heat maps, for example,
whereas they would not even be noticeable as hard numbers on a
spreadsheet.

Data science is a multidisciplinary field that requires knowledge in

computer science, mathematics, statistics, operations research,
and machine learning. The work itself can involve a wide range of
methods and techniques. For each project, the data scientist must
determine which are the most appropriate and effective methods
to apply. This can be a difficult process, since there are literally
dozens to choose from. The most commonly used include logistic
regression, data visualization, time series analysis, neural
networks, cross-validation, clustering, decision trees, and random
forests.

Big users of data science, like Amazon, do not depend on these

common methods. The company uses over 20 different data
science systems to operate its business. Being such a massive
business, their problems are very complex and simple algorithms
just will not do. Instead they utilize lesser known techniques like
the Markov chain models and Monte-Carlo simulations.

Generalist to Specialist

Many of the jobs available to new data scientists call for a

generalist. A data science generalist usually starts out working
with an established team of more experienced data scientists.
Generalists are expected to be familiar with all common tools
designed for big data analysis and have some experience working
with unstructured data sets. They typically start out spending
most of their time cleaning data.

When data scientists achieve more experience or change jobs,

they often get a chance to start specializing in a particular niche.
Data science is needed by nearly every government agency,
business, and organization in the world, so there are certainly
plenty of opportunities for specialization. For example, a data
scientist hired by the Defense Department might specialize in
analyzing threat levels, while another carves out a reputation for
helping small startups build and retain a customer base. Within
the business world, data scientists can choose to work in specific
industries, such as automotive sales, financial services, or
pharmaceutical manufacturing. They may focus on certain types
of business problems like pricing, logistics, or marketing.

Exactly what a data scientist does will vary among industries and
different companies within the same industry, or among
governments and different departments within the government.
A cell phone company, for example, may need a data scientist to
create an algorithm to look at millions of data points to figure out
what is the most likely reason customers will switch carriers. This
could be a stand-alone project or require ongoing monitoring. A
streaming service might need a real-time data management
program to instantly view past usage and make
recommendations that will keep viewers or listeners engaged.

Just about every federal agency relies on data science, and it is

common for agencies to partner with other agencies on
important issues. For example, NOAA (National Oceanic and
Atmospheric Administration) is at the forefront of combating
climate change. Its data scientists build data-driven tools that
other federal agencies use to prepare for the impact of climate
change in specific communities, businesses, and populations.

Though lagging behind the private sector and federal government,

data science is being used in local government as well. A data
scientist working for a big city government might gather and
analyze both internal administrative data, and data from outside
organizations, to find ways to improve municipal policy and
performance. With the local government, data scientists in the
housing department might look for ways to remediate blight,
while those in the health department might predict community
health issues.
The Professionals Speak
I Was Hired Right After College

“It has been about two years since I graduated from college and
started working as a data scientist. After earning a computer
science degree, I was hired by a consulting firm that specializes in
AI projects. It is a really fun job where I get to do all the cool stuff I
love, like coding and researching emerging trends.

I came to my first job with a lot of different expectations. Some

were met and some weren’t. Data science has its own pros and
cons. If you go into it with your eyes open, it is indeed a rewarding
field. It’s a rare job that has you reading a research paper in the
morning, writing an algorithm in the afternoon, and coding it up in
the evening. It’s like getting paid for doing my favorite hobby.

On the downside, if you are a code jockey, you might be

disappointed that you won’t get to showcase your Ninja coding
chops. If you are a math lover who relishes diving into complex
problems, the data wrangling and monotony of routine
programming will likely bore you. The trick is to strike the right
balance and appreciate the value of the different aspects of the
job in getting to the results you’re after.

The best thing about getting into data science, I think, is that the
experience can be a stepping stone to other endeavors. For
example, the amount of time you spend doing experimentation,
reading technical papers, and debating ideas with colleagues
could prove quite valuable in academic research. You could also
use your data science experience to work for government or
global organizations that need a sophisticated understanding of
AI to make policies around sustainable development or climate
change measures. And of course, if you’re the independent type,
you could always start your own consulting firm offering
recommendations to politicians, marketers, or other startups.”

