✦clariva

Research 101

A High-School Student's Guide to Getting Started

Maybe someone told you to "do research" or "read some papers", and now you have 17 Google Scholar tabs open and absolutely no clue what you're supposed to do with them.

Fair.

This guide starts from zero.

Made for students who have never done research before.

Research sounds much more complicated than it actually is when you're starting.

At its core, it's basically:

a question + evidence + your analysis of that evidence

You find something you're curious about, turn it into a question, see what other people already know, collect or examine evidence, and work out what that evidence tells you.

For a first project, you can start much smaller than that. A literature review, survey, local data project, or simple experiment still counts.

You just need somewhere to start.

Quick note: This is a starting guide, not a rulebook. Different research programs, competitions, journals, and schools may have their own requirements.

1.So... what actually is research?

Say you're interested in inflation.

Searching "What is inflation?" and reading a few websites is useful, but that's mostly learning about a topic.

Research gets more specific.

Research question

How has food-price inflation affected how teenagers spend their monthly allowance?

Now you have something you can actually investigate. You could read previous studies, look at price data, survey students, or compare spending patterns.

AI / CSHow do Grade 11-12 students use generative AI when doing mathematics homework?
HistoryHow did two newspapers describe the same historical event differently?

The difference is that research isn't just collecting information. You're trying to answer a specific question using evidence.

A school essay often asks you to explain something we already know. Research usually asks you to investigate something.

2.Finding something worth researching

A lot of students get stuck here because they think they need a genius idea.

You don't need one.

"I noticed this. Why does it happen?"

That is enough to begin.

Write down three subjects or areas you actually like. Then write a few things about each one that make you curious.

Psychology

Why do I procrastinate more at night?

Does scrolling before bed affect sleep?

Why do some students participate in class more than others?

Possible question: Is late-night short-form video use associated with sleep among students aged 15-18?

EnvironmentHow has extreme summer heat affected school attendance in my city?
ScienceHow does light intensity affect the output of a small solar panel?

Your topic is probably too broad if...

...you could write three books about it and still not finish.

Too broad: Climate change
Better: How has extreme summer heat affected school attendance in [location]?
Too broad: Social media
Better: Is late-night short-form video use associated with sleep among students aged 15-18?

You can make a topic smaller by adding a specific age group, place, time period, platform, event, community, or type of evidence.

Smaller usually makes research easier, not weaker.

3.Turning your idea into a research question

Your research question is the specific question your whole project is trying to answer.

How / Why / To what extent + issue + specific group, place, or time

You don't have to follow that perfectly. It just helps when your question is still something like "AI in education?"

From vague to useful

Too vague: Is AI good for students?

Better: How do students use AI for homework?

Stronger: How do Grade 11-12 students use generative AI for mathematics homework, and what benefits or problems do they report?

Your research question can change later. Actually, it probably will.

You might start with "How does social media affect students?" Then read a few papers and realize sleep is the part you find interesting.

So your question becomes: Is late-night short-form video use associated with sleep among students aged 15-18?

That's normal. Your question usually gets sharper once you've actually started reading.


4.Finding and reading existing research

Before collecting your own data, find out what other people already know.

Good places to search include:

  • Google Scholar
  • government websites
  • university websites
  • reports from credible organizations
  • JSTOR, if you have access
  • other academic databases your school or library provides

Regular Google can still help, especially when you're learning what terms people use. Just don't let the first random blog you find become the foundation of your entire paper.

Search more than one phrase

Say your topic is digital payments and teenagers. Don't only search teenagers money.

Try instead

teenage spending habits digital payments
adolescent financial behaviour
youth digital payment use

Researchers may use words you wouldn't have thought of. Once you find one useful paper, look at the terms it uses and search those too.

You do NOT need to read every paper from page 1 to page 27

Title → Abstract → Conclusion → Relevant sections

The abstract is the short summary near the beginning of a research paper. It usually tells you what the researchers studied, how they studied it, and what they found.

If the abstract has almost nothing to do with your question, move on.

Your goal isn't to collect 87 PDFs. Your goal is to find a smaller number of useful sources you actually understand.

Can I trust this source?

  • Who wrote it?
  • Where was it published?
  • Is there evidence?
  • Is it recent enough for my question?
  • Does it cite sources?
  • Can I confirm the main claim somewhere else?

Academic papers, government reports, university research, and reports from established organizations are often useful starting points. News articles can help with recent events.

Wikipedia can help you understand a topic and find references, but it usually shouldn't be the main evidence in your paper.

5.Literature review + research gap, minus the scary wording

You're going to hear literature review a lot.

It simply means understanding what previous researchers have found and connecting those findings together.

Not: "Paper A says this. Paper B says this. Paper C says this. Goodbye."

