Resources / IRIS Mentors

The IRIS
Mentor Handbook

A field guide for teachers taking a student from a first spark of curiosity to national finals, and, sometimes, to a stage on the other side of the world.

Edition 01 · 2026 Exstemplar · IRIS For Educators

Start reading

A note before you begin.

If you are reading this, someone in your school has walked up to you, probably shyly, probably at the end of a lunch break, and asked if you would help them do “real” science. Or perhaps you have been the one asking, quietly, for years, whether there was more you could do beyond the syllabus.

This handbook is written for you, whether it is your first student or your fiftieth. It is not a rulebook. It is a companion.

Mentoring a young researcher is not the same as teaching a class. In a class, you already know the answer. In research, neither of you does. Your job is to help a student stay in that uncertainty long enough to find something honest, not to rescue them from it. Almost every mistake first-time mentors make comes from stepping in too soon.

The IRIS National Fair is Exstemplar's pathway from that first question all the way to the Regeneron International Science and Engineering Fair (ISEF), the world's largest pre-college science competition. That is a real journey, and it is worth doing well. Everything in this handbook exists so that when your student stands next to their poster, they can answer any question they are asked because they, not you, actually did the work.

“The best mentors are the ones who ask better questions than they answer.” A judge from the IRIS Scientific Review Committee

A short note on how to use this book: sections 01–07 cover the arc from an idea to a fully designed, ethics-approved project. Sections 08–10 cover the fair itself, the poster, the abstract, the interview. Sections 11–12 are for after: the pitfalls we see year after year, and what to do once the medals are handed out. The appendices contain the rubric, sample abstracts, logbook templates, poster templates, and the ISEF-aligned forms bundle. Read this cover-to-cover once, then keep it on your desk.

What IRIS actually is.

Most teachers who hear about IRIS for the first time picture a school science exhibition, only bigger. That is not quite right.

IRIS, Initiative for Research and Innovation in STEM, is a multi-stage research competition that follows the framework of the Regeneron International Science and Engineering Fair (ISEF). It is India's official pathway to ISEF, and the projects that reach the National Fair are held to the same categories, ethics rules, judging structure, and safety review as any ISEF project anywhere in the world.

Practically, that means IRIS is not an exhibition of models. It is a research fair. Judges are not looking for the neatest volcano or the most elaborate working prototype. They are looking for a well-framed question, a defensible method, honest data, and a student who can explain their own reasoning under pressure.

The stages of the journey

A typical IRIS project runs across roughly six to twelve months, in this shape:

School registration & project selection. Students propose a research question with a mentor's support. The school shortlists projects to submit.
Regional / district evaluation. Shortlisted projects go through a preliminary review, often based on submitted abstracts and research plans.
The IRIS National Fair. Selected projects present in person: poster, logbook, and a live judging interview.
ISEF selection. A subset of national finalists represent India at ISEF in the United States, with additional grand and category awards.

The IRIS JourneyRoughly 6 to 12 months from school registration to national finals.

Timelines shift year to year. Always check iris.exstemplar.com for the current cycle's dates and portal. Everything else in this handbook, categories, rubric, ethics rules, is anchored to the ISEF framework and does not change often.

Finding a research question.

This is where almost every first-time mentor tries to help too hard.

A student walks in with “I want to do something on cancer” or “I want to build an app for farmers” and the mentor, wanting to be encouraging, says “yes, great, let's do that.” Six weeks later they are stuck, because “cancer” is not a research question and “an app for farmers” is not a hypothesis.

A research question is a specific, testable, honest thing you do not yet know the answer to. The gap between a topic and a question is where the entire project lives.

The four filters

A good IRIS question passes four filters. Ask your student each of these, and do not move on until you both have honest answers.

Is it specific? “Does music affect plants?” is not specific. “Does 4 hours of daily 60 Hz sinusoidal sound exposure at 70 dB change the germination rate of Vigna radiata compared to silence?” is specific.
Is it testable with what we have? You have a school lab, maybe a college lab willing to help, a small budget, and eight to sixteen weeks. Can you actually collect data on this in that world?
Is the answer unknown to you? If a student can predict the answer before the experiment, the experiment is a demonstration, not research. Real questions have real uncertainty.
Does anyone care what the answer is? Not “is it globally important,” but “can we explain in one sentence why this matters to a farmer, a doctor, a policymaker, or the next student”? If not, the question is probably too small or too disconnected.

