TL;DR
You’ll find everything you need to write a research abstract below: the five sentences every abstract needs, a fill-in template, word limits for ISEF, JSHS, and four student journals, and three annotated examples from a lab project, a survey, and an English paper. Jump to the five sentences every abstract needs, word limits by venue, three worked examples, and the mistakes that get abstracts rejected. All six venues cap abstracts between 200 and 250 words, so you’re writing one tight paragraph, not a page. Check your venue’s limit before you draft, because at JSHS an abstract over 250 words gets shortened by someone else. Submitting soon with no one to read your draft? Apply to Horizon and work on yours with a PhD or post-doc instructor.
What a research abstract is
A research abstract is a short summary of your entire project that makes sense without anyone reading the rest. An abstract has to stand alone, because many readers, and some judges, will read nothing else. It isn’t a teaser. It gives away the ending. Someone reading only your abstract should know what you asked, how you looked for the answer, what you found, and why it matters.
A few terms first, so you don’t get stuck later. A venue is wherever you’re submitting: a competition, a symposium, or a journal. A submission portal is the website or form where you upload your work. Keywords are the short search terms some venues ask you to list under the abstract.
An abstract summarizes the finished paper, results included, while an introduction opens the paper, builds the case for your question, and saves the findings for later.
Why does this matter so much? Society for Science says Regeneron ISEF’s special award organizations and Grand Award judges review the abstract to decide whether a project stands out, and visitors read it for a quick overview. The abstract is often the only part a judge reads before deciding whether to read the rest. Treat it as the front door to your project, not a chore you finish at midnight.
If your project is a literature review instead of original research, the same structure works, and your “result” is what the literature shows. Our guide to writing a literature review in high school shows you how to get there. Check first that your venue accepts reviews, though, since JSHS doesn’t.
Types of research abstracts
For a high school competition or journal, you’ll almost always write an informative abstract, one that includes your results. The other types are worth recognizing so you don’t copy the wrong model.
| Type | What it includes | Where you’ll see it |
| Informative | Context, question, method, results, and conclusion | Every venue in the table below. This is the one you’ll write. |
| Descriptive | Topic, purpose, and approach, but no results | Some humanities papers and conference proposals written before results exist |
| Structured | An informative abstract split under labels like Background, Methods, Results, and Conclusions | NHSJS and many medical journals |
| Unstructured | An informative abstract in one plain paragraph | ISEF, JSHS, IJHSR, and JEI |
Abstract word limits by competition and journal
Your venue sets your word limit, and the right structure follows. Here are the limits as each venue publishes them right now. Limits change, so confirm yours on the venue’s own page before you submit.
| Venue | Word or character limit | Structured or unstructured | Where the limit is published |
| Regeneron ISEF | 250 words or less | Unstructured paragraph covering purpose, procedure, results, and conclusions. No bibliography, no acknowledgments. | How to Write an ISEF Abstract, Society for Science |
| Junior Science and Humanities Symposium (JSHS) | 250 words maximum (the header above the abstract is separate) | Unstructured paragraph that must include your hypothesis and conclusions | Core Rules of Competition 2026-27, Section IV |
| International Journal of High School Research (IJHSR) | 200 words | Unstructured paragraph, plus keywords | IJHSR submissions page, which links to the research article template |
| National High School Journal of Science (NHSJS) | 200 to 250 words | Structured: Background/Objective, Methods, Results, Conclusions, then keywords | NHSJS Submission Guidelines |
| Journal of Emerging Investigators (JEI) | Under 250 words | Unstructured summary on its own page | JEI Manuscript Format & Content |
| Journal of Student Research, High School Edition (JSR) | 250 words maximum | Not specified. One paragraph summarizing the work from introduction to conclusion. | JSR How to Submit |
Source: each venue’s own submission page, re-checked November 8, 2026.
Reading the table, notice that the limits cluster between 200 and 250 words. That’s why this page builds its main examples around those two sizes. Only NHSJS asks you to label the parts, so for the others you choose the order, and the five-sentence order below is a safe default.
A few notes on the table. JSHS regions run their own competitions, so your regional director can add rules on top of the national ones. JSHS also asks for a header above the abstract with your title, name, school, and teacher or mentor. ISEF tells finalists to leave out acknowledgments, including mentors and awards. If you’re still choosing where to send your paper, our list of 15 research journals for high school students compares more options, and our guide to publishing in the International Journal of High School Research walks through that journal’s process.