I Mentor Future Data Scientists

“I work with a nonprofit group of senior data scientists that

mentor students interested in pursuing a career in the field. It is
often difficult for aspiring data scientists to know where to invest
their time and effort to make a successful transition from school
to the workplace. There is no one-size-fits-all solution, and the
field keeps evolving so fast that recommendations made today
may be out of date in the near future. But I do have some general
advice on how to get job-ready.

For those with no coding experience or STEM background, start by

learning what data science actually is before jumping into a
different major. Reach out to some data scientists and offer a cup
of coffee in exchange for a chat. You can find plenty of them in
your city on LinkedIn who will be happy to share their insights.
Follow a data science podcast or two. There are at least a couple
dozen of them to choose from. Once you’re sure you want to
proceed, learn Python and start taking data science MOOCs
(Massive Open Online Courses). They are all over the internet.

When you are ready to start your job search, don’t target a full-
fledged data science position. There are better ways to break into
the field. Data analyst or data visualization roles, for example, are
in high demand and are much easier to land. In either of these,
you will likely work on a team alongside senior data scientists. Pay
attention and you can pick up some tips and tricks that make it
possible to make a lateral move into your own data science role.
One important piece of advice is to read data science job
descriptions carefully. Data scientist is a broad title that means
different things to different companies. By carefully reading the
requirements for skill sets and experience levels, you can find a
role that is a good match for what you have to offer. You may be
more qualified than you think!”

I Am a Chief Data Scientist in a Nonprofit Organization

“I began my career in quantitative political science. It was mostly

research work, interviewing people, looking at historic
documents, and experimenting with statistics. I thought it was a
good gig until I did an internship with a nonprofit abroad that
worked on election monitoring and disaster relief. To connect
what was actually happening on the ground to the data collected
from real people in truly dangerous situations was amazing. It
opened my eyes to what data science could accomplish.

Today I work for a different nonprofit, this time in an organization

that is trying to make healthcare accessible for vulnerable people
like senior citizens. The best thing about my work is doing
something real that matters in people’s lives. In short, I have a
chance to make things better for people rather than just help a
company find ways to make a profit.

As the head of the data science group, I do a lot of the hiring, and
I interview a lot of people looking for their first data science jobs.
They are very often confused and intimidated by the hiring
process in this field. They see all these job posts wanting 10 years
experience and wonder if there is any such thing as a junior data
science path that doesn’t require specialization in an industry and
mastery of a hundred different tech tools.
I’m here to say there are definitely a lot of junior roles out there.
You just have to know where to look. Do not look at the smaller,
feistier startups. They might sound exciting, but it’s not a good
career move, especially if you’re their first data science hire.
Instead, look for mid to larger companies. Ideally, you will be the
one junior person on a team of six senior people. Working with
some really experienced professionals is the best possible
scenario. Your learning would be massive and your opportunities
for advancement will escalate faster than you can imagine.”
Personal Qualifications
Data scientists are highly educated and are fully versed in the
technological tools needed to do the work, but good data science
is much too nuanced for any software alone to arrive at
meaningful conclusions. There are many unique characteristics
that apply to data science. The most sought after data scientists
possess a special combination of technical mastery, analytical
skills, and business acumen. They have the ability to think
critically without making premature assumptions as they
transform massive amounts of raw intelligence into concise,
actionable plans. Other valuable soft skills include curiosity,
creativity, focus, and determination.

Critical thinking in data science is absolutely essential. Successful

data scientists have many more questions about the data than
the analytical tools can answer. They never dump raw data into
software and accept what comes out. On the contrary, the
primary aspect of critical thinking is the ability to question the
data, remove misleading outliers, and make valid adjustments.

Communication skills are needed to clearly translate complex

technical information for non-technical users, such as a
marketing or sales team. Most users are not interested in the data
or how you analyzed it. They only want to know how it can impact
their business and help them meet their goals. Effective data
scientists use “data storytelling” to make it easy for anyone to
understand their findings.

Business acumen is a helpful trait. The point of data science is to

help businesses improve their competitiveness and bottom line.
Think like an entrepreneur. You will need a deep understanding of
the industry you are in, how your company operates, and which
problems most urgently need solutions. That knowledge can help
you identify data points that the business can leverage in new
ways.