From topic to questionEach filter narrows the question. Any filter can send you back to the topic to try again.

The resultSpecific, testable, genuinely unknown, and clearly matters to someone.

Refining a topic into a question

Here is a small ritual that works well. Take an hour with your student. Ask them to write down their vague topic in the middle of a page. Then ask “why does this matter?” five times. Each answer moves them up towards significance. Then ask “how could we test that?” five times. Each answer moves them down towards a method. Somewhere in the middle sits the actual research question.

Case in point

Topic: Plastic pollution.

Why does it matter? → It is choking rivers → the plastic breaks into microplastics → microplastics enter drinking water → nobody in our town is testing for them → local health impact is unknown.

How could we test that? → Collect water at five points along the Yamuna in Gurgaon → filter → count particles under a stereo microscope → compare to distance from an industrial discharge point.

Research question: Does the concentration of microplastic particles in the Yamuna correlate with proximity to identified industrial discharge points across a 20‑km stretch through Gurgaon?

Notice that the question is now doable in a school setting, is specific enough to be judged, and is honestly unknown, nobody has counted particles at those exact five points before.

Engineering projects need a design goal, not a hypothesis

If your student is building something, a device, a piece of software, a low-cost diagnostic, the shape of the question changes. Instead of a hypothesis, they need a design goal with measurable criteria. “I will build an app” is not a goal. “I will build an Android app that lets a rural pharmacist verify a drug's authenticity in under thirty seconds using only the phone camera, tested against a bank of 200 real and counterfeit samples with a target >90% accuracy” is a goal.

Engineering IRIS projects are judged on whether the built thing meets the design criteria and whether the student can explain each design decision. Make the criteria explicit up front.

Reading the field.

Before a single measurement is taken, your student needs to know what other people have already found out. This is the literature review. In India, most school teachers never formally learned to do one, not because they are not capable, but because school curricula never asked for it. This section is for you and your student, together.

Why bother?

Three reasons. First, so the student does not repeat, badly, an experiment somebody else already did well. Second, so they know what “normal” results look like in the field, when a judge asks “is a 40% germination rate high or low?” the student needs to know. Third, so they can situate their small piece of work inside a bigger conversation. That framing is a huge fraction of what makes a project feel like research rather than a school activity.

Where to look

Most Indian schools do not have paid database subscriptions, and they do not need them. Free sources are enough:

Google Scholar. Start here. Search with keywords, then filter by year (last 5–10 years for most fields).
PubMed. Essential for anything life-sciences or medical.
arXiv. Physics, mathematics, computer science, and increasingly biology preprints.
PLOS ONE, eLife, Frontiers. Open-access journals with strong peer review.
Semantic Scholar. Uses AI to surface influential papers and citation trails; often finds things Google Scholar misses.
Government portals. ICMR, DBT, MoEF&CC, and the census for Indian-context data.

If a paper is paywalled, check the authors' personal or university pages, researchers routinely post preprints. If still stuck, a polite email to the author asking for a PDF works surprisingly often. This is a good, teachable moment in scientific culture.

How to read a paper

Nobody, not even a professor, reads a paper front to back on the first pass. Teach your student the three-pass method:

Pass one (5 minutes): title, abstract, figures, conclusion. Is this paper relevant at all?
Pass two (20 minutes): intro and discussion. What did they set out to do and what did they find?
Pass three (as long as it takes): methods and results. Only if this paper directly informs your student's design.

Have them keep a simple reading log: one row per paper, columns for citation, one-sentence summary, and “what this changes about our project.” Ten rows is a solid literature review for a school-level project.

Citations

Pick a style early, APA or a numbered style like Vancouver are both fine, and stick to it. Use a free reference manager: Zotero is the standard, works on any browser, and imports citations with one click. Learning Zotero in an afternoon saves your student twenty hours of manual bibliography formatting later.

A common mistake

Do not let your student cite Wikipedia in the final references. It is a perfectly good starting point for background, every professional researcher uses it, but it is not a primary source. The trick is to follow Wikipedia's citations down to the actual studies and read those.

Designing the experiment.

A well-designed experiment is a boring experiment. There are no surprises in the method, only in the results. The moment you find yourself saying “let's just try it and see,” stop. Design first, measure second.

Variables

Every experiment has three kinds of variables. Make sure your student can name each one for their project, out loud, without notes.

