Hiring glossary
What is adaptive testing?
Adaptive testing is a way of delivering a test in which each question is chosen by the computer based on how the test taker answered the previous ones, so the difficulty adjusts to the person’s level.
A correct answer usually leads to a harder question and a wrong one to an easier question. Behind it sits a bank of items with calibrated difficulty, usually built with item response theory. The score estimates the level of difficulty a person handles reliably, which is why two people can answer the same number of questions correctly and receive different scores.
The advantage is precision with fewer questions, since nobody spends long on items far too easy or too hard for them. Candidates also see different question sets, which makes shared answers less useful. The cost is that calibrating the item bank takes a lot of response data, so adaptive formats are most common in cognitive and knowledge tests.
Also called: computerised adaptive testing, CAT.
Example
Two candidates take an adaptive numerical reasoning test and both start on a medium question. One answers correctly and moves to harder items; the other misses it and gets an easier one. Each finishes with a score on the same scale, from different questions.
Go deeper
Related terms
Cohesyve · Skill assessments for hiring
Add a test of the real work to your process
Cohesyve takes a job description, finds candidates, reaches out to them and runs an AI job simulation of the real work, then returns a ranked shortlist.
1,500+
assessments completed
50%
faster time-to-hire
5 min
from JD to assessment
No credit card · 10 free candidates · Plans sized to your hiring volume