Can a student do research without a lab

Yes, and for a high school student it is usually the better choice. Lab access is the single hardest thing to obtain and the single most common reason a project stalls before it starts — waiting on an institution, a supervisor’s schedule, safety approval or equipment. Most published high school research is not bench science, and the routes below are not consolation prizes. They are how a large share of the social sciences, humanities, economics, public health and computer science literature is produced.

What follows is six routes, what each actually requires, and what each is bad at.

1. Secondary data analysis

You analyse a dataset somebody else collected. Government statistics, census and labour data, public health releases, climate and weather records, sports statistics, election results, transport and crime data, open scientific repositories.

Needs: a spreadsheet, or Python or R if the dataset is large. Patience with documentation — most of the work is understanding how the data was collected and what its gaps mean.

Good for: a first project, and for anyone on a compressed timeline. There are no participants to recruit and nothing to wait for.

Bad at: questions nobody has collected data about. You are limited to what exists, which is why the question and the dataset have to be chosen together.

2. Systematic review

You survey the existing literature on a defined question with explicit search terms, explicit inclusion and exclusion criteria, and an explicit account of what you found and excluded. The rigour is in the method, not in new measurements.

Needs: database access — a school or public library usually provides enough — and discipline about documenting the process.

Good for: a student who reads well, and for any field where the primary research is out of reach. It also produces exactly the background knowledge that makes a second, original project much stronger.

Bad at: producing a novel finding. A review synthesises; it does not discover. Be honest about that in the write-up rather than dressing it up.

3. Survey research

You design an instrument, recruit respondents, and analyse the responses.

Needs: far more care than it looks. Use validated scales from the published literature wherever one exists rather than writing your own questions — a home-made scale is the most common flaw in high school survey work. Human subjects means consent, and for minors that means informed consent from guardians, anonymised data, and a plan for storage.

Good for: questions about attitudes, behaviour and self-reported experience, especially in a population the student can actually reach.

Bad at: anything where a convenience sample of one school will not generalise, which is most things. The fix is not a bigger claim; it is a smaller, honestly stated one.

4. Computational and modelling work

You build or analyse a model, an algorithm, a simulation, a proof or a formal argument. Mathematics, computer science, physics, economics and philosophy all have serious routes here.

Needs: a laptop. Real skill in the method, which the student may need to build first.

Good for: students who already program or who are strong in mathematics. Reproducing a published result and then extending it in one direction is an excellent, under-used project shape.

Bad at: being judged by a general reader. The work needs framing so that someone outside the subfield can see what was done and why it matters.

5. Qualitative and textual analysis

Interviews, open-ended responses, or thematic and close analysis of texts, media, archives and documents.

Needs: a defined corpus, a coding scheme applied consistently, and the honesty to report what contradicts the reading. Digitised archives and newspaper collections have made this dramatically more accessible than it was a decade ago.

Good for: humanities and social science students, and for local history where the source material exists and nobody has looked at it.

Bad at: looking rigorous to readers who expect numbers. It is rigorous, but the method section has to earn it explicitly.

6. Comparative and observational study

You compare cases, texts, policies or places, or you observe and record something systematically over time.

Needs: a defensible basis for the comparison and consistent measurement. Systematic observation — of a public space, a species, a media pattern, a market — needs a protocol written before observation starts, not after.

Good for: questions rooted in the student’s own environment, which are frequently the most original questions available to them.

Bad at: causal claims. Comparison shows difference; it does not explain it, and a paper that pretends otherwise will be marked down by any careful reader.

Choosing between them

Work backwards from what you can actually get. If a dataset exists, take route 1. If the student reads better than they compute, take 2 or 5. If they program, take 4. If the question is about people they can reach and the ethics are manageable, take 3. If the interesting thing is nearby and nobody has documented it, take 6.

Then check the choice against the question using the FINER test in what makes a good research question — feasibility is the test that fails most often, and the method is where feasibility is decided.

What “no lab” does not excuse

Not having a lab lowers the barrier to starting. It does not lower the standard.

The work still needs a real question, a method described precisely enough that someone else could repeat it, results reported including the ones that were inconvenient, and limitations stated plainly rather than buried. A survey of ninety students at one school is a legitimate study with a stated limitation; the same survey presented as evidence about teenagers in general is not.

Reviewers at venues that actually review are looking for exactly this. Honest scope is not a weakness in a high school paper — it is the clearest signal that the student understands what they did.

When the lack of a lab is a real constraint

Be straight about the exception. If the student wants to be a bench scientist and the entire point is learning wet-lab technique, none of the above substitutes. In that case the right moves are a university outreach program, a summer research placement, or a local lab that takes volunteers — and a paper is not the goal that year.

For everyone else, the absence of a lab is not the obstacle. Choosing a question that needed one is.

Published 19 August 2026. Written by the Meridian Scholars team.