Independent variable. The one thing you deliberately change. There should be exactly one.
Dependent variable. What you measure to see the effect. Ideally one primary measurement plus one or two secondary ones.
Controlled variables. Everything else you hold constant so that any change in the dependent variable can be blamed only on the independent one. This list is often ten to twenty items long.

Anatomy of an experimentOne independent variable, one primary dependent variable, everything else controlled.

Controls

A control is a version of your experiment that isolates the effect of the independent variable. If you are testing whether a plant grows better with a fertiliser, your control is a plant with everything the same, same soil, same water, same light, same species, same age, but no fertiliser. Without a control, you cannot know whether the fertiliser did anything at all, or whether the plants just grew in seven weeks anyway.

For engineering projects, the analogue is a baseline: what does performance look like without your invention? If your new sensor detects a disease with 92% accuracy, that is only impressive if you also know that the existing standard achieves 78%.

Sample size and repetition

One data point is an anecdote. Three is a suggestion. Ten begins to be evidence. Whatever your student is measuring, they need enough repetitions to see through the noise. For biology-type experiments, a sample size of at least 10 per condition, with three biological replicates, is a defensible minimum at school level.

If sample size is tight because of time or budget, be honest about it in the discussion. Judges deeply respect students who say “n = 5 per group is a limitation of this study” and deeply distrust those who pretend it is not.

Blinding and randomisation

Whenever the person measuring the outcome could be biased, and this is more often than you think, especially in behavioural, plant, and taste-test experiments, blind the measurement. If your student is scoring how “green” a leaf is, they should not know which fertiliser condition each leaf came from. A friend with a coding sheet is enough.

Similarly, when assigning subjects (plants, mice, participants) to conditions, use randomisation. A random number generator on a phone is fine. Do not assign “the first ten to Group A”, that quietly introduces order effects.

Pilot before you commit

Before running the full experiment, run one condition, once, at half the scale. This will surface every practical problem, the thermometer that reads wrong, the software that crashes at midnight, the reagent you did not order in time, while it is still cheap to fix. A day spent piloting saves a week spent salvaging.

The logbook.

The logbook is the single most important physical object in an IRIS project. Judges look at it, ethics reviewers may request it, and years later it will be the only record of what actually happened. Treat it accordingly.

What a good logbook looks like

A good logbook is a bound, page-numbered notebook, not loose sheets, not a Google Doc, not photos on a phone. Bound because judges want to see continuity; page-numbered because gaps become obvious. Every entry should have at minimum:

Date and time.
A one-line summary of what was done that day.
The raw observations, in whatever form: numbers, sketches, taped-in printouts.
A short reflection: what worked, what did not, what to change next time.

Handwritten is fine. Mixed handwriting and printed-and-taped-in tables is fine. Mistakes crossed out with a single line and initialled is fine, and correct scientific practice. Whited-out mistakes are a red flag. So are perfectly clean logbooks with no crossings-out. Real research is messy and the record should show it.

Entries the student should never skip

The first entry. The research question, exactly as it stands on Day 1, before it has been polished.
The pilot day. What was tried, what broke, what changed as a result.
Every deviation from the plan. If the temperature was 3 °C off, if a subject dropped out, if a reagent was substituted, log it, and log why.
Meetings with the mentor. Date, what was discussed, what was decided. This is also useful evidence of appropriate mentor involvement rather than mentor-does-the-work.
Ethics or safety milestones. When each form was signed, by whom.

Mentor discipline

Look at the logbook every week. Not to grade it, just to see it exists and is being kept. Students who know their mentor will glance at Wednesday's entry on Thursday are much less likely to reconstruct three weeks of “work” the night before the fair. That reconstruction is easy for a judge to spot and it will sink an otherwise strong project.

Ethics & safety.

This is the section most first-time mentors want to skim. Please do not.

Every IRIS project must comply with the ISEF International Rules for Pre-College Science Research. These rules exist because the fair sits inside a global research community that takes ethics seriously, and because a well-run ethics process protects your student, your school, and the people or organisms they work with. Skipping a form does not just risk disqualification, it teaches the student the wrong lesson about what science is.

