Toward Intelligent Assessment: An Integration of Constructed Response Testing, Artificial Intelligence, and Model-Based Measurement
- Author(s):
- Bennett, Randy Elliot
- Publication Year:
- 1990
- Report Number:
- RR-90-05
- Source:
- ETS Research Report
- Document Type:
- Report
- Page Count:
- 44
- Subject/Key Words:
- Artificial Intelligence, Constructed-Response Tests, GIDE, HOST (Model), Hybrid (Model), MicroPROUST, Models, Test Theory
Abstract
In this report a new assessment conception is described that integrates constructed-response testing, artificial intelligence, and model-based measurement. The conception incorporates complex constructed-response items for their potential to increase the validity, instructional utility, and credibility of standardized tests. Artificial intelligence methods are invoked to produce item-level partial-credit scores and diagnostic analyses similar to those of a human expert. Finally, cognitively-grounded measurement models provide diagnostic statements based on commonalities in performance across items. Progress toward achieving this conception is examined, with emphasis on two automated scoring programs. Potential applications of intelligent assessment are discussed. (44pp.)
Read More
- Request Copy (specify title and report number, if any)
- http://dx.doi.org/10.1002/j.2333-8504.1990.tb01342.x