What Admissions Teams Seek in Top Data Science Master's Students
Tufts University is among the most selective universities in the nation. While this exclusivity energizes some prospective students, others find it daunting—particularly when they aren't sure they meet the admissions criteria for technical programs like the Tufts School of Engineering's online Master of Science in Data Science (MSDS).
Overcoming intimidation is, in large part, a matter of remembering that there's nothing adversarial about the admissions process. Tufts wants students to thrive both in and out of the classroom, academically, professionally, and personally. The university's MSDS standards are in place not as an exclusionary measure, but to ensure that every student admitted to the program has what it takes to join the next generation of leading data scientists.
First and foremost, admissions teams want to see MSDS applicants who meet the essential prerequisites of the program and submit each of the required application materials. Strong quantitative skills and a demonstrable aptitude for advanced mathematics are critical. Attention to detail, discipline-specific foundational knowledge, and soft skills are also important. From there, admissions teams look for evidence that applicants will thrive in a data science master's program. This evidence can take many forms: a high undergraduate GPA within math coursework, a track record of leveraging analytical skills, professional experience working with data, and a relevant academic background.
Taking a critical inventory of whether you meet all the Tufts School of Engineering master's in data science prerequisites is the best way to improve your odds of admission to this rigorous and highly technical program.
Who typically earns an MSDS from Tufts?
One way to learn more about what admissions teams seek in MSDS applicants is to look at the program's student body. The Department of Computer Science and the Department of Electrical and Computer Engineering jointly administer Tufts' online Master of Science in Data Science program, with courses also available from the Department of Mathematics. Incoming students are excited by numbers and analytics. They have bachelor's degrees in STEM fields and programming skills—but see themselves as analysts first—and enroll in Tufts' MSDS program to gain specific skills and credentials necessary to advance in data analysis and data-intensive science.
Some students in the online data science master's program are senior software engineers, program managers, or in technology-focused leadership roles, familiar with the topics in the interdisciplinary MSDS curriculum (e.g., data analytics, probabilistic systems analysis, algorithms, machine learning, data mining, and Big Data). Others see earning a data science master's degree as part of a more extensive research-focused academic journey that will culminate in a top-rated PhD program. These highly independent problem-solvers want to use data to make a difference in the world—and the caliber of their accomplishments can be unnerving for those who are unsure they have what it takes to succeed in a data science master's degree program.
If that's you, keep in mind that data science is a collaborative discipline. Students in top MSDS programs benefit from the achievements of their peers as they tackle real-world challenges together.
What are the basic requirements for admission to the Tufts MSDS program?
The basic master's in data science requirements include a bachelor's degree—preferably in a computational or STEM discipline with a strong math background and coursework in calculus, linear algebra, and statistics—and a competitive GPA. Applicants with undergraduate degrees conferred by accredited U.S. or Canadian colleges and universities no longer have to submit GRE scores but can if they feel that will make their applications stronger. International students who earn undergraduate degrees in other parts of the world still submit GRE scores with the MSDS application.
Applicants submit transcripts from each educational institution attended, which admissions teams use to measure academic performance and intellectual curiosity. GPA is just one metric. Admissions officers also consider how much upper-level coursework applicants completed in their undergraduate years and whether their grades show patterns of improvement. They scan transcripts for academic honors, independent work, and the content of each applicant's senior thesis.
Beyond undergraduate transcripts, Tufts' master's in data science application requirements include:
- Three letters of recommendation
- A résumé highlighting work experience plus publications and other related activities
- TOEFL, IELTS, or Duolingo test scores for Non-Native Speakers of English
- A personal statement describing why applicants want to study in the data science graduate program at Tufts
What makes an MSDS application strong?
The strongest master's in data science applications go above and beyond the basic requirements to tell a compelling story. Differentiating your application is often more important than having the highest GPA or a long list of accolades. Most MSDS applicants will have similar academic and professional backgrounds. Your personal statement, résumé, and letters of recommendation show the admissions team what makes you an excellent candidate for this degree and why you're a good fit for Tufts.
Personal statements should clarify that you are excited about the program and likely to enroll if accepted. Two points to highlight are the classes in the master's in data science program that inspire you and how your interests align with the research projects or specializations of professors and faculty. Do you have experience working with data sets, data management, or data visualization? A background that aligns with the MSDS degree requirements will make your application stronger. On the flip side, if you hope to use your MSDS to transition from a data-adjacent career (like healthcare informatics, for example) to data science, the personal statement is your opportunity to explain this.
