Six major research programmes are genuinely free, and two of them publish acceptance rates — RSI at under 2.5% and the Simons programme at about 5%. Three widely listed as free are not: Garcia charges $4,000, UT Austin's HSRA $4,000, and MIT Beaver Works $3,000 above a $200,000 income threshold.
The awkward thing about writing this comparison is that we sell the paid option. So here is the free side first, in as much detail as we can give it, including the bits that make our own product look unnecessary.
Which research programmes are genuinely free?
Six, and the differences in what "free" covers matter. Several are covered in more detail in our guides to MIT research programmes and summer science research programmes.
| Programme | Host | What free means | Who can apply |
|---|---|---|---|
| RSI | Center for Excellence in Education at MIT | Everything: tuition, housing, meals | Juniors, 16+ by 1 July; US direct, international via partner bodies |
| MIT PRIMES | MIT Mathematics | Free, year-long, after-school | Sophomores and juniors, US residents only |
| MITES Summer | MIT | Free except travel; travel scholarships exist | 11th grade, US citizen or green card |
| MITES Semester | MIT | Free, six months online | 11th grade, US citizen or green card |
| Stanford SIMR | Stanford Medicine | Free; $50 application fee, waivable | Juniors and seniors, 16+, US citizen/PR, must live with a guardian |
| Clark Scholars | Texas Tech | Free, plus a $750 stipend | 17+ at start, rising seniors and recent graduates, US citizen/PR |
| Welch Summer Scholar | Welch Foundation, Texas campuses | Free and residential | Texas students only, sophomores and juniors |
Two of these are better than anything money can buy. RSI is free, covers housing and meals, and states on its own FAQ that it "receives over 2,500 U.S. applications and 600 international applications each year and accepts less than 2.5% of applicants." Stony Brook's Simons Summer Research Program states about 5%.
Those are the only two published acceptance rates in the category. Every other percentage you will see for these programmes was invented somewhere.
Which ones only look free?
Three, and they appear on free lists constantly.
Garcia Summer Program, Stony Brook — a $4,000 laboratory usage fee, plus $50 registration and $66 health services. Housing, if needed, is $2,599. No stipend. It is a strong seven-week materials science programme and one of the few that accepts international students. It is not free.
Summer High School Research Academy, UT Austin — $4,000 plus a $400 deposit, no housing, no transport, Texas residents prioritised. We cover it in detail in our guide to the UT Austin programme.
MIT Beaver Works — free below $200,000 in household income and assets, $3,000 above it, and no housing at either price.
What does free actually cost you?
Eligibility and timing, which is the part the free lists leave out.
Citizenship. MITES, Stanford SIMR, Clark Scholars, the Simons programme and MIT PRIMES all require US citizenship or permanent residency. PRIMES directs international students to CrowdMath instead. RSI admits international students only through country-specific partner bodies, not direct application. For most of the world, most of this table is closed before merit is considered.
The calendar. These close between November and February for the following summer. MIT PRIMES closes 2 November 2026. A student who starts looking in April has missed the year.
The odds. RSI rejects roughly 3,000 students annually, most of them excellent. A plan consisting only of free applications has a very high chance of producing nothing.
And one that catches people out: NASA's OSTEM internships require enrolment at an accredited college or university. They appear on dozens of high-school lists. Applying costs you a deadline you could have spent elsewhere.
So what does paying actually buy?
Three things, and we would rather name them precisely than gesture at value.
Availability. No citizenship requirement, no geography requirement, rolling admission. For an international student, this is often the entire proposition — not that a paid programme is better than RSI, but that RSI is not an option.
A deadline with a reader attached. Most students who attempt independent research do not fail on intellect. They fail because nothing external requires a draft by a particular Thursday, and nobody with expertise reads it when it arrives. That is the mechanism you are buying.
Ownership of a question. Lab placements are genuinely valuable and usually teach more about research life. They rarely produce a paper with a school student's name on it, because the work belongs to a larger project on a longer timeline.