Intellectual curiosity is the fuel that pushes data scientists to

constantly pursue more knowledge. You are expected to have a
head for math, but what are you going to do with all those
numbers? When first looking at a mountain of data, it is
impossible to make any sense of it. There are so many data points
to consider! Data scientists spend 80 percent of their time in
discovery mode, asking questions about the data that have not
been asked before. It is inherent curiosity that drives them to sift
through it all, uncover opportunities for insights, and realize the
data’s full potential.

Collaboration is key. Data science is not a solo endeavor. Data

scientists work with just about everyone in the organization, from
company executives to customers. They collaborate with
management to develop strategies, designers and work product
managers to create new and better products, marketers to launch
dynamic campaigns, and developers to improve tools and
customer experiences.

Determination will get you past the many dead ends, wrong turns,
and bumpy roads that define data science. Quick answers do not
exist. Projects are usually long term and can be ongoing. It takes
grit and patience and maybe even a bit of stubbornness to get
past the frustration and arrive at the “Eureka!” moment.
Attractive Features
Survey after survey in recent years have proclaimed data scientist
the number one best job in America. Glassdoor voted it the “Best
Job in America” four years in a row based on high reports of job
satisfaction and the number of job openings, which exceeds most
other occupations. What really stands out, however, is the pay.
Data science is a very lucrative career option. Overall, half of all
recent college graduates are earning less than $45,000 a year.
Another 25 percent are making less than $35,000. Data scientists
are the clear winners, starting off with a median income of over
$90,000 and the potential to advance into positions earning
$250,000 a year.

Data scientists are in demand. New job seekers will find

numerous opportunities in this highly employable job sector. The
demand for data scientists is well above the national average. In
fact, it is the fastest growing job on Linkedin. This does not appear
to be a transient feature. It is one of the few jobs that is not in
danger of being thrown on the trash heap of obsolescence. On
the contrary, as technology progresses, these skilled
professionals will be increasingly needed to help their companies
keep up and move in the right direction.

Data scientists are highly respected for their expertise. Employers

rely on them to make smarter business decisions and provide
better results for their customers as well as their own bottom
line. This puts data scientists in a prestigious position within any
company. Their high standing is often well-deserved. In fact, their
work can even save lives. Data scientists have helped medical
science move to a whole new level by applying their skills to
medical imaging, genetics, drug creation, predictive medicine, and
a plethora of other healthcare advances.

Excellent pay and growing career opportunities will only take you
so far if the work is tedious and boring. Data science is anything
but. Data scientists report that their high satisfaction is based
more on the intellectual stimulation of the work than the big
paychecks. Data scientists are smart people who thrive on
challenge in their daily activities. Their work involves a wide
variety of tasks and a constant flow of interesting problems to
work on. This is a versatile field, too, with opportunities to work
on projects that touch many different industries.
Un-Attractive Features
Data science is a field with clearbenefits, but like any career, it
also has its share of disadvantages and limitations. Consider the
sheer enormity of the field and its cross-disciplinary nature. While
there are vast opportunities and an extreme variety of projects
that offer exciting new challenges, the sheer volume of subject
matter can mean never being able to fully master a specific
function. If you try, you will find that the technologies are evolving
so quickly that the systems and software you have learned are
suddenly obsolete. Now you need to start with a whole new
system. Determining which new system is best is a challenge
unto itself that can lead to confusion and wasted time.

Is it possible to master data science? Not really. While the initial

training may only take the usual five or six years to earn a
master’s degree, it can take years more to gain proficiency. This is
a dynamic, ever-evolving field that is a mashup of many other
fields. It is nearly impossible to master each and become an
expert in all of them. At minimum, data scientists need to keep
learning and try to fill the skill-gap through online courses.

Good data science is dependent on deep domain knowledge. It is

not enough to have the technical skills to manipulate and
interpret data. That data has to specifically relate to a certain
industry. Without a solid knowledge of genetics and molecular
biology, for example, it would be impossible to work on an
analysis of genomic sequences. A data scientist with a
background in the healthcare industry can do it. However, it
would be extremely difficult for someone working in the financial
world to successfully migrate to healthcare.
The competition for jobs is growing. There is great demand, but
only for experienced professionals. It can be very hard for a new
graduate to find a data science position without getting some
experience first. To outsmart the competition, new data scientists
are finding their way into the field through the back door by taking
closely related, yet lower-level jobs such as junior developer or
statistical analyst.