You're looking for patterns. Maybe most studies found one thing. Maybe two studies disagree. Maybe lots of research exists about adults, but barely anything about teenagers.

Research gap

A research gap is something existing research hasn't fully answered.

It does not mean nobody in human history has ever thought of it before.

A gap could be a different country, younger age group, newer data, a local version of an existing question, a comparison previous studies didn't make, or studying something again after a major change.

Existing research: Digital payments and spending among adults.
Your question: Digital-payment use and spending behaviour among high-school students.

That younger age group could be the gap.

And for a first project, you don't need to panic about finding some enormous original gap anyway. You're learning how research works.

6.Choosing how you'll answer your question

The word methodology basically means: How are you going to answer your research question?

Your method depends on the question.

Literature / review researchYou use existing studies and sources rather than collecting new data. Good for: What does existing research suggest about social-media use and adolescent sleep?
Historical / source-based researchYou study newspapers, speeches, laws, letters, archives, or old government documents. A primary source comes from the time or event itself; a secondary source is later analysis of that evidence.
SurveysGoogle Forms enters the chat. Useful when you want to understand people's behaviours, experiences, or opinions.
InterviewsYou speak directly to people with relevant experience. Useful when you want more detailed answers than a survey can give.
Existing data analysisYou use public datasets instead of collecting everything yourself — for example, comparing unemployment and inflation data across countries.
ExperimentsYou change something and observe what happens — for example, how light intensity affects the output of a small solar panel.
Computational researchYou use code, models, simulations, or datasets. This could be anything from analyzing data in Python to testing a simple machine-learning model.
Choose the method that answers your question. Don't pick machine learning because it sounds impressive and then desperately search for a question to attach to it. The method is supposed to serve the question, not your résumé.

Quick survey rules

  • keep questions neutral
  • don't push people toward the answer you want
  • don't collect private information you don't need
  • test the survey on two or three people first
  • keep it reasonably short
Bad: Don't you think AI makes students lazy?
Better: How, if at all, has using generative AI affected the amount of time you spend on schoolwork?

There is no magical perfect sample size. More responses can strengthen your evidence, especially if the people responding actually represent the group you're studying.

But a small student project can still be useful. Just be honest about its limits.

Not 100 responses after some "creative mathematics."

Research involving people

If your work involves surveys, interviews, minors, health information, mental health, sensitive personal information, or anything that could affect participants, talk to a teacher, school, supervisor, or another appropriate adult first.

People should understand what they're agreeing to. Don't collect personal information for fun. Don't pretend responses are anonymous if they aren't.

And don't conduct risky medical or psychological experiments without proper supervision.


7.Keep your sources organized

A basic table is enough to keep track of your sources.

SourceAuthorYearMain ideaUseful evidenceHow it helps meLink
Example paperA. Researcher2026One-sentence summaryKey findingWhy it mattersSave it here

You can use Google Sheets, Excel, Notion, a document, Zotero, Mendeley — whatever you'll actually keep using.

If you save nothing because you think "I'll remember where I found it later" — you probably won't.

Also, do not spend three hours making the world's prettiest Notion research dashboard before reading your second paper.

If organizing everything starts taking longer than the research, simplify it.

8.Do you need a mentor?

No.

You do not need a professor before you're allowed to start research.

You can independently do things like literature reviews, historical research, basic surveys, public-data analysis, simple coding projects, and some safe experiments.

A mentor becomes useful when your project includes difficult statistics, advanced lab work, highly technical topics, specialist equipment, complicated datasets, or ethics concerns.

And "mentor" does not automatically mean professor. It could be your teacher, a PhD student, a researcher, a professor, or another subject expert.

A person who actually has time to help you can be much more useful than a famous professor who never replies.

If you want to contact someone

Don't send the exact same email to 100 academics. Find someone whose work genuinely connects to your question. Read a little about what they do. Then send something short.

Hi Dr. Rao, I'm a Grade 11 student exploring a small research project on digital payments and teenage spending. I came across your work on consumer payment behaviour and wanted to ask whether you think a survey would be a reasonable way to approach this question. I completely understand if you don't have time to reply, but I'd appreciate any brief guidance.

A specific email usually works better than a fancy one.

9.Make a simple research plan

Before doing anything huge, answer these:

  • My broad topic: What am I interested in?
  • My research question: What exactly am I trying to answer?
  • What have other researchers already found?
  • What do I still want to understand?
  • My method: Survey? Historical sources? Existing data? Experiment? Literature review?
  • What evidence do I need?
  • Where can I get it?
  • What might go wrong?
  • When do I want to finish?

Maybe you can't get enough survey responses. Maybe a dataset is missing. Maybe your original question is impossible to answer.

Will something go wrong halfway through? Probably. Welcome to research.

A basic project might take around 8-10 weeks:

  1. Weeks 1-2

    Choose and narrow your question.