What if you’re over the limit? Don’t hope nobody counts. JSHS states that abstracts over 250 words will be shortened, which means a stranger decides what stays. Cut it yourself until you’re under, starting with background and hedging words. The order is in the cutting section below.
The five sentences every abstract needs
Almost every abstract, in every field, does the same five jobs in the same order. Give each job one sentence, two at most, and you’ll have a draft in an hour. If your venue wants a structured abstract, the five jobs just become labeled sections. The word counts below suit an abstract of roughly 200 to 250 words, so scale them down if your limit is lower.
| Sentence | What it answers | Roughly how many words | Common failure |
| Context | What’s the problem, and what’s missing? | 25 to 40 | Starting too broad, like “Climate change is a big problem” |
| The question | What did you ask or predict? | 25 to 40 | Burying the question inside the method, or leaving out the hypothesis |
| The method | What did you do to find out? | 40 to 60 | Listing equipment instead of naming the approach and sample |
| The result | What did you find? | 60 to 80 | Vague words like “significant” with no actual finding |
| The so what | Why does it matter, or what comes next? | 30 to 45 | Leaving it out, or repeating the question |
Context. Start with the gap, not the history of your field. One sentence on what’s known and a clause on what’s missing is plenty. Context is the first thing you’ll cut when space runs short, so don’t fall in love with it.
The question. Say what you tested or asked in plain words. If you ran an experiment, state your hypothesis, which is your testable guess about what will happen (our guide to what a hypothesis is and how to write and test one shows how). JSHS requires the hypothesis in the abstract.
The method. Name your approach and the size of your data, not your equipment list. A phrase like “we surveyed [number] students” does the job. ISEF tells finalists not to go into detail about experimental procedures or statistical methods.
The result. Report the actual finding, with a number if you have one. “Seedlings under blue light grew taller” is a result. “Results were significant” is not.
The so what. End with what your finding means or what should happen next. ISEF asks for conclusions in one or two sentences that can cover applications and implications. The “so what” sentence is the one most students leave out, and the one judges look for. JSHS’s judging criteria ask what was learned and whether you recognize your contribution to the field.
Three annotated examples
Here are three abstracts written from scratch for this page, at three lengths. Each one follows the five sentences. Anything in [square brackets] is a placeholder figure, so swap in your own numbers and never copy these as real results.
Example 1: Lab science (249 words)
Indoor growers often choose grow lights by color, but small-scale growers rarely have data on which color helps young herbs most. Light color affects how fast young plants establish, yet most herb growers choose fixtures by price or habit. This project tested whether basil seedlings grown under blue-heavy light grow taller than seedlings grown under red-heavy light over 14 days. We grew [60] basil seedlings in two matched growth chambers, [30] per light condition, with identical soil, water, temperature, and 16-hour light cycles. Light intensity was matched at [200] micromoles per square meter per second, so color was the only planned difference. Each seedling was measured every other day with a ruler, and final heights were compared using an independent-samples t-test. Seedlings under blue-heavy light reached a mean height of [9.8] cm, compared with [7.1] cm under red-heavy light, a difference of [2.7] cm ([p < 0.01]). Seedlings under red-heavy light produced [more] leaves per plant, but that difference was not statistically significant. Stem thickness did not differ between the two groups. These results support the hypothesis that blue-heavy light increases early seedling height in basil, though leaf count and stem thickness did not follow the same pattern. Both chambers used one basil variety and a short growing window, so the results cannot yet be applied to other herbs or to full-grown plants. Growers who want tall, fast-starting seedlings may benefit from blue-heavy light, and a follow-up test should check whether leaf production catches up at later growth stages.
How it works. Sentences 1 and 2 provide context, and sentence 3 poses the question as a testable comparison. Sentences 4 to 6 cover the method. Sentences 7 to 9 report results, including one that didn’t pan out, which makes the abstract more believable. Sentence 10 ties the result to the hypothesis, sentence 11 states a limit, and sentence 12 is the so what.
Before and after. The first draft ran 287 words, which is over ISEF’s 250 limit. Three cuts brought it to 249:
Context (40 words to 21), before: Indoor growers, hobbyists, and commercial greenhouse operators often choose LED grow lights by color, and there is a large body of research on light and plant growth, but small-scale growers rarely have data on which color helps young herbs most.
