What equating processes does the AP Program use?
The AP Program uses two different types of equating processes:
- In common-item equating, a set of common test questions is used across different versions of the exam administered to different student populations.
- In random groups equating, the different versions of the exam are administered to groups of examinees that are statistically equivalent but randomly assigned.
In both of these equating designs, the statistical procedures adjust for differences in difficulty among exam versions that are built to be similar in difficulty and content. This allows the AP Program to establish “cut scores” for multiple versions of the exam that result in comparable AP Exam scores of 5, 4, 3, 2, and 1. While the cut scores may differ by a few points depending on the difficulty of the exam versions, the exam scores will represent the same performance standards (and hence, a student’s score will represent the same level of achievement) no matter which version of an exam a student takes.