Is a PhD Worth It? How to Decide for Your Field

A PhD is worth it when the degree is required or clearly useful for the research work you want, and when you can name the cost in years and lost income before you start. It is usually not worth it when you want the title more than the work, or when a master's, a staff job, or direct employment gets you to the same place faster.

What is a PhD actually for?

A PhD is training in how to produce research. You learn to define a question, find what is known, design a study, handle evidence, and defend a conclusion. That training fits work where research judgment is central: academic research, industry R&D, some government labs, and roles that ask for doctoral training. It is a poor default if you mainly want a credential. Check the roles you want and see whether they ask for a PhD or whether they ask for a master's, a license, or experience. The guide on whether you need a PhD to be a scientist can help you sort that out.

A PhD also has a narrow purpose. It does not guarantee a faculty job, a high salary, or a particular title. It gives you a supervised research apprenticeship and a credential that some employers require. If the work you want does not use that training, the degree can cost more than it returns.

How long does a PhD take, and what do you give up?

The federal Survey of Earned Doctorates for 2025 gives a useful baseline. For science and engineering, the median research doctorate recipient took 8.0 years from a bachelor's degree, 6.8 years from starting graduate school, and 5.7 years from starting the doctoral program. Field medians vary. Physical sciences took 6.8 years from a bachelor's degree; health sciences took 11.0. Biological and biomedical sciences took 8.0 years from a bachelor's degree, and computer and information sciences took 8.1. The guide to how long a PhD takes can help you compare programs without treating one number as a rule.

Forgone income is not a single number in this material. Your stipend is your income while you train, so compare the written offer with the pay you would give up. The NSF Graduate Research Fellowship Program pays a $37,000 stipend for each of three years of support. The NIH Kirschstein-NRSA predoctoral stipend is $29,364 a year. Those figures are not a salary, and they do not include fees, health coverage, or local costs unless your offer says so. Use the PhD stipend guide to read an offer line by line.

The time cost is not only the years in the program. It is also the work you turn down, the savings you delay, and the location you accept for training. If you are changing careers, add the cost of catching up in a new field. If you are already earning a salary, compare the stipend with your current pay and your current obligations. If the stipend does not cover your costs without heavy borrowing, that is a serious warning.

What can you earn after a PhD?

The same federal survey reports expected salaries for new doctorate recipients who had a US postgraduation plan and a reported salary. For science and engineering, the median expected basic annual salary was $75,000 overall, $109,000 for employment, and $65,000 for postdoctoral study. For all fields it was $75,000 overall, $100,000 for employment, and $65,000 for postdoctoral study. Those are expected salaries at graduation, not lifetime earnings.

Field matters. For computer and information sciences, the median expected salary was $150,000 overall, $170,000 for employment, and $75,000 for postdoctoral study. For biological and biomedical sciences, it was $66,737 overall, $95,000 for employment, and $64,000 for postdoctoral study. For physical sciences, it was $76,000 overall, $115,000 for employment, and $67,600 for postdoctoral study. Health sciences was $80,000 overall, $96,000 for employment, and $65,000 for postdoctoral study. Geosciences, atmospheric, and ocean sciences was $70,000 overall, $85,500 for employment, and $68,000 for postdoctoral study.

A postdoc is common in some fields and rare in others. Among science and engineering doctorate recipients in 2025, 59.8% reported a definite postgraduation commitment, and 42.5% of those with a commitment went to a US postdoc. In biological and biomedical sciences, 65.7% of those with a commitment went to a US postdoc. In computer and information sciences, 19.0% did. Use those shares to check whether a low postdoc salary is a short stop or a likely path.

Later salaries depend heavily on sector. The National Science Foundation's Survey of Doctorate Recipients looks at US-residing full-time employed doctoral scientists and engineers with a science doctorate. In 2023, their median annual salary was $130,000 overall, $100,000 in 4-year educational institutions, $170,000 in private for-profit employers, and $144,000 in the federal government. For biological, agricultural, and environmental life sciences, the overall median was $129,000, $100,000 in 4-year institutions, $168,000 at private for-profit employers, and $139,000 in the federal government. For physical sciences, geosciences, atmospheric sciences, and ocean sciences, the overall median was $140,000, $100,000 in 4-year institutions, $165,000 at private for-profit employers, and $149,000 in the federal government. For computer and information sciences, the overall median was $185,000, $119,000 in 4-year institutions, $212,000 at private for-profit employers, and $175,000 in the federal government.

The Bureau of Labor Statistics gives another check on occupation-wide pay in the field. In May 2025, medical scientists, except epidemiologists, had a median annual wage of $103,410. Chemists had a median of $91,240, and biological technicians had a median of $57,510. These are not PhD-specific figures, so do not treat them as the payoff for a doctorate. Compare them with the actual jobs you would target. The salary pages and the academia versus industry guide can help you read the difference between a training salary, a first job, and a mid-career median. Outcomes vary, and no figure here is a promise.

When is a PhD not worth it?

A PhD is probably not worth it if you cannot name the research work you want to do and the jobs that require it. It is also a weak choice if you would take on heavy debt for a stipend that does not cover your costs, or if you are using the degree to delay a career decision. The degree takes years, and the opportunity cost is real.

It is also not worth it if your target job rewards experience more than a doctorate. Some science-adjacent roles value a master's, a license, or years on the job. Some research roles do require a PhD, and some do not. Read the postings rather than the general advice. If the postings you want do not ask for a PhD, ask why you would spend the time and money.

A PhD can still be worth it if the work itself matters to you, the funding is workable, and the path after graduation is plausible in your field. That is a judgment about fit and working conditions. The degree is a tool for a particular kind of work. It is not a universal upgrade.

What about a PhD in the UK?

For a UK doctorate, use the UKRI studentship as a benchmark. From 1 October 2026, the UKRI minimum doctoral stipend for full-time students is £21,805 a year, or £23,805 with London weighting. It says funding for full-time studentships usually lasts between three and four years. That is the training income, not the salary after the PhD. The material here does not include UK salary outcomes, so do not map the US salary figures onto the UK. Ask for the written funding offer and check what it covers.

What to do next

Write down the job you want and collect current postings for it. Mark which ones ask for a PhD and which ask for a master's, a license, or experience.
Get the written funding offer and compare it with the pay you would give up. If the stipend does not cover your costs without heavy borrowing, treat that as a warning.
Ask the program for its median time to degree and funding length. Ask people in the sector you want how they spend their days and what degree the work actually uses.
If the only reason to do the PhD is that you are unsure what else to do, pause. Test your plan against real sectors with the sector and salary pages.