Data science is one of the strongest-looking majors on paper.
The Bureau of Labor Statistics reports a $120,230 median annual wage for data scientists in May 2025 and projects employment in the occupation to grow 35% from 2025 to 2035.
Those are exceptional numbers.
They are also not a promise that every new data science graduate walks into a $120,000 job.
The degree itself is relatively new, programs vary enormously, and many employers still hire data scientists from computer science, statistics, mathematics, engineering and graduate programs.
So the answer is: data science can be a very good major, but the curriculum matters more than the label.
The job market is a real reason to take data science seriously
BLS expects demand for data scientists to grow as organizations use larger datasets, machine learning and analytical models in more decisions.
The occupation itself is not a fad.
U.S. Bureau of Labor Statistics: Data Scientists
But degree data is the better measure for evaluating college.
We show reported earnings for data science programs where College Scorecard has sufficient data, along with how many programs are suppressed or too new to have mature outcomes.
That missing-data rate is itself important.
New majors create a data problem
Computer science has decades of established programs.
Data science bachelor's degrees expanded much more recently.
That means:
- some programs have small cohorts;
- some have limited federal earnings history;
- program names differ;
- CIP classifications can change or be used inconsistently;
- graduate outcomes may reflect only early cohorts.
We do not punish a good new program simply because federal earnings are not reportable yet.
It should label it:
Federal program earnings not yet reportable
Then show institution-level context and curriculum information separately.
Data science vs. computer science
Computer science is broader in software and computing fundamentals.
Data science usually combines statistics, programming, data management and modeling.
Computer science may be the safer choice if you want maximum flexibility across software engineering, systems, AI and data.
Data science may be more direct if you already know you enjoy working with data, statistics and models.
Internal link: Is Computer Science a Good Major?
Do not decide from salary alone.
Open the curriculum.
Data science vs. statistics
Statistics gives deeper grounding in inference, probability and mathematical modeling.
Data science programs often add more programming, databases, machine learning and applied computing.
A strong data science degree should not be "statistics lite plus Python."
A strong program needs enough depth in both.
Useful questions:
- How much calculus and linear algebra are required?
- Is probability required?
- Are statistical inference and experimental design covered?
- How much programming is required?
- Do students learn SQL and data systems?
- Is machine learning taught with mathematical depth?
- Is there a serious capstone?
The answers tell you more than the program's marketing language.
The biggest program risk is shallow curriculum
Data science is marketable.
That makes it easy for colleges to launch.
A weak program can string together introductory courses from statistics, business analytics and computer science without giving students enough depth in any one area.
We do not attempt to auto-score curriculum quality unless the source data is reliable and the methodology is transparent.
Instead, provide a checklist.
A student should be able to audit a program in ten minutes.
Experience matters almost as much as coursework
Employers want evidence that you can work with messy data.
A strong student portfolio might include:
- a reproducible analysis;
- SQL work;
- a predictive model;
- data visualization;
- a team capstone;
- an internship;
- a project using real-world data.
Kaggle competitions can be useful.
A serious internship is usually more useful.
Look for colleges where students can work with faculty labs, companies, hospitals or public agencies on real datasets.
Graduate school is not always necessary
BLS lists a bachelor's degree as the typical entry-level education for data scientists while noting that some employers require or prefer a master's or doctoral degree.
That nuance matters.
A strong bachelor's degree can be enough for many analytical roles.
Highly specialized research and machine-learning positions may prefer advanced education.
Do not automatically add a master's degree because the field sounds technical.
Work experience can be more valuable.
Cost matters because there are many routes into the field
Data science competes with:
- computer science;
- statistics;
- mathematics;
- economics;
- engineering;
- information systems;
- graduate certificates;
- employer training.
That means an expensive data science degree has to justify itself.
If a public university offers a technically rigorous data science program at half the net cost of a private competitor with similar reported outcomes, that is a meaningful payoff difference.
The highest BLS salary is not your starting salary
Repeat this because search results often blur it.
The $120,230 BLS figure is the median for people employed as data scientists across experience levels.