The forms

Some forms are required for every project; others depend on what the student is doing. The most common ones, in the order you will encounter them:

Form 1:Adult Sponsor Checklist. Signed by you, the mentor, before any experimentation begins. Your commitment that the plan is safe and appropriate.
Form 1A:Student Checklist & Research Plan. The student's written research proposal, submitted before starting.
Form 1B:Approval Form. Signed by the student, parent/guardian, and adult sponsor.
Form 1C:Regulated Research Institutional/Industrial Setting. If any work happens in a college, hospital, or company lab.
Form 2:Qualified Scientist. Required for certain regulated research (human subjects, vertebrates, hazardous agents).
Form 3:Risk Assessment. For anything involving hazardous chemicals, activities, or devices.
Form 4:Human Participants. Any study involving people, surveys, taste tests, behavioural studies, exercise studies. Requires informed consent (or assent) and IRB-equivalent review.
Form 5A / 5B:Vertebrate Animals. Any experiment on vertebrates. School-based vertebrate research is heavily restricted, read these rules carefully before promising a student they can do it.
Form 6A:Potentially Hazardous Biological Agents. Microorganisms, recombinant DNA, human/animal tissue and body fluids.
Form 6B:Human and Vertebrate Animal Tissue. Required whenever tissue is used.
Form 7:Continuation Projects. If this year's work builds on last year's project.

Some of these forms must be signed before experimentation begins. This is not a technicality. If Form 4 is signed after a human-subjects study is already done, the study is invalid and the project is disqualified, even if the study was harmless. Approval is prospective, not retrospective.

Practical guidance

If it involves people, even a survey among classmates, it needs Form 4 and prior review.
If it involves vertebrates, even goldfish, assume it needs Forms 5A/5B and review; assume it will be restricted.
If it involves microorganisms grown in the school, even from a swab of a doorknob, it needs Form 6A. Cultures from unknown environmental sources must be handled at BSL‑2 or done in a properly equipped institution.
When in doubt, assume a form is needed. The cost of an extra signature is nothing. The cost of disqualification is a broken teenager.

A hard truth

The most common reason talented Indian IRIS projects get held back at ISEF is an ethics form problem, a Form 4 signed after the survey ran, a Form 5A skipped for “just observing” ants in an aquarium, an environmental sample cultured on a classroom bench. The science is often good; the paperwork sinks it. Spend Week 2 on the forms as if it were as important as the experiment. It is.

For the current, authoritative version of every form, see the Society for Science ISEF rules and forms page at societyforscience.org/isef/international-rules. IRIS uses these forms directly.

Data, analysis, and negative results.

The experiment ends. The data sit there. This is where students, especially bright ones, are most tempted to shape the story rather than let the story shape itself.

Recording data

Raw data lives in two places: the logbook (as they were observed) and a spreadsheet (for analysis). Never touch the raw file; always work on a copy. Every spreadsheet needs a metadata row at the top describing what the columns mean, the units, and the date of collection. Six months from now, at ISEF, your student will thank themselves.

Basic analysis

For most school-level projects, four kinds of analysis cover almost everything:

Descriptive statistics. Mean, median, standard deviation, range. Report these for every group.
A comparison test. t‑test for two groups; ANOVA for three or more; chi‑square for counts. Free tools like JASP or the Google Sheets “=TTEST” function are enough.
A correlation or regression. If two things vary together, quantify how strongly.
Error bars on every chart. Standard deviation or standard error, but label which.

Your student does not need to have taken a stats course. They do need to know what test they used and, in one sentence, why. “I used a two-sample t‑test because I was comparing the means of two independent groups and my sample sizes were small.” If they can say that, they can answer 90% of stats questions from a judge.

Negative results are results

Sometimes the fertiliser does nothing. Sometimes the new sensor performs worse than the old one. Sometimes the survey shows no effect. Every year we watch bright students panic and start p‑hacking, slicing the data six different ways until something looks significant.

Do not let them. A negative result, honestly reported, is scientifically valuable. It also demonstrates something rarer than a positive result: integrity. Judges know this. A student who stands next to a poster titled “No significant difference was observed, and here is what that tells us” and can defend the design is stronger than one waving a colourful bar chart with a manufactured p‑value.

A useful phrase

Teach your student to say, out loud: “These are the limitations of the study.” Every mature researcher says this. Every honest project has them. Naming them does not weaken the work, it is what qualifies it as work.

The poster and the abstract.

The poster is not the project. The poster is a map to help a judge navigate the project in the twelve minutes they will spend in front of it. Design accordingly.

