Columbia's MS in Data Science is one of the most sought-after data science master's degrees in the US — run jointly by the Fu Foundation School of Engineering and Applied Science (SEAS) and Columbia's Data Science Institute, and sitting in the middle of Manhattan, a short subway ride from the quant-finance and tech firms that recruit heavily from the program. For applicants weighing Columbia against Berkeley's MIDS, NYU's MSDS, or Cornell Tech, the pull is the combination of Columbia's brand, SEAS's rigorous technical core, and direct access to New York's finance and media industry. This guide walks through exactly what it costs, what it takes to get in, and what happens after you graduate.
Acceptance Rate: What We Actually Know
Columbia does not publish a program-specific acceptance rate for the MS in Data Science, and this dataset's own graduate acceptance-rate field for Columbia is left null for the same reason — no official university-wide graduate figure exists either, since each of Columbia's 19 schools reviews applications independently. Be skeptical of any site that states a precise "MS Data Science acceptance rate" to one decimal place; none of that is coming from Columbia itself.
What we can say responsibly:
- Columbia's undergraduate acceptance rate (4.9% for the Class of 2029) is frequently misused as a stand-in for graduate selectivity. It is not a useful proxy — master's programs at SEAS admit at meaningfully higher rates than the undergraduate college, because they are larger, cohort-based, and partly self-funding.
- Third-party admissions consultants and applicant forums have, over recent cycles, estimated the MS Data Science admit rate somewhere in the 15–20% range, but this is not an official Columbia figure and should be treated as a rough, unverified indicator rather than a hard number.
- The program is explicitly capped by cohort size and is more selective than most other SEAS master's programs, because it draws from an unusually large, high-quality global applicant pool (CS, statistics, math, and engineering graduates worldwide).
Bottom line: treat this as a highly competitive but not ultra-low-single-digit program — closer in spirit to a strong specialized graduate program than to Columbia's undergraduate admissions.
Tuition and Fees
Columbia's SEAS master's programs, including the MS in Data Science, are billed per credit rather than a single flat annual rate, so your actual cost depends on how many credits you take per term and how quickly you complete the 36-credit degree (typically 3–4 semesters).
| Cost Item | Amount (USD) |
|---|---|
| Representative graduate tuition (GSAS reference rate, per year) | $76,212 |
| Living expenses (per year, Columbia's published estimate) | $31,773 |
| Total cost of attendance (per year, reference estimate) | $107,985 |
| Application fee | $125 |
These figures are Columbia's published Graduate School of Arts and Sciences reference cost of attendance, used here as a representative benchmark since SEAS does not publish one all-in annual sticker price for a per-credit program. Budget for the full 36-credit degree to land somewhere in the $85,000–$100,000 tuition-only range across program length, before living costs — confirm the current per-credit rate directly on the SEAS or Data Science Institute admissions page before committing, since per-credit rates are revised annually.
Admission Requirements
| Requirement | Detail |
|---|---|
| Minimum GPA | No published hard minimum; strong quantitative GPA expected |
| GRE | Not required (SEAS has dropped the GRE requirement for most master's programs) |
| Prior coursework | Programming (Python or Java), probability & statistics, linear algebra, data structures |
| Letters of recommendation | Typically 2–3, ideally from academic or research supervisors who can speak to quantitative ability |
| Statement of purpose | Required — should be specific about which data science subfield you're targeting |
| Resume/CV | Required |
| TOEFL/IELTS | Required unless exempted (see below) |
| Application fee | $125 |
Columbia doesn't publish a required undergraduate major, but the overwhelming majority of admitted students come from computer science, statistics, mathematics, electrical engineering, or a closely related quantitative field. Applicants from non-CS backgrounds are expected to show demonstrated programming competence — a coding bootcamp certificate alone is generally not viewed as sufficient preparation.
How to Get In
-
Lead with a coherent technical narrative, not a broad one. Columbia's Data Science Institute wants applicants who can explain specifically why they want data science (as opposed to a general CS master's) and what subfield — NLP, ML systems, computational biology, quantitative finance — they're aiming at. A vague "I love data" statement of purpose is the single most common weakness reviewers cite.
-
Show real, verifiable programming and math depth. Because GRE is no longer required, your transcript and project history carry more weight than they used to. A GitHub portfolio, a research paper, or a documented independent project (even a solid Kaggle track record) meaningfully strengthens a borderline profile.
