Assessment science
Most tools tell you who passed.
We tell you if the exam was any good.
Every Foresight item is tagged by NREMT® domain, format, and clinical-judgment step. The moment your students start answering, you can see which questions are doing real work, which are quietly broken, and which skills the class hasn't locked in yet — long before the real exam finds out for you.
What we measure
Real psychometrics, in plain English.
You don't need a stats degree to read these. Each card below starts with a question you already ask about your exam — the psychometric term is just the small label above it.
Difficulty (p-value)
How hard was this question, really?
We track the share of students who get each item right. Too easy and it's a freebie; too hard and it may be confusing rather than rigorous — either way it isn't telling you much.
Example: 96% of your medics nailed the scene-safety item — a gimme that won't separate a strong class from a weak one. The rhythm item only 31% got right is doing real work.
Discrimination (point-biserial)
Are the right students getting it right?
A good question is one your strongest students tend to get right and your struggling students tend to miss. When that pattern flips, the question is probably broken or has the wrong answer marked.
Example: Foresight flags a 12-lead item where your top performers all chose B but the key says C — a mis-keyed question caught before it unfairly tanks a grade, not after a student complains.
Distractor analysis
Which wrong answers are actually working?
For every question we show how many students each answer choice pulled. A wrong answer nobody picks is dead weight; one that half the class picks reveals a real misconception you can teach to.
Example: On an anaphylaxis item, 40% chose 'diphenhydramine first' instead of epinephrine. That distractor isn't just wrong — it's a teaching moment your class clearly needs.
Field-test (pilot) items
Try a question before it counts.
Drop a new question into a real exam as a field-test item — students answer it like any other, but it doesn't affect their grade. You gather real performance data and vet the question before you ever rely on it.
Example: Field-test a new ordered-response stroke scenario across two cohorts; once 30+ students have answered, Foresight shows whether it's fair and sharp — then you make it count. (The same way well-run standardized exams pilot new items before scoring them.)
Domain / blueprint coverage
Cover the blueprint, not just your favorites.
Every item is tagged to its NREMT content domain, so you can see your exam — and your class's performance — mapped against the same blueprint the national exam is built on.
Example: The domain heatmap shows your cohort is solid on Trauma but consistently soft on Medical and Cardiology — so you know where to spend next week's lab time before the real exam finds the gap.
Clinical Judgment
Measure thinking, not just recall.
Foresight tags items to the six clinical-judgment steps a provider moves through on a real call — Recognize Cues, Analyze Cues, Define Hypothesis, Generate Solutions, Take Action, Evaluation — and charts where your class is strong or stalls.
Example: The radar shows your students Recognize Cues well but stumble at Generate Solutions — they spot the sick patient but freeze on the next move. (For AEMT and Paramedic, Clinical Judgment is a scored NREMT area; for EMT it runs through task-based items. Foresight uses these six steps as a teaching lens at every level — it does not claim the NREMT scores all six.)
Plain-English glossary
Difficulty (p-value): the share of students who got a question right. High means easy.
Discrimination (point-biserial): whether your strongest students are the ones getting a question right. A negative value means it is likely broken or mis-keyed.
Distractor: a wrong answer choice. Good ones catch real misconceptions; useless ones nobody picks.
Field-test (pilot) item: a question students answer for data only — it does not count toward their grade until you have vetted it.
NREMT domain: the blueprint areas the national exam is built from, tracked per level. EMT (2025 blueprint): Scene Size-up and Safety, Primary Assessment, Secondary Assessment, Patient Treatment and Transport, and Operations. AEMT and Paramedic: Airway/Respiration/Ventilation, Cardiology/Resuscitation, Trauma, Medical/OB-GYN, EMS Operations, and Clinical Judgment.
Clinical-judgment steps: the six-step path from sizing up a patient to acting and re-evaluating — used here as a lens on clinical reasoning.
Clinical judgment
The six steps a provider runs on every call.
Foresight tags items to each step of clinical reasoning, so you can see where your class moves smoothly and where it stalls — not just whether the final answer was right.
- 01
Recognize Cues
Notice what matters in the scene and the patient.
- 02
Analyze Cues
Connect those findings into a clinical picture.
- 03
Define Hypothesis
Name what you think is going on.
- 04
Generate Solutions
Lay out the interventions on the table.
- 05
Take Action
Choose and carry out the right one.
- 06
Evaluation
Check whether it worked, then adjust.
For AEMT and Paramedic, Clinical Judgment is a scored NREMT area; for EMT it runs through task-based items. Foresight uses these six steps as a teaching lens at every level — it does not claim the NREMT scores all six.
This is the layer graduate students learn to build into.
Knowing what makes an item work is one thing. Teaching the next generation of EMS educators to write items that hold up to it is the next. That's what the Graduate Programs add-on is for — a separate module for degree-granting EMS programs.
See the Graduate Programs add-onSee your own cohort's numbers.
Request a walkthrough and we'll show you the difficulty, discrimination, and clinical-judgment views running on realistic sample data.
Request a walkthrough