The abstract

Write the abstract last, when everything else is done. It is 250 words, roughly, and it should contain, in this order:

One sentence of context: why this matters.
One sentence of gap: what was not known.
One sentence of question: what you set out to find.
Two to three sentences of method.
Two to three sentences of result, with actual numbers.
One sentence of interpretation and significance.

Read good published abstracts from PLOS ONE or Nature to internalise the rhythm. A weak abstract is often a sign the underlying story is unclear, use the abstract-writing exercise to force clarity, not just describe it.

The poster

ISEF-style posters are usually a tri-fold display (roughly 122 cm × 76 cm total when open). Standard sections, roughly left to right:

Title, name, school, category.
Introduction / background (short).
Question / hypothesis.
Methods, with a clear visual: a flow diagram is worth 200 words.
Results, with two or three well-chosen charts.
Discussion & conclusion.
Limitations & next steps.
References (small type is fine).

Design principles that judges reward, quietly:

Every chart has axes labelled with units. A shocking number of posters fail this.
The eye can travel the poster in a Z‑shape. Left-to-right, top-to-bottom, top of next column.
Text sizes: title 100‑pt, section headers 40‑pt, body 24‑pt. Readable from a metre away.
White space is not wasted space. A crowded poster reads as an anxious project.
Colour serves data, not decoration. Two data colours plus a neutral background is almost always right.

ISEF tri-fold poster layoutRoughly 122 cm wide by 76 cm tall when open. Body text at 24‑pt so it reads from a metre away.

Templates for tri-fold and A0 posters, editable in Google Slides and PowerPoint, are in the appendices.

The judging interview.

The interview is where the fair is decided. In the ISEF rubric, it is worth 25 points, a full quarter of the score, and more than any single other component. Your student can have a perfect poster and lose here. They can also have a shaky poster and win here.

What judges are actually doing

A judge's job in twelve minutes is to work out whether the student in front of them understands their own project. That is it. Everything else, the poster, the abstract, the logbook, is corroborating evidence. If the student can explain their design choices, defend their method, and reason about their data in real time, the rest falls into place.

The 3-2-1 answer

Teach your student a small trick: for any question, structure the answer in three parts. Three sentences of direct answer. Two sentences of evidence from the project. One sentence of what it means. Practise this out loud, ten times, on ten different questions. It becomes automatic and it prevents rambling.

Questions to rehearse

You already know most of the questions judges will ask. Drill them:

Tell me about your project in two minutes.
Why did you pick this question?
Why this method and not another?
What is your independent variable? Dependent? Controlled?
Why is your sample size what it is?
What is the biggest limitation of your study?
What surprised you?
If you had six more months, what would you do next?
Who helped you, and with what?

That last question matters more than students realise. Judges ask it to distinguish student work from mentor work. The right answer is honest and specific: “Dr Kaur at the college let me use her stereo microscope on three Saturdays and taught me how to prepare wet mounts. She did not touch the samples. My teacher, Mr Sharma, helped me refine the research question in the first two weeks and reviewed my logbook every Thursday.” This is exactly the kind of collaboration ISEF respects. Pretending you did it all alone is what triggers scepticism.

Rehearsal, not scripting

Do practice interviews with three different people your student does not know well, another teacher, a college student, a parent from another class. Ask them to be a little sceptical. If your student cannot handle warm scepticism from friendly strangers, they will freeze in front of a stranger with a clipboard. Better to freeze in the rehearsal.

Common pitfalls.

Every year, we watch the same handful of things go wrong. Not exotic mistakes, ordinary ones, made under pressure, by good people. Reading through this list once in September saves a lot of pain in March.

Mentor writes the abstract.

The abstract sounds subtly wrong: too polished, sentences the student cannot explain. Judges notice within one question. Let the student draft it, redraft it, and keep every version in the logbook. Edit for clarity, not for voice.

Ethics forms signed after the fact.

The single most common cause of disqualification at national and international level. Any human-subjects, vertebrate, or hazardous-agent work needs prior approval. No exceptions. Look at the calendar, not the excitement.

Wrong category.

Strong biology-adjacent projects placed under “Biomedical Sciences” land in a room of clinical trials and look thin. Read every category description before choosing. If in doubt between two, pick the one where the project's core method fits, not the one that sounds more impressive.

Sample size of three.

Three data points do not survive a t‑test. If time is short, cut variables rather than replicates. It is better to answer one question well than three questions badly.

No control condition.