Your résumé should concisely outline your education, relevant work history, internships, publications, and any extracurricular activities related to data science, computer science, information systems, or analytics. Don't feel like you have to list every single accomplishment. A one-page résumé highlighting strengths and achievements specific to data science is almost always more compelling than a longer, less-focused résumé.
Letters of recommendation should come from supervisors or colleagues who can speak to your abilities, work ethic, and impact. People who know you and your work well—even if they aren't data scientists themselves—write the most powerful letters of recommendation. Prestige isn't important. A recommendation from a high-profile data science or computer science expert only carries weight if you've worked with or studied under them.
Every component of your application should highlight your quantitative abilities. Proof of strong mathematical skills—either professionally or academically—is critical. One way to accomplish this is with undergraduate transcripts showing high scores in mathematics courses. Letters of recommendation from people who can speak to your quant skills will also go a long way.
What are the requirements to complete the MSDS program?
The School of Engineering delivers the 10-course, 32-credit hour MSDS for part-time students 100 percent online over less than two years. The same faculty members behind the university's full-time, on-campus program develop (and teach) the core courses in the MSDS curriculum. The below links contain sample syllabi from recent offerings of the courses at Tufts; versions in the online program will be similar:
- Principles of Data Science in Python
- Probabilistic Systems Analysis
- Introduction to Machine Learning
- Reinforcement Learning
- Stochastic Processes, Detection, and Estimation
- Aspects of Data Analysis
- Big Data
- Artificial Intelligence
Students also complete a two-course capstone project, proposing a substantial data science project, carrying it through to completion, and sharing their results with faculty and peers in a professional presentation.
The financial investment required to earn an MSDS at Tufts—about $54,000—is not insubstantial, but tuition costs are competitive, and the investment pays off. When you enroll in a school like Tufts, the ROI of a data science master's earned online is identical to that of one completed on campus. Data scientists with advanced degrees—which is almost all of them—earn just over $100,000 per year. Data science professionals in managerial and executive roles can earn $200,000 or more, provided they have the right technical skills and soft skills.
What technical skills do you need to succeed in the MSDS program?
The requirements that data science master's students must meet to succeed are challenging to enumerate in a shortlist of prerequisites. Admissions teams look for MSDS applicants with programming and statistical modeling skills, but the nature of those skills differs from applicant to applicant. Python, for instance, is the programming language of choice in much of the data science world, but MSDS candidates may also enter the program with R, Scala, Clojure, Java, C++, and Octave skills. Students learn to apply their programming skills to real-world reinforcement learning and probabilistic systems challenges throughout the program.
Admissions teams also look for applicants with the foundational computer science aptitude necessary to learn the machine learning skills used to build predictive models in engineering, manufacturing, business, and operations research environments. Some incoming students have relatively simple linear and logistic regression modeling skills, while others enter the MSDS program with more advanced abilities. Even the most basic machine learning skills offer a window into which technologies to apply to what situations to get timely and accurate results. Tufts' online MSDS builds on this knowledge in COMP 135 Introduction to Machine Learning, which teaches supervised learning, unsupervised learning, reinforcement learning, and knowledge extraction from large databases.
Mathematics skills are also critical. Many people don't realize just how math-intensive most data science master's programs are, assuming data science is more like computer science than statistics when the inverse is true. Much of data science involves leveraging math expertise to develop complex statistical models. Data scientists need to be comfortable with calculus, linear algebra, probability, discrete mathematics, graph theory, and information theory. Tufts’ MSDS MATH 166 Statistics course reviews probability and covers estimation, parametric inference, Bayesian inference, regression (linear and nonlinear), tests for independence, and the role of regularization in statistics.
Don't erroneously assume you won't succeed in the Tufts Online Master of Science in Data Science program if you aren't strong in all the areas outlined above. Where students start in the MSDS program is based to some degree on their backgrounds. Those who enroll with little prior experience in Python programming, for instance, can elect to take COMP 205 Principles of Data Science in Python. Students with no previous linear algebra experience can take COMP 206 Linear Algebra for Data Science. They can also fulfill those and other knowledge requirements related to data science hard skills through self-study before entering the program.
What soft skills does it take to be successful in the MSDS program?
