Data scientists have the analytical chops to solve problems and the curiosity to be the first to discover the problems that need to be solved in the first place. Straddling both the IT and business worlds, they identify patterns in vast sets of scattered and apparently unrelated data, drawing out meaningful insights capable of solving real-world problems.
From Pittsburgh to Harrisburg, Pennsylvania companies and government agencies are drowning in valuable data and building out teams of data scientists to make sense of it all. With vast troves of data just waiting to be spun into gold and powerhouse organizations desperate for systems and processes capable of handling the torrents of new data pouring in each day, anybody with those capabilities is a white hot commodity throughout the Keystone State right now.
Penn Medicine is just one of a growing number of Pennsylvania healthcare systems using big data to improve patient outcomes, and save lives. Their predictive analytics project, dubbed Penn Signals, relies on a homegrown data warehouse that holds more than a decade’s worth of medical records on three-million-plus patients and counting. Drawing from both historical data and new daily input, data scientists are creating predictive models using AI to cross reference hundreds of co-morbid factors to actually predict the likelihood of deadly diseases in individual patients.
It’s resulted in the development of a sepsis early warning system, which has led to a 4% reduction in mortality rates, and an algorithm for detecting patients at risk of cardiac failure, upping the early detection rate of patients trending toward heart failure by more than 20%.
If big money is your motivation in getting into data science—and the money is good, with Robert Half’s 2020 Tech Salary Guide pegging the high-end salary for data scientists in Philadelphia at more than $207,000—this field lets you do well by doing good. But you’ll need to start off by doing some work on your education. And that typically means earning a master’s degree in the field.
Preparing for a Master’s Degree in Data Science in Pennsylvania
The best preparation for pursuing or advancing a career in data science is to earn a master’s degree. However, these popular degree programs have strict admission requirements in place to ensure that students possess the technical skills required to handle the curriculum.
Admission requirements for master’s degree programs in data science often include:
- Undergraduate degree in a related discipline (or the completion of specific undergraduate prerequisites)
- Minimum undergraduate GPA
- Minimum GRE/GMAT scores in the quantitative reasoning section
- Previous experience in a quantitative discipline
- Admissions essay/interview
- Professional letters of recommendation
Undergraduate Degree and Master’s Prerequisite Courses
Undergraduate degrees are an important component when determining a candidate’s qualifications for entering a master’s degree in data science. Some of the desirable undergraduate majors for candidates interested in pursuing a data science master’s degree commonly include:
- Applied mathematics
- Statistics
- Physics
- Engineering
- Operations research
- Computer science
- Business
While some graduate schools require candidates to hold a bachelor’s degree in a quantitative discipline, others do not; instead, they may only require candidates to have completed specific undergraduate courses, such as:
- Programming languages (C++, Python, Java, R)
- Linear algebra
- Data structures
- Algorithms and algorithm analysis
Data Science Bootcamps to Pave the Path to a Master’s or for Entry-Level Work
Of course, you may have somehow gotten all the way through your undergraduate career without having taken any of those courses or earning a relevant degree. But fear not! You still have an option when it comes to building your credentials to the point where a master’s admission’s committee will be willing to look at you: enroll in a data science bootcamp.
Bootcamps cram an enormous amount of practical learning into a course that lasts only a few weeks or months. Rather than dive into the theoretical background that academic studies supply, bootcamps stress learning by doing: you get the practical training you need to actually execute data science tasks by undertaking projects that work on live data, using the most cutting-edge tools and techniques available in the business today.
It’s usually done by putting you through a series of projects in concert with your cohort, working as a team under the supervision of instructors who have years of experience in the field. You will build your skills on tools such as:
- Hadoop and NoSQL data stores
- Traditional SQL and SQL databases like SQL Server
- Python, R, and JavaScript programming
- Numpy, D3.js, and other statistical and visualization libraries
- Tableau and other off-the-shelf visualization tools
Bridge Programs and Massive Open Online Course (MOOC) Options
Some colleges and universities accept students into their data science graduate programs who may not have satisfied all of the program’s required undergraduate courses, instead allowing them to satisfy the prerequisites through bridge courses. Bridge courses allow students to satisfy the educational gaps in their undergraduate work after being admitted to the graduate program but before beginning courses. It’s typically done through summer courses that replicate the kinds of classes you would have taken in a normal quantitative program at that level.
Another option available to students with undergraduate deficiencies is massive open online courses (MOOCs). Students with an interest in pursuing graduate work in data science may use MOOCs to build knowledge in key areas before applying to a graduate program. MOOCs offer flexible scheduling options from a variety of providers. Students complete them according to their schedule and through a completely online format.
Relevant Personal and Work Experience
In addition to having completed specific undergraduate coursework, data science graduate school candidates are also often expected to have relevant work experience. Many colleges and universities look for experience in areas like:
- Basic analytic skills
- Programming
- Information visualization
A number of Pennsylvania firms employ entry-level data scientists, allowing those with undergraduate degrees in a quantitative major to begin building their resumes.
