Adaptiertest

Adaptiertest refers to an assessment approach in which the items presented to a test taker are chosen adaptively based on the person’s previous responses. Instead of a fixed sequence of questions, an algorithm estimates the test taker’s ability and selects subsequent items of appropriate difficulty. The goal is to achieve comparable measurement precision with a shorter test duration. Adaptiertests are most commonly implemented as computer-based tests, collectively known as computer adaptive testing (CAT).

The underlying methodology typically relies on item response theory (IRT) or Bayesian approaches. Items in a calibrated item bank have known difficulty and discrimination parameters, allowing the test to update an ability estimate after each response. In CAT, the next item is selected to maximize information around the current ability estimate, while in other adaptive formats, multistage designs may present short sets of items whose composition depends on earlier results. These methods aim to balance precision, test security, and content coverage.

Applications and examples vary, but adaptiertests are widely used in education, professional licensure, and large-scale assessments. Standardized tests such as the GRE and GMAT employ adaptive formats to efficiently measure verbal and quantitative abilities. They are also used in some career qualification exams and digital learning platforms to tailor item difficulty to individual learners.

Limitations and considerations include the complexity of item bank construction and calibration, ensuring test security, and addressing fairness and accessibility concerns. Critics point to potential biases, differential item functioning across populations, and the need for transparency about scoring and exposure control.