-
Get recommendation letters from people who can speak to technical rigor. A letter from a manager who liked your "attitude" is far weaker than one from a professor or technical lead who can describe specific quantitative work you did and compare you to peers.
-
Target New York-relevant experience if you have it. Columbia's placement strength is heavily tied to NYC's finance and tech ecosystem. If you have relevant fintech, quant, or applied ML experience, make that connection explicit in your application — it maps directly onto how Columbia pitches the program to employers.
-
Apply early relative to the deadline window. GSAS/SEAS deadlines for Fall entry typically fall between December and January; rolling elements of review mean applying near the opening of the cycle, rather than at the final deadline, can help with cohort planning and scholarship consideration where available.
-
Don't neglect the English test if you need one. SEAS reviewers use the English proficiency score as a threshold check, not a differentiator — but falling short of it can stall an otherwise strong file procedurally. Clear it early so it isn't a last-minute bottleneck.
English Proficiency Requirement (IELTS/TOEFL)
Columbia's SEAS graduate programs follow the same English-proficiency policy as the university's Graduate School of Arts and Sciences: IELTS Academic 7.5 or TOEFL iBT 100 (legacy scale; 5.5 on ETS's new 1–6 CEFR-aligned scale for tests taken on or after January 21, 2026), with waivers available if your prior degree was taught entirely in English. Scores are valid for two years, and Columbia does not offer conditional admission for applicants who fall short.
For the full breakdown of section-by-section expectations, waiver conditions, and how Columbia's scale compares to other Ivy League schools, see Gabble's dedicated guides: Columbia IELTS requirements and Columbia TOEFL requirements.
Career Outcomes
Columbia does not publish a program-specific employment rate or average starting salary for the MS in Data Science (the field is left blank even in Columbia's own published graduate outcomes data), but the university-wide graduate recruiter list gives a strong directional signal:
Top graduate recruiters at Columbia: McKinsey & Company, Goldman Sachs, Boston Consulting Group, J.P. Morgan, Bain & Company, and international organizations including the United Nations.
For Data Science Institute graduates specifically, the most consistently reported outcome categories (based on program alumni networks and LinkedIn placement patterns rather than an official Columbia figure) are:
- Quantitative and data roles at investment banks and hedge funds in New York (a distinct advantage of Columbia's location)
- Applied ML/data science roles at major tech companies (Google, Meta, Amazon, and similar)
- Data science and analytics roles at consulting firms, particularly for graduates who combine the degree with prior industry experience
Graduates commonly report total compensation in the $120,000–$170,000 range for New York-based tech and finance roles at the entry level, though this reflects commonly cited market ranges for the degree and location rather than an official Columbia-published statistic — treat it as directional, not guaranteed.
FAQ
Does Columbia publish an official acceptance rate for the MS in Data Science? No. Columbia does not release program-specific graduate acceptance rates, and its own graduate-wide acceptance rate is also unpublished. Any precise percentage you see cited online is a third-party estimate, not an official figure.
Is the GRE required for Columbia's MS in Data Science? No, SEAS has dropped the GRE requirement for most of its master's programs, including this one, though a strong GRE score can still be submitted optionally to strengthen a borderline quantitative profile.
What undergraduate background do I need? A quantitative degree — computer science, statistics, mathematics, or engineering are the most common — with demonstrated programming ability (Python or Java) and coursework in probability, statistics, and linear algebra.
How long is the program? The degree requires 36 credits, typically completed over 3–4 semesters (roughly 1.5–2 years) depending on course load per term.
What IELTS or TOEFL score do I need? IELTS Academic 7.5 or TOEFL iBT 100 (legacy scale), following Columbia's Graduate School policy. See our full Columbia IELTS and TOEFL requirement guides for section details and waiver conditions.
Is Columbia's MS in Data Science better for finance or tech careers? It's genuinely strong for both, but the New York location gives it a distinct edge for quant finance and fintech roles compared to West Coast-heavy programs like Berkeley's MIDS — while still placing well into FAANG-adjacent tech roles.
Prepare for IELTS with Gabble or prepare for TOEFL — clearing Columbia's 7.5 IELTS / 100 TOEFL threshold early means one less variable to manage while you build the technical portfolio that actually decides admission.