It is not the median starting salary of data science bachelor's graduates.
We put the occupation number in a clearly labeled card:
Occupation context: not graduate starting salary
Then lead with field-of-study earnings for the degree.
When data science is probably worth it
The major looks strongest when:
- the curriculum is mathematically and technically serious;
- you enjoy both statistics and programming;
- the price is reasonable;
- you can build a portfolio;
- internship access is strong;
- the program has enough depth to keep options open beyond one job title.
When to think harder
Be cautious if:
- the curriculum is mostly business analytics with little computing;
- the degree is very expensive;
- the school markets the BLS median as an expected graduate salary;
- you dislike math;
- you would rather build software than analyze data;
- computer science or statistics offers a stronger program at the same college.
How to compare two data science programs
Data science needs a more technical comparison than most majors.
| Program A | Program B | |
|---|---|---|
| Net price | Not available | Not available |
| Reported program earnings | Not available | Not available |
| Federal debt | Not available | Not available |
| Calculus / linear algebra | Not available | Not available |
| Probability / inference | Not available | Not available |
| Programming depth | Not available | Not available |
| Databases / SQL | Not available | Not available |
| Machine learning | Not available | Not available |
| Capstone / internship | Not available | Not available |
If the cheaper program is stronger on fundamentals, that is a major signal.
If the more expensive program has substantially better employer access and stronger reportable outcomes, that deserves weight too.
Avoid using "AI" as a quality signal
A program can add "AI" to several course names without teaching students much more.
Students should look for:
- linear algebra;
- probability;
- optimization;
- model evaluation;
- algorithms;
- data systems;
- software practices.
Those fundamentals age better than the current names of popular models and tools.
We do not reward buzzword density.
Data science can lead to jobs that are not called data scientist
This is important.
Graduates may work as:
- data analysts;
- business intelligence analysts;
- analytics engineers;
- machine-learning analysts;
- product analysts;
- research analysts;
- junior data scientists.
That means judging the degree only by the BLS occupation "Data Scientists" is too narrow.
The College Scorecard field-of-study outcome is valuable precisely because it follows degree holders across occupations.
Portfolio quality is hard to standardize
A program may advertise capstones and portfolios.
DegreeVerdict cannot easily score them nationally without subjective judgment.
Do not fabricate a portfolio-quality metric.
Instead, give readers a checklist:
- Is the capstone public?
- Do students use real data?
- Are projects individual or team-based?
- Is code version-controlled?
- Are employers involved?
- Can you see examples from recent graduates?
A few minutes on the department site can reveal whether the "capstone" is meaningful.
Accreditation and program designations need careful treatment
Some computing programs have recognized accreditation or government/industry designations.
Those can be useful context.
They should not be turned automatically into a quality ranking.
Accreditation may signal that a program meets defined standards. It does not tell you what the program costs, what graduates earn or whether the curriculum fits your goals.
If DegreeVerdict adds these fields later, show them as verifiable program attributes alongside outcomes rather than awarding unexplained bonus points.
The DegreeVerdict verdict
Data science is a strong major when the program is strong.
The labor-market case is unusually good. The risk is that the degree name has grown faster than the quality of some programs, and faster than the federal outcome data can fully measure.
Look for rigorous fundamentals, real projects and sensible cost.
Compare data science programs by earnings, debt and cost →
Keep exploring on DegreeVerdict
- Is Computer Science a Good Major?
- Best Degrees for Jobs
- Highest-Paying Bachelor's Degrees
- What Degree Should I Get?
- Is Cybersecurity Worth It?
Sources and further reading
- U.S. Department of Education: College Scorecard Field of Study technical documentation
- U.S. Bureau of Labor Statistics: Data Scientists
- U.S. Census Bureau: American Community Survey PUMS documentation
About the data
We use College Scorecard field-of-study outcomes where reportable and disclose missing/suppressed program data. The BLS Data Scientists profile provides current occupation context, including the May 2025 median wage and 2025–35 projection, but must not be labeled as the starting salary of data science majors.