Elo difficulty
Answer correctly and your rating climbs, pulling in harder questions. Miss and it eases back. You stay near your edge instead of drilling things you already know or drowning in things you do not.
AdaptivStudy · one of three study modes
Reason
the test mode, built on adaptive testing
Questions Elo-matched to push you toward exam-level reasoning.
See what makes it differentWhat is adaptive testing?
Adaptive testing is an assessment method that adjusts question difficulty in real time, estimating a learner's ability from each answer and choosing the next question to sit near the edge of that ability.
Reason implements this with an Elo-style rating, treating difficulty like a chess rating. You start at 1.0 and the engine raises or lowers the challenge in real time across your set's questions, mapping the range 1.0 to 6.9 onto Bloom's levels of thinking.
How does Reason use adaptive testing?
Mapped to Bloom
The 1.0 to 6.9 range maps onto Bloom's taxonomy, so a higher rating is not just harder facts. It demands higher-order thinking, like analysis and evaluation, the way a real exam does.
Streak multiplier
A run of correct answers raises a streak multiplier, from 0.3 on your first correct in a row up to 0.7 at five or more. The higher it climbs, the more of the gap between your rating and the question's difficulty each correct answer captures. A separate floor still guarantees at least a small climb whenever you clear a question harder than your current rating, so progress never stalls.
Confident or unsure
On each answer you flag whether you were confident or unsure. A confident correct answer moves your rating more than an unsure one, and a confident wrong answer is penalized harder than an unsure miss, so your rating reflects genuine knowledge rather than lucky guesses.
Learning Momentum
Your Elo, normalized to a 0 to 100 scale, gives one readable number for how you are trending over a session rather than a raw rating you have to interpret.
Why does adaptive testing work?
Questions that are too easy waste your time, and questions far out of reach just teach guessing. Adaptive testing avoids both by matching difficulty to your estimated ability and updating that estimate after every answer. Measurement research shows why this is efficient: an adaptive test can reach the same precision as a fixed test while asking substantially fewer questions, because it stops spending questions on material that is clearly too easy or too hard for you (Weiss, 1982). Reason estimates skill and difficulty with an Elo rating, the same math built to rank chess players, an approach research on adaptive learning systems has found simple, robust, and effective, with accuracy comparable to more complex models (Pelánek, 2016). Keeping each question near the edge of your ability also reflects a long-standing principle in learning research: difficulty helps while you can still succeed at it, and stops helping once material is out of reach (Bjork and Bjork, 2011). And as your rating climbs, Reason moves the questions up Bloom's taxonomy toward analysis and evaluation (Krathwohl, 2002), so harder means deeper thinking, not just more obscure facts.
Sources
- Weiss (1982). Improving measurement quality and efficiency with adaptive testing. Applied Psychological Measurement, 6(4), 473-492.
- Pelánek (2016). Applications of the Elo rating system in adaptive educational systems. Computers & Education, 98, 169-179.
- Bjork and Bjork (2011). Making things hard on yourself, but in a good way: Creating desirable difficulties to enhance learning. In Psychology and the Real World, 56-64.
- Krathwohl (2002). A revision of Bloom's taxonomy: An overview. Theory Into Practice, 41(4), 212-218.
See how all three modes fit together, and everything else inside AdaptivStudy, on our about page.
At a glance
- Rating
- Elo, start 1.0, range 1.0 to 6.9
- Mapping
- Elo bands to Bloom levels
- Adapts
- In real time to each answer
- Streak
- Multiplier 0.3 to 0.7 on a run
- Confidence
- Confident or unsure gates the move
- Momentum
- Elo normalized to 0 to 100
Frequently asked questions
How does Elo rating work for quiz difficulty?
Elo is the rating system originally built to rank chess players, where beating a higher-rated opponent gains you more rating than beating a lower-rated one. Reason applies the same math to questions: your rating rises when you answer correctly and falls when you miss, so the difficulty of the next question tracks your current rating in real time.
What is Bloom's taxonomy?
Bloom's taxonomy is a framework for classifying levels of thinking. In its modern revision (Anderson and Krathwohl, 2001), the levels run from remembering and understanding facts up through applying, analyzing, evaluating, and creating. Reason maps its Elo range onto these levels, so a higher rating does not just mean harder facts, it means questions that demand higher-order thinking like analysis and evaluation.
How does adaptive difficulty know when to get harder or easier?
Each answer updates your Elo rating: correct answers push it up and pull in harder questions, incorrect answers pull it down, and how confident you said you were scales the size of the move. On a run of correct answers a streak multiplier, from 0.3 up to 0.7, lets each one capture more of the gap to the question's difficulty, keeping the challenge near the edge of what you can currently handle.
Try for freeContinue studying
Free, on your device, no account.