Tech-savvy applicants may have the edge over those with less technical backgrounds, but that isn't all it takes to succeed in Tufts' data science program. Soft skills are important. Application teams look for prospective MSDS students who communicate effectively, actively participate on teams, and deal with uncertainty—or who have traits that suggest they can refine those skills in a graduate program. As you prepare your application, don't forget that top data science master's students typically demonstrate:
Strong communication skills
As lecturer J Singh explains in the DS 205 Principles of Data Science in Python syllabus, "Doing data analysis is the easy part of Data Science. The hard part is showing others how you got there and persuading them to take action."
Data scientists need well-developed written, verbal, listening, persuasion, and presentation skills. Success in data science requires understanding others' needs, helping them supply you with the information you need, and sharing your findings with stakeholders who have no analytics or statistics experience. Professor Singh goes on to say that his introductory course emphasizes "not just analysis, but also communicating results to others in a manner that allows reproduction of results and critical analysis."
Critical thinking abilities
Problem-solving and troubleshooting are essential elements of data science, and both require critical thinking skills. Problems in this discipline tend to be open-ended and unstructured, which means it's up to data scientists to answer questions—and determine what, exactly, those questions should be.
Tufts online MSDS courses like COMP 135 give students machine learning challenges in the context of real-world, open-ended use cases designed to foster creativity and critical thinking. As Associate Teaching Professor Ming Chow, quoting Jeffrey J. Selingo, writes in his undergraduate capstone course syllabus, "Generally speaking in the real world, you will not be given detailed specifications as you have been in most classes. You will also not be given details on how you will be graded on a project because there is no such thing as grades in the real world. The professional world 'is unstructured, with competing priorities and decisions that need to be made on the fly... The workplace is more of a mash-up of activities with no scheduled end.'"
Data scientists are skilled decision-makers who give clear feedback at all levels, so the School of Engineering admissions team looks for signs of leadership potential in MSDS applicants. Master's program graduates should know how to help executives and other stakeholders drive strategy, making them de facto leaders. However, they can only lead if they feel comfortable with the latest techniques and infrastructures in data science, which is why courses like COMP 119 Big Data are vital to success.
Positivity and an ethical bent
Confidence, cooperation, friendliness, honesty, and patience are important in data science, where students and professionals alike tackle open-ended and occasionally frustrating challenges. The COMP 131 Artificial Intelligence syllabus makes it plain: behavior and attitude matter more than most students realize, and "integrity and honesty are equally important qualities of any future academic, scientist, or engineer."
Sometimes code doesn't work. Sometimes a formula delivers spurious conclusions. And sometimes, stakeholders don't want to hear what the data has to say. None of that should interfere with a data scientist's diligence, dedication, or attention to detail. The work in Tufts' MSDS program is challenging in much the same way data science work is in the real world, so students get a taste of what's to come and learn to insulate themselves against discouragement. They also learn about the various ethical and social considerations data scientists must know to avoid discouraging or invalidating others and their experiences.
Teamwork is about more than shared goals. The best data scientists work with people across teams and departments, accept feedback and criticism with grace, and let others take the reins when it makes sense to do so. Incoming MSDS candidates often have academic and professional experience demonstrating their ability to work collaboratively, and the group work in the program further nurtures their teamwork skills. Students enhance their interpersonal and networking skills by exceeding classroom participation goals, making a point of dropping into office hours, and being opportunistic about taking advantage of opportunities to work with others.
Do you have what it takes to succeed?
The hard skills and soft skills used by data scientists evolve alongside technology—and just as rapidly. Tufts' online MSDS program is an "agile degree program" designed to change quickly in response to developments and innovation. In other words, the soft and hard skills admissions teams look for today may not be the exact skills they look for in the future.
Nonetheless, the best way to overcome application anxiety is to look critically at how your skills and experience align with the content of the data science master's program as it exists right now. You may not have all the competencies outlined in the master's in data science requirements above, and that's okay. The goal of any graduate-level program is, after all, to help students identify gaps in their knowledge and develop skills they don't possess.
Another goal of this particular master's program is to prepare students for employment in a field where 90 percent of working professionals have advanced degrees and nearly half of all job postings require that applicants have master's degrees in addition to experience. Employers want evidence that new hires can tackle real-world challenges and make a difference with data. That's why Tufts built a student experience that gives every MSDS candidate the hard and soft skills required to succeed not only in the program itself but also in their working lives.