What paying does not buy: selectivity as a signal. Our overview of research opportunities sets out what each provider publishes and what it does not. Paid programmes publish acceptance rates between "about 30%" and nothing at all, and none is audited. Being accepted to one tells an admissions officer nothing about you.
What do universities actually say?
One institution is specific, and it is worth reading exactly.
Caltech reports that 42% of admitted students submitted research, and specifies that it should come from "a laboratory in a university or research-affiliated organization" with "a letter of recommendation from site of research." Independent research and internships are directed to a portfolio instead.
That is not a dismissal of independent work. It is a statement that Caltech distinguishes supervised institutional research from self-directed projects and routes them differently. Most universities publish nothing this precise — which means anyone telling you confidently how research is weighted is guessing, including companies selling it.
Where we sit, stated plainly
The Oxford Centre for Advanced Research runs 1:1 mentorships with PhD candidates and early-career researchers at Oxford, Cambridge, other leading UK universities and Ivy League institutions. The Oxford Scholar Programme is £2,000 — about $2,600 — for ten contact hours across ten to fourteen weeks. The Oxford Publication Fellowship is £3,400, roughly $4,400, and supports submission to the Oxford Journal of Student Scholarship, the Journal of Research High School or Vanderbilt's Young Scientist Journal.
We accept approximately 60% of applicants, self-reported and unaudited like every figure in this sector.
If RSI, PRIMES, MITES, SIMR or the Welch programme are open to you, apply there and do not pay us. They are better and they cost nothing. We would say the same on a call, and we have.
We are the right choice in a narrow case: the free routes are closed to you by citizenship or calendar, you have tried the local labs, and the alternative is another year of intending to do a project. That is a real situation and it describes most of the students we work with. It is not the same as being the best option available in the abstract.
The order to do this in
Check eligibility first — ten minutes, highest-return ten minutes in the process. Apply to every free programme you qualify for, in the autumn. Email five academics in January with something specific to say about their work — how you approach them matters more than how many you contact. Give it six weeks.
If nothing lands by March, decide deliberately. Students who make a decision in March finish something by September. Students still waiting for a reply in June generally do not.
Frequently asked questions
Which research programmes are actually free?
RSI covers everything including housing and meals. MIT PRIMES is free and runs a full year. MITES Summer and MITES Semester are free apart from travel. Stanford SIMR is free with a waivable $50 application fee. Clark Scholars is free and pays a $750 stipend. The Welch Summer Scholar Program is free and residential, but Texas students only.
Which ones only look free?
Stony Brook's Garcia programme charges a $4,000 laboratory fee plus registration and health fees, with optional housing at $2,599. UT Austin's High School Research Academy is $4,000 plus a $400 deposit. MIT Beaver Works is free only for families under $200,000 in household income and assets, and $3,000 above it.
How competitive are the free ones?
Only two publish a figure. RSI states it receives over 2,500 US and 600 international applications a year and accepts under 2.5%, for a cohort of 100. The Simons Summer Research Program states about 5%. The others publish cohort sizes at most — Clark Scholars takes twelve students, Stanford SIMR around fifty.
So why would anyone pay?
Three reasons that hold up. Eligibility — most free US programmes require citizenship or permanent residency. Timing — they run on a fixed winter calendar and a student who misses it waits a year. And a deadline with a reader, which is what most students who fail to finish a project actually lack.
Do paid programmes look worse to admissions officers?
Not inherently, but the artefact matters more than the badge. Caltech reports that 42% of admitted students submitted research and specifies it should come from 'a laboratory in a university or research-affiliated organization' with a letter from the research site. Independent work is directed to a portfolio instead.
What is the honest order to try things in?
Apply to every free programme you are eligible for, in the autumn, because those applications expire. Email five local academics in January. Give it six weeks. If nothing lands by March, choose a paid option deliberately rather than drifting into one — students who decide in March finish something by September.