Especially common in engineering and applied projects. “My app got 85% user satisfaction” means nothing without a baseline. Judges will ask; have the number.

The logbook is written in one sitting.

Ink from the same pen. Uniform handwriting. All entries oddly self-consistent. Judges glance at logbooks. They know. Keep the logbook in real time or not at all, a genuine, patchy record beats a fabricated tidy one every time.

Buying an off-the-shelf kit and presenting the assembly as research.

Ordering a “water purity testing kit” from Amazon and reading its values is not research. Building the sensor yourself and calibrating it against a known standard is. If the student did not design the measurement, they did not do the science.

Overselling significance.

“My project could solve India's water crisis” is the sentence most likely to make a judge stop trusting the student. Small, honest, well-defined contributions earn respect. Grandiose framing does the opposite.

Practising the poster instead of the project.

Students memorise their opening pitch and then freeze on the second question, because they never rehearsed being pushed off the script. Practise the follow-ups, not the pitch.

Peaking too early.

The state or district round is not the final. Watch your student's energy. The best projects treat the National Fair as a working presentation of an ongoing piece of work, not the closing scene of a play.

After the fair.

The medals get handed out on a Sunday afternoon. On Monday morning, your student is back in your class. What happens now matters more than most mentors realise.

If they win

Say congratulations, and then very quickly steer the conversation back to the science. Winning is a signal, not a finish line. The best follow-up questions are “what would you do differently?” and “what is the next question this raises?” Both are things ISEF judges will ask if the student advances. Both keep the student oriented towards the work rather than the plaque.

If your student is selected for ISEF, you are about to enter a much larger process, visas, additional ethics review, a bigger poster, sometimes a research paper. Reach out to the IRIS team at Exstemplar early. The support system exists and is used to first-time-abroad students and their teachers.

If they do not win

Say something honest before you say something encouraging. “The result is disappointing and it makes sense to feel that way” lands better than “great effort, we'll get them next year.” Students see through cheerful deflection instantly.

Then, when the moment has passed, ask two things. First: what did they learn about doing research that they will carry into whatever they do next? Second: is there a version of this project that could go somewhere else, a school journal, a state fair, a paper, a college application essay? Very few IRIS projects are one-off. Most of them can travel.

Retire the project properly

Archive the logbook, the poster, the abstract, the data, and the ethics forms in one place at the school. Digital scans are fine. Two years from now the next student in your school will benefit hugely from being able to see a complete, real project as a reference. This is how you build institutional memory. This is how the second cohort at your school is always stronger than the first.

Look after yourself

Mentoring a research project on top of a full teaching load is expensive, emotionally and practically. If you got to the end of a cycle with a student, you did something rare and difficult. Take a week. Then decide whether to do it again next year, ideally with a colleague from another department, because the most durable school science-fair traditions are the ones held up by two or three teachers, not one.

“You are not raising a winner. You are raising a person who knows how to sit with a question long enough to find out.” Exstemplar Scientific Review Committee

Appendices

Reference & templates.

The main handbook is deliberately short. Everything below is a separate resource, download only what you need for the stage you are at.

The ISEF-aligned judging rubric

This is what judges score against, out of 100. Show it to your student in Week 1 and again the night before the fair.

ISEF Judging RubricWeightings are stable across years. Practise the interview accordingly.

Component	Points
Research Question Clear, focused, testable; addresses a real gap.	10
Design & Methodology Well-designed plan, appropriate controls, variables identified.	15
Execution: Data Collection, Analysis & Interpretation Systematic data collection, appropriate analysis, honest interpretation.	20
Creativity Original thought in question, method, or interpretation.	20
Presentation: Poster Clear layout, logical flow, self-standing when read alone.	10
Presentation: Interview Depth of understanding, response to questions, ownership of the work.	25
Total	100

Note the weighting. The interview alone is worth two and a half times the poster. Rehearse accordingly.

Authoritative external sources

Society for Science: ISEF International Rules & Guidelines. The current-year rulebook, forms, and category descriptions. Read the version for the year your student is competing.
iris.exstemplar.com. Current cycle dates, submission portal, and IRIS-specific announcements.
contact@exstemplar.com. If you get stuck, ethics questions, category questions, logistics, write to us. Real humans read this inbox.

Take the handbook with you.

Print it, share it with a colleague, or keep it as a bookmark. It is free to use in any school, and we will update it every year as the ISEF rules evolve.

All Resources