The money is still great, but there are signs of salaries beginning
to flatten. It is a matter of supply and demand. As the word
spread that data science is the hottest career of the 21st century,
enrollments in college data science programs exploded. Online
learning hubs like Coursera and Fast.ai further fueled the rush to
acquire data science skills. While there is still a huge demand for
experienced professionals, there are limited entry-level jobs
available for new candidates. That discrepancy is holding down
wage growth, particularly for new careerists.
Required Education and Training
Data scientists are highly educated professionals. Nearly nine out
of 10 have at least a master’s degree and nearly half have PhDs.
Before enrolling in a graduate degree program in data science,
roughly one third of data scientists started with a bachelor’s
degree in computer science. Another 20 percent majored in math
and 16 percent studied engineering. Other common degrees are
in social sciences, physical sciences, and mathematics, statistics,
and other majors related to quantitative fields. There is a growing
number of universities offering bachelor’s degree programs in
data science. However, it is still more common for universities to
offer majors in applied computer science, business
administration, or business analytics with a specialization in data
science.

Regardless of where and how data scientists get their education,

there are some important skills that they must learn. The top five
skills employers expect to see are data visualization, artificial
intelligence, machine learning, modeling, and optimization. Some
fluency in at least one programming language is needed. The
most common are Python, R, SQL, and Scala. They should also be
adept at using the tools of the trade like Spark, BigQuery, and
Hive. Interestingly, despite the constant introduction of new and
more advanced tools, what employers want most is expertise
with Excel. Data scientists can obtain training in all of these in a
number of ways, including traditional college courses, boot
camps, and online certificate programs. A good way to hone skills
is by participating in competitions hosted by online platforms like
Kaggle.
Traditionally, the tech industry was the sole domain of data
science. That is no longer the case. Today’s data scientists work in
a variety of industries, such as financial services, logistics,
manufacturing, healthcare, security and fraud detection, and
government. As more and more businesses come to understand
that data science is a critical factor in their competitiveness, the
role will become ubiquitous. There may come a time when data
scientists can be generalists, but for now, employers are looking
for industry-specific experience. It is, therefore, vital that aspiring
data scientists balance their robust academics with on-the-job
training.

Students should research their area of interest early on and

decide what they will specialize in. They can develop the specific
skill sets for a particular industry by attending meetups,
participating in internships, accepting lower-level (temporary)
jobs, taking professional development courses, and going through
industry-specific boot camps.

Certifications

A good strategy for getting the attention of employers is to earn

certifications. There are immersive programs that lead to
certification in data science offered by MIT, Harvard, and other
top universities. Each program is offered online, making it
possible to participate from anywhere, anytime.

There is a general professional certification, the Certified Analytics

Professional (CAP), but most certifications are offered by leading
analytics software providers, such as Microsoft, Oracle, IBM, SAS,
EMC, and Apache. Each offers training and credentials for each of
the prominent tools in their product line. Two of the most
common examples are EMC Data Science Associate (EMCDSA)
and AS Certified Predictive Modeler.

Industry-specific certifications for data scientists are just

beginning to emerge. For example, the Certified Health Data
Analyst (CHDA) credential from the American Health Information
Management Association (AHIMA) is specific to the healthcare
industry.
Earnings
The government labor experts have not yet started tracking
salaries for data scientists, but industry leaders are watching
closely. Every one of them confirms that these professionals earn
substantial incomes and can expect to do so for the duration of
their careers. Currently, the average data scientist salaries offered
on the major job search sites are 80 percent higher than the
average salaries for all other job postings nationwide. The average
salary is an impressive $115,000, with a range of $85,000 to
$170,000. Those who prefer to work as independent contractors
do equally well (or better). The hourly pay range for contractors is
$75 to $100, depending on skills and project requirements.

A data scientist’s income depends on several factors including

experience, job title, industry, company size, and location. With
the right combination, a talented data scientist can earn well over
$250,000 per year with salary plus incentives.

Experience

Experience has the biggest impact on a data scientist’s earnings.

For every year in the field, these professionals earn an average of
$2,000 to $2,500 more. A recent study found that due to the high
demand, entry-level data scientists are often offered top dollar –
$95,000 right out of college. Those who stick with it for at least
nine years can see their earnings grow to $150,000 or more.