After: Indoor growers often choose grow lights by color, but small-scale growers rarely have data on which color helps young herbs most.
Method detail (40 words to 23), before: We grew [60] basil seedlings (Ocimum basilicum, Genovese variety) in two matched growth chambers built from storage bins, [30] per light condition, using the same potting mix, the same tap water, the same temperature range, and identical 16-hour light cycles.
After: We grew [60] basil seedlings in two matched growth chambers, [30] per light condition, with identical soil, water, temperature, and 16-hour light cycles.
Hedging (28 words to 26), before: These results seem to suggest that blue-heavy light may possibly increase early seedling height in basil, though leaf count and stem thickness did not follow the same pattern.
After: These results support the hypothesis that blue-heavy light increases early seedling height in basil, though leaf count and stem thickness did not follow the same pattern.
Example 2: Social science (202 words)
Teachers often assume that students who spend more time on homework are working harder, but late-night homework may cost sleep, and sleep affects learning. This study asked whether weeknight homework time is related to weeknight sleep among high school students, and whether the relationship differs by grade level. We surveyed [180] students in grades 9 to 12 at [one public high school] using an anonymous online questionnaire about typical weeknight homework hours, bedtime, and wake time. Participation was voluntary, no names were collected, and surveys with missing answers were removed before analysis. We calculated each student’s sleep duration, tested its relationship with homework time using a correlation analysis, and compared the pattern across grades. Students who reported more homework slept less: each extra hour of homework was linked to about [22] fewer minutes of sleep (r = [-0.41]). The link was strongest among juniors and weakest among freshmen. The survey was self-reported and drawn from one school, so the results show a link, not a cause, and may not describe other schools. Still, schools reviewing homework policies may want to count sleep as a cost of heavy workloads, and a follow-up study could track the same students across a full school year.
How it works. Sentence 1 is the context and sentence 2 is the question. Sentences 3 to 5 are the method. Sentences 6 and 7 are the results: the first gives a number and the second gives a pattern. Sentence 8 names a limit, and sentence 9 is the so what. Social science abstracts earn trust by saying what the data can’t show, so “a link, not a cause” is doing real work.
Example 3: Humanities (197 words)
Readers usually remember the creature in Mary Shelley’s Frankenstein as a monster made in a laboratory, and critics have often focused on Victor Frankenstein’s ambition. This paper asks a quieter question: what does the creature’s self-education, which includes reading Paradise Lost, Plutarch’s Lives, and The Sorrows of Werter, reveal about how Shelley imagines learning? Using close reading of the 1818 text’s middle chapters, the paper compares what the creature takes from each book with what he is able to do with it. Particular attention goes to the scenes in which he learns language by listening to the De Lacey family, since those scenes tie learning to a household. It argues that the creature absorbs the language and ideals of these books but has no one to practice them with, so his education produces a reader and not a member of a community. This reading moves the blame in the novel away from the creature’s nature and toward the absence of care around him. For readers of the novel, the argument suggests that Shelley treats education as a social act, not a private one, and it connects the book to modern questions about what schools owe their students.
How it works. Sentence 1 is the context, and sentence 2 is the question. Sentences 3 and 4 are the method: close reading, with the specific scenes named. Sentences 5 and 6 are the result, which in the humanities is your argument. Sentence 7 is the so what. A humanities abstract has no numbers to lean on, so the argument itself has to be the result: state your claim in a sentence a stranger could disagree with. If your venue’s page doesn’t ask for an abstract at all, follow the page.
How to cut an abstract to the word limit
Most first drafts run long. Cut in this order, and stop the moment you’re under the limit. Keep the question and the result for last.
- Background. Keep one sentence of context and delete the rest.
- Hedging words. Phrases like “seems to,” “may possibly,” “somewhat,” and “in order to” almost always go.
- Method detail. Brand names, equipment, and step-by-step procedure go. The design and the sample size stay.
- Adjectives and adverbs. Words like “very,” “extremely,” “carefully,” and “important” add length and no information.
- The so what, shortened to a clause. This is a last resort, and you should never cut the question or the result.
Cut context before you cut findings, because the question and the result are the abstract. Here’s one sentence cut three ways, so you can see the moves:
Full (38 words): To find out whether students who get less sleep might perform somewhat worse in school, we conducted a long anonymous online survey of [180] students at one public high school and then carefully analyzed the data using statistics.