  2. Weeks 3-4

    Find and read sources.

  3. Week 5

    Choose your method.

  4. Week 6

    Collect evidence.

  5. Week 7

    Analyze.

  6. Weeks 8-10

    Write, revise, fact-check, and cite.

Some projects take much longer. That's normal.

10.Writing the paper

A common research-paper structure looks like this:

  1. 1.

    Title

    Tell the reader clearly what the project is about.

  2. 2.

    Abstract

    A short summary of the whole paper.

  3. 3.

    Introduction

    Your question, why it matters, and what you're investigating.

  4. 4.

    Literature Review / Background

    What previous research already tells us.

  5. 5.

    Method

    How you answered the question.

  6. 6.

    Results

    What you found.

  7. 7.

    Discussion

    What those findings might mean.

  8. 8.

    Conclusion

    Your answer, limitations, and possible next questions.

  9. 9.

    References

    The sources you cited.

History and humanities papers may look different. You might use argument-based sections instead of neat "Method → Results → Discussion" headings. That's completely fine.

Results not what you expected?

Maybe annoying, but not a failed project. If you predicted that heavier AI use would mean lower test scores and your data shows no relationship, report that. Your job is to follow the evidence, not prove yourself right.

Correlation does not equal causation

If students who use more social media also sleep less, the two are correlated. That doesn't prove social media caused the lack of sleep. Maybe poor sleepers use their phones more; maybe stress affects both.

Research often gives you clues, not perfect proof.


11.Things you really should NOT do

✕Plagiarize

Plagiarism is presenting someone else's work or ideas as your own. If you use another person's idea, data, argument, or exact words, you usually need a citation showing where it came from. Changing three words is not proper paraphrasing. Understand the idea, explain it yourself, and cite it. Use the citation style your school, competition, or journal asks for — often APA, MLA, or Chicago.

✕Change your data

If 43 people responded, you have 43 responses. If the result doesn't support your hypothesis, report it anyway.

✕Hide inconvenient results

A result that contradicts your expectation can still lead to an interesting discussion. It belongs in the project.

✕Overclaim

If you surveyed 30 students from one school, don't write "Teenagers around the world believe..." You studied 30 students from one school. Say that.

✕Confuse complicated with good

You do not need sentences that sound like they were discovered inside an ancient philosophy textbook. Clear writing is good writing.

✕Use AI as fake evidence

AI can help explain concepts, brainstorm search terms, organize ideas, and debug code. It should not invent citations, fabricate data, create fake interviews, make up findings, replace original sources, or be treated as evidence. Verify facts and check the AI rules wherever you plan to submit.

12.What happens after you finish?

You've finished the paper. Now what?

Depending on the project, you could:

  • present it at school
  • enter a research competition
  • adapt it for an essay competition
  • submit it to a student journal
  • enter a science fair
  • share it at a student conference
  • include it in a portfolio
  • share useful findings with an organization
  • continue the research later

You do not need publication for the project to count.

If you do look at journals, check who runs them, whether the review process is explained, what the fees are, whether previous papers are available, and whether they make suspicious promises.

"Pay us and we guarantee publication" is not exactly the academic green flag you're looking for.

13.Start today

Do this before you close this guide:

  • Pick 3 topics you actually like
  • Write 5 questions about them
  • Pick your favourite one
  • Search it on Google Scholar
  • Try 2-3 different search phrases
  • Save 5 useful sources
  • Read their abstracts
  • See whether your question needs narrowing
  • Write down one possible method
That's enough for Day 1. Seriously. You do not need a 20-page proposal before you're allowed to begin.

14.Your research roadmap

  1. 1.

    CURIOSITY

    Something catches your attention.

  2. 2.

    TOPIC

    Choose the general area.

  3. 3.

    QUESTION

    Decide what you actually want to know.

  4. 4.

    READ

    See what other researchers have already found.

  5. 5.

    REFINE

    Make your question clearer or narrower.

  6. 6.

    CHOOSE A METHOD

    Work out what evidence could answer it.

  7. 7.

    COLLECT EVIDENCE

    Sources, data, surveys, interviews, experiments — whatever fits.

  8. 8.

    ANALYZE

    Ask what the evidence actually shows.

  9. 9.

    WRITE

    Explain your reasoning clearly.

  10. 10.

    EDIT

    Fix weak arguments, missing citations, unclear wording, and errors.

  11. 11.

    SHARE

    Submit it, present it, publish it, or simply use what you learned for your next project.

Ready for the next step?

Once you know what you're interested in, head back to Clariva to explore research programs, competitions, journals, summer programs, and other research-related opportunities available to you.

Still stuck?

Have a research question you're stuck on? Sometimes the problem isn't the research itself. You're just stuck on what to do next.

Research 101 | Clariva Guides