Job Title

In most salary reports, those surveyed are in a job role titled “data
scientist,” but there are almost as many who have earned
promotions or gained expertise in some specialized area. As they
move up the career ladder, they often shed the title data scientist
for something that more closely describes their role such as big
data engineer or software development expert. These are just two
of many evolved roles that earn more than the average for data
scientists.

Data science professionals who have managerial responsibilities,

such as leading team projects or interacting with company clients
typically earn the most. In general, the more managerial tasks, the
higher the salary. Experienced data scientist managers who lead
teams of at least 10 earn the most in this field. Their median cash
compensation (salary plus bonuses) is $250,000.

Data scientists do not need to become managers to earn higher

salaries. Becoming a subject matter expert can be a separate
career track that does not involve leadership. The fact is, one of
the biggest challenges for employers is finding data scientists that
thoroughly understand their industry and its business model.
Those that do, commonly known as domain experts, are a rare
commodity and their compensation reflects their value. In fact,
their annual salaries can be up to $50,000 more than a general
data scientist gets.

Industry and Company Size

Data scientists are paid well no matter who they work for, but
some industries are better than others. The top three industries
that pay the most are:

Cloud services, hosting, and CDN (content delivery network)

Search and social networking

 Banking and finance

Naturally, the highest paid data scientists work at leading tech

companies. Currently, the average salaries at some high-profile
organizations are:

Google: $153,000

Apple: $146,000

Twitter: $136,000

Facebook: $135,000

PayPal: $133,000

Airbnb: $128,000

Microsoft: $123,000

As for company size, the larger the organization, the larger the
salary. For example, in a company with at least 10,000 employees,
a data scientist would likely earn at least 20 percent more than
someone in the same role at a company with fewer than 1,000
employees.

Geographic Location

Salaries depend greatly on the region in which you live. Because

so many tech companies are headquartered on the West Coast,
earnings tend to be higher than the national average. Data
scientists on the West Coast earn about 13 percent more than
their Northeast counterparts, and even more when compared to
non-coastal states. The highest salaries are in California. The
Pacific Northwest, namely Washington and Portland, Oregon,
have fewer data scientists than California and the salaries are a
little lower. However, the region can still boast of paying the
second highest salaries in the country. States with the lowest
salaries are Florida, Maryland, and Iowa.

Nationwide, the two cities that offer the best salaries for data
scientists are Houston and San Francisco with salaries averaging
from $140,000 and $165,000 respectively. If your goal is to put the
most money possible in your pocket, Houston would be a better
bet because the cost of living is lower than the national average,
while in San Francisco it is 60 percent higher.
Opportunities
Data scientist is the number one most promising job in America
today, with professionals in the field reporting a winning
combination of high demand, high salaries, and high job
satisfaction. The government does not yet track employment in
this field, but the major job posting sites have definitely taken
notice. LinkedIn, for example, sifts millions of member profiles to
gather stats of job growth. The site (which employs its own army
of data scientists) saw a dramatic upswing of over a 50 percent
increase in job openings in the US over just one year.

The time is ripe for aspiring data scientists to enter the job
market. A look at the job boards quickly reveals thousands of job
openings. Top companies like Facebook, Twitter, Apple, and
PayPal are competing with each other to sign up available talent.

Data science is virtually future-proof. The field is expected to

continue growing exponentially and to provide unprecedented job
security for a long time coming. Experts are predicting that 163
zettabytes (163,000,000,000,000,000,000,000 bytes) of data will be
in existence by 2025, causing data science career opportunities to
shoot through the roof. This represents rapid growth in data
collection. Businesses big and small are wooing data scientists to
help them make sense of the oceans of data being generated.

The rise of machine learning and AI, the hottest technology out
there, is another factor in the rising demand for data scientists.

Fueling the hot job market is a widening supply gap that makes it
hard for employers to find qualified candidates. As the number of
job openings increases, the supply of skilled applicants is growing
at a slower pace. Data science has been hyped as the hottest job
of the 21st century for nearly a decade now and college students
have taken notice. The challenge for universities is finding quality
faculty to provide the necessary training. Most who are qualified
are out in the workforce where salaries are much higher than
those in academia. Some experts predict the supply gap will
narrow as more people graduate with computer science degrees
or get “nanodegrees” from the huge number of open online
courses. They also do not believe the gap will close anytime soon.