Shorter (24 words): To test whether less sleep is linked to lower grades, we surveyed [180] students at one public high school and analyzed the data statistically.
Shortest (14 words): We surveyed [180] students to test whether less sleep is linked to lower grades.
Notice what survived every cut: the sample, the question, and the link being tested. What went first was the story around them. After each round of cuts, recount, since it’s easy to cut fifty words and still be ten over.
Common mistakes
Here are the slips to avoid, each with its fix.
- Writing it before the project is finished. Write it last, from the finished paper, so it reports what you found and not what you expected
- Including citations or credits where the venue forbids them. ISEF’s guidance says no bibliography and no acknowledgments. Check your venue’s rules and cut anything like “(Smith, 2021)” unless the page allows it.
- Vague results. “Results were significant” tells a reader nothing. Say what changed, in which direction, and by how much.
- No method. One sentence naming your design and your sample is enough.
- Copying sentences from the introduction. An abstract built from pasted opening lines has no results in it. Write it fresh from the finished paper.
- Going over the limit. Count the words the way your portal does, then use the cutting order above.
- Including information that isn’t in the paper. Check every sentence against the paper, and either remove the claim from the abstract or add it to the paper.
- Submitting an abstract that isn’t your own words. Write every sentence yourself, and read your venue’s AI policy before you touch an AI tool.
The mistake that can get an entry disqualified, not just marked down, is an abstract that isn’t your own work. JSHS lists using AI to write the abstract as a prohibited use and says plagiarism leads to disqualification; ISEF’s rules say the abstract must be in your own words and will be run through a plagiarism checker, and IJHSR bars AI from writing the abstract too.
If you want an instructor reading your draft before you submit, here are the facts on Horizon Academic Research Program. It’s online, and its 1-1 Labs format runs year-round, pairing you with a PhD or post-doc instructor. It includes four writing coach sessions and four publication specialist sessions, and it ends in a written capstone of roughly 5,000 words graded on a university rubric. You keep all intellectual property. Tuition is $6,950, and need-based aid is available for households under $75,000, which Horizon describes as very competitive. Plan for 8 to 10 hours a week. Horizon doesn’t promise publication, and its program page says a rigorous academic program cannot and should not. You can see the full details in the Horizon program catalogue.
Frequently asked questions
Most research abstracts run 150 to 300 words, but your venue sets the real number. ISEF and JSHS cap abstracts at 250 words, NHSJS asks for 200 to 250, and JEI wants under 250. Check the venue table for the full list.
An abstract summarizes the whole finished paper in one paragraph, including your results. An introduction opens the paper, explains why your question matters, and holds the findings for later.
Usually not, but your venue’s rules decide. ISEF says no bibliography and no acknowledgments in the abstract. Check your venue’s page, which the table above links. If you need to mention a source, name it in the paper itself and keep the abstract clean.
Last. Write it from the finished paper so it reports what you actually found, not what you planned to find. You can sketch a rough version early to keep your project focused, but expect to rewrite it.
It’s an abstract split into labeled sections, such as Background, Methods, Results, and Conclusions. Of the venues in our table, NHSJS is the one that spells out labeled parts, while the others take one paragraph.
Yes, with edits. Re-cut it to the new word limit, add anything the new venue requires (JSHS wants your hypothesis, for example), and remove anything it forbids. Two journals in our table, JSR and JEI, also say you can’t send the same paper to more than one journal at a time.
Cut it yourself, starting with background and hedging words. Our cutting section gives the order and shows how to cut one sentence three ways.
The paragraph that gets read first
Submitting soon with no one to read your draft? Your abstract is the paragraph a judge or editor reads first, so it deserves a second pair of eyes before the deadline. Apply to Horizon to work on your research and writing with a PhD or post-doc instructor, or join an info session to see how the program runs first.
Resources
External sources
- How to Write an ISEF Abstract, Society for Science
- Core Rules of Competition 2026-27, Junior Science and Humanities Symposium
- IJHSR Submissions, International Journal of High School Research, Terra Science and Education
- Submission Guidelines, National High School Journal of Science
- Manuscript Format & Content, Journal of Emerging Investigators
- How to Submit, Journal of Student Research, High School Edition
More from Horizon
- 15 Research Journals for High School Students
- Publishing in the International Journal of High School Research
- How to Write a Literature Review in High School
- What Is a Hypothesis? How to Write and Test One in High School
Image source: Horizon Academic Research Program