For example, GlaxoSmithKline in Upper Providence employs bachelor’s level clinical data scientists to help manage data-intensive clinical studies. Likewise, JUNO Search Partners in Philadelphia employs entry-level data scientists with engineering, math, or related bachelor’s degrees for their Strategic Operations team.
Preparing for Success on the GRE/GMAT
Many graduate schools require candidates to submit GRE or GMAT exam scores when applying to a master’s degree in data science. Naturally, they consider the quantitative sections of these exams, often looking for candidates who scored in the 85th percentile or higher. Preparation for these exams allows students to improve their chances of hitting that mark:
The Quantitative Reasoning Section of the GRE exam assesses a candidate’s skills in four areas:
- Arithmetic
- Algebra
- Geometry
- Data analysis
Individuals may review the Math Review document, which provides a detailed overview of the exam’s quantitative section.
The quantitative reasoning section of the GMAT measures a candidate’s ability to analyze data and draw conclusions using reasoning skills. With 37 questions, there aren’t any opportunities to blow a few and still hope to score well, so preparation is particularly important with this exam.
Study tools, sample questions, and videos help test takers prepare for the exam.
Earning a Master’s Degree in Data Science in Pennsylvania
Online: A number of institutions offer data science master degree programs in an online format. Students of these programs complete all required curriculum remotely for the best in flexibility and convenience. Most programs require students to visit the campus on just one or two occasions to complete an immersion experience, which allows students to meet their peers, their professors, and to engage in networking activities. Online programs are designed to rival their campus-based counterparts with videos and live stream lectures, interactive case studies, and self-paced lectures.
Part-time/accelerated: Some students find that completing the program in a part-time or accelerated format better fits their personal or professional goals. Full-time programs in data science take between 18 and 24 months to complete, whereas accelerated programs take about 12 months and part-time programs take about 32 months.
Depending on the institution, these programs go by titles like:
- Master of Computational Data Science
- Master of Information and Data Science (MIDS)
- Master of Science in Statistics: Data Science
- Master of Science in Data Science
Curriculum and Core Coursework
A master’s degree in data science consists of about 30 credits of coursework designed to prepare tomorrow’s data science leaders in area such as:
- Applied machine learning
- Data mining
- Experimental statistics
- File organization and data management
- Statistical sampling
Program Competencies and Objectives
Data science master degree programs prepare students to utilize the latest tools and analytical methods and to interpret and communicate their findings. A multidisciplinary curriculum draws on the social sciences, computer science, statistics, management, and the law.
Graduates of these programs are able to identify patterns in—and extract insights from—complex datasets. Their expertise lies in their ability to utilize the latest statistical and computational methods to make predictions, communicate their findings, and appreciate the ethical and legal considerations associated with working with real-world data.
Career Opportunities for Data Scientists in Pennsylvania with Advanced Degrees
Data science is a field that is growing explosively everywhere in the country, and that most certainly includes Pennsylvania, which is ranked 10th in the nation for technology employment by the 2020 DICE Tech Jobs Report. That’s up one slot since 2019.
Data scientists play a major role in that expansion. The report also puts data engineers and data scientists as two of the top three fastest growing tech occupations in the country for 2020, with engineer roles expanding by a mind-blowing 50 percent year over year.
The value of data scientists in Pennsylvania is reflected in the many job postings for these technically savvy professionals. Although the following job descriptions do not provide a guarantee or assurance of employment, they do highlight the abundance of job opportunities for today’s master’s prepared data scientist:
Data Scientist, Siemens: Warrendale, PA
Responsibilities:
- Collaborate with data engineers to ensure availability of data
- Define and drive requirements for adapters to enterprise data source
- Create and implement algorithms to process healthcare data for descriptive and predictive analytics
- Translate proof-of-concept analyses into scalable pipelines
Requirements:
- Skills in modern digital product design and delivery
- Design thinking and customer experience mapping
- Master’s degree in a relevant field, clinical experience a plus
- Strong working knowledge of medical imaging SW engineering
- Expert knowledge of data mining algorithms including decision trees, probability networks, associate rules, clustering, and neural networks
Data Scientist, Hershey: Hershey, PA
Responsibilities:
- Root cause analyses
- Forecasting
- Growth opportunity identification via analytical modeling
Requirements:
- Master’s degree with a STEM focus strongly preferred (majors could include data science, statistics, analytics, or a closely related field)
- At least eight years of experience in statistics, machine learning, information retrieval, or graph analyses
- At least four years of experience with small and large scale data mining technologies
- At least four years of experience in predictive analytics or statistical modeling, Bayesian modeling, Time Series modeling, Panel modeling, Marketing Mix modeling
- At least four years of experience with large data sets and big data/distributed computed platforms
Data Scientist, Magento: Philadelphia, PA
Responsibilities:
- Develop data models and analysis methods
- Using statistical techniques to design scalable solutions to business problems
- Analyze and extract insights from large, unstructured datasets
- Author whitepapers and case studies
Requirements:
- At least two years of professional experience, specifically in data modeling, statistical interference, and analysis
- Track record of learning new skills and putting them to use immediately
- Ability to communicate complex quantitative analysis and analytic approaches in a clear, precise, and actionable manner