Employer expectations are also contributing to the supply gap.

Organizations are often searching for the “unicorn” data scientist
that combines the skills of data engineer, AI expert, and business
executive. These people are very rare in real life. That is why most
large organizations and those with experience with data scientists
utilize the team setting that pulls together domain experts, AI
specialists, data analysts, and engineers. The best approach for
job seekers dealing with smaller and less experienced employers
is to showcase some relevant experiences on their résumés that
touch on these different areas.

Data scientists are coveted professionals in the tech industry.

Companies like Apple, Salesforce, Linkedin, and Amazon have
long understood the game-changing advantages of capitalizing on
Big Data. Until recently, only large digital enterprises were willing
and able to make significant investments in data scientists. Today,
almost everyone is, from Google to the humble online retail store.

One of the hottest new areas for data scientists is healthcare,

with its massive amounts of patient records, medical research
data, and profit-hungry big pharma. Other industries aggressively
recruiting data scientists include government agencies, big-box
retailers, and the US military.
Getting Started
According to the experts, aspiring data scientists must do three
things in order to land their first job in the field:

Take the time to get some real world experience

Get ready to show what you can do

Interact with other data scientists

The first two are crucial because even though the demand for
data scientists is great, employers want to make sure they are
offering top dollar to someone who can deliver.

The third is common to all fields, but especially important in this

one. Making connections will widen your network and create
future opportunities. One more bit of advice: consider relocation.
This is a practical step since the vast majority (75 percent!) of
available jobs for data scientists are located on the East and West
Coasts.

Are you experienced?

You will graduate with many of the skills needed to become a data
scientist, but without some real-world experience you will find
employers are reluctant to give you a chance. You do not have to
wait until graduation to get this experience. In fact, you should
start no later than your third year in college. Participate in
internships, at least two or even three. These can usually be
completed during the summer and may eventually turn into full-
time job offers. Look for lower level entry positions, such as
statistical assistant or research technician. These jobs can provide
valuable training and experience that are easily transferable to
data science. Take as many certificate training classes as possible,
especially those involving analytical software programs, advanced
analytics techniques, and big data management. These are likely
to arm you with needed skills not taught in college.

Put yourself out there.

Build yourself a website and create a portfolio. That is a bare

minimum. You can accomplish both with GitHub Pages, with free
hosting. Find something employers are interested in and find a
way to do a data science project around that using real data. Your
portfolio projects should highlight your coding skills in Python, R,
and SQL. Polish that code until it is clean, readable, and elegant.
Do not be afraid to comment on your process. In fact, you should
comment a lot. Include a write-up that tells the story, whether it is
a readme or a blog post. Consider using Blogdown to write your
website in R. That takes care of two goals in one. Create a
LinkedIn profile and include links to your website and portfolio
projects.

Skip the job boards. People who are currently working in data
science jobs will tell you there are much better ways to land a job
in this field. First, make yourself visible online. You never know
what opportunities will come to you if you simply show up.
Second, start building up your network. Building relationships
with recruiters is one of the best ways to break into the field. You
can also leverage other connections like friends and classmates,
recruiting events, career fairs, and active participation in
professional association memberships. After doing this you
should consider looking at company websites and job boards.
Take the extra steps. Join meetup groups for data scientists.
These usually have a community Slack team for sharing
information online. Offline, most of these meetups start with
attendees sharing job openings at their companies. This is an
easy way to get inside information, meet people in the field, and
learn how to “talk the talk.” Newsletters are another source of real
job opportunities that you will not find posted on job boards.
There are many data science newsletters like the Daily Science
Weekly and The Daily Science Roundup. Subscribe to all of them
and get job postings emailed to you automatically.
Associations
Data Science Association
International Institute for Analytics (IIA)
Certified Analytics Professional (CAP)
American Health Information Management Association

Listed below are some trade periodicals

Data Science Weekly

Web Sites

Data Science Central

Burtch Works
Data Elixir
Hackathon
Kaggle
Python
Udacity (Nanodegrees)

Careers Research Monographs Copyright 2015 by The

Institute For Research CHICAGO