It happens to almost all first-time researchers: You download your first academic paper, read the abstract twice, and still can't say what the authors found. By page three you're looking up every other word. Your first academic paper can feel like it's written in another language. That's normal, and it gets easier fast once you stop reading front to back. Here's the order researchers actually read in, and how to take notes you'll use later.
Key takeaways
Feeling lost in your first few papers is normal, and it eases with every paper you read.
If you’re new to a field, public science writing is a gentle way to build context first.
Reading in three passes helps you spend your time on the papers that matter most.
Figures are often the quickest route to a paper’s main finding.
Consistent notes from day one become the backbone of your literature review.
PubMed Central, arXiv, and Google Scholar are reliable free places to start.
Why papers feel hard at first
Research papers are written by specialists for other specialists. They assume you already know the field's vocabulary, methods, and recent debates, so as a high school student you're joining the conversation halfway through. Some confusion is expected.
School habits add to it. Tests reward finding the single right answer, while papers ask you to skim, jump around, and live with open questions. We wrote more about that shift in From Practice Tests to Original Ideas.
At Scholar Launch's Irvine panel, the speakers gave advice worth tapping above your desk: read as much as you can, and don't get discouraged when your first paper gives you a headache. Read more from that panel in What the Experts Say.
How do I understand scientific papers without a science background?
A good first step is building context before opening a journal. For students who haven’t settled on a field yet, the panel recommended starting with science writing for the general public, like National Geographic or the news section of Science. These pieces often link to the original study, and reading the story first makes the paper’s jargon much easier to decode.
The three-pass method
One of the most useful habits our Faculty Advisors pass on is reading a paper in layers instead of front to back. A well-known version comes from computer scientist S. Keshav’s guide, “How to Read a Paper,” which describes reading each paper up to three times, a little deeper with each pass. Most papers only need one or two.
Pass 1: The skim (5 to 10 minutes). The first pass is about the big picture. Keshav suggests reading the title, abstract, and introduction, then the section headings, the conclusion, and a quick glance at the references. By the end, you should have a rough sense of what he calls the five Cs: category, context, correctness, contributions, and clarity. If a paper turns out not to fit your topic, it’s perfectly fine to set it aside and save yourself an hour.
Pass 2: The careful read (up to an hour). On the second pass, you read the whole paper more closely while letting heavy details like proofs wait. This is a good moment to study the figures, jot key points in the margins, and mark references you may want later. A helpful check at the end: could you explain the main point and its evidence to a friend?
Pass 3: The deep read (several hours). Keshav describes this pass as re-creating the work in your head, using the same assumptions the authors made, and he estimates four or five hours for beginners. Our TAs usually suggest saving this level of reading for the two or three papers a project builds on most directly.
Read the figures before the text
Our Teaching Assistants often suggest starting with the figures, since that’s where most papers show their main result. Here’s a simple routine that works well:
Start with the caption.
Look at the axes and units to see what was measured and what’s being compared.
Notice the biggest difference or trend.
Try summing it up in one sentence: “This figure shows that ___.”
Then compare your sentence with the results section. If the two don’t line up, you’ve found exactly the spot worth rereading or bringing to your mentor.
Take notes that turn into your literature review
A habit that pays off later is using the same short template for every paper. When it’s time to write a literature review, students who do this compare tidy notes instead of rereading 20 PDFs. Here’s a version our TAs often share:
Citation: authors, year, title, journal, link
Question: What were they trying to find out?
Method: How did they test it, in one or two lines?
Finding: the main result, in your own words
Limitation: what the authors admit they couldn’t do (usually in the discussion section)
Connection: how this relates to your question
Words to look up
One more tip from our mentors: put any exact phrase you copy in quotation marks with the page number, an easy way to avoid accidental plagiarism later. Keeping every paper in one spreadsheet also helps patterns stand out. A limitation that appears in paper after paper often points to a question nobody has answered yet, which is frequently where a strong research question begins.
Where to find papers for free
High school students have more free research available online than many realize. Of all the options, three come up most often when our Faculty Advisors and Teaching Assistants recommend where to begin: PubMed Central, arXiv, and Google Scholar.
| Source | Best for | What to know |
|---|---|---|
| PubMed Central | Biology, medicine, health, and psychology | A free full-text archive run by the U.S. National Library of Medicine at NIH. Every article opens in full. |
| arXiv | Physics, math, computer science, statistics, economics, engineering, and quantitative biology | Hosts preprints, which are papers shared before or during peer review. Great for the newest work, but check whether a published version exists. |
| Google Scholar | Every field | A search engine, so not every result is free. Look for [PDF] links on the right side, or click "All versions" to find an open copy. |
How can Scholar Launch help?
Reading papers is one of the first skills our Teaching Assistants build with students. In 1-on-1 TA sessions, students learn academic writing from the ground up: how a paper is structured, how to read one with confidence, how to find and evaluate the right sources, how to write an academic paper of their own, and how to get it ready to submit for publication. Students do the reading, thinking, and writing themselves, with an experienced guide beside them so no one stays stuck on the same page for a week.
FAQ
What if I don’t understand the methods section?
You’re in good company, since it’s one of the sections students ask our TAs about most often. On a first pass, it’s fine to skim past most of it. On the second pass, it helps to focus on three things: what was measured, on whom or what, and how the results were compared. Whatever is still unclear makes a great question for your mentor, and a specific one, like “Why did they use this control group?”, usually gets a far more helpful answer than “I don’t get it.”
Can I use AI to summarize papers?
AI can be a useful starting map, as long as you check it against the paper itself. These tools sometimes misstate findings, skip limitations, or invent sources that don’t exist. Our TAs suggest using a summary to decide what to read closely, then writing your notes from the original. An AI summary shouldn’t be cited as a source.
How many papers do I need before I start?
Usually fewer than students expect. Keshav suggests beginning a literature survey with three to five recent, highly cited papers in your area and then following the references they share. Once you’ve read those core papers carefully and can name a gap they leave open, you’re in a good position to start shaping a question. Reading continues throughout a project, so there’s no need to wait.
How do I know if a paper is reliable?
A few questions go a long way. Was it peer reviewed and published in an established journal, or is it a preprint? How recent is it? Do other researchers cite it? Is the sample large enough to support the claim? Searching the title in the Retraction Watch database is another quick way to confirm it hasn’t been retracted. Over time, our Faculty Advisors help students build a sharp instinct for telling strong sources from weak ones.
Read your first papers with expert guidance this winter
If you’d like to build these skills alongside experienced mentors, Scholar Launch’s Winter 2027 Group Research Program runs January to March 2027 for high school students. Over 12 weeks, you’ll explore a research question of your own with guidance from a Faculty Advisor and 1-on-1 support from a Teaching Assistant, with nine tracks across STEM, Humanities and Social Science.
Apply by November 22, 2026 to receive 2 bonus TA sessions. Early Bird Round II closes November 30 with one bonus session.
