Computer-based Adaptive Speed Tests

Research output: Contributions to collected editions/worksArticle in conference proceedingsResearchpeer-review

Authors

The assessment of a person's traits is a fundamental prob-lem in human sciences. Compared to traditional paper & pencil tests, computer based assessments not only facilitate data acquisition and processing but also allow for adaptive and personalized tests so that competency levels are assessed with fewer items. We focus on speeded tests and propose a mathematically sound framework in which latent compe-tency skills are represented by belief distributions on com-pact intervals. Our algorithm updates belief based on di-rectional feedback; adaptation rate and difficulty of the task at hand can be controlled by user-defined parameters. We provide a rigorous theoretical analysis of our approach and report on empirical results on simulated and real world data, including concentration tests and the assessment of reading skills.
Original languageEnglish
Title of host publicationThe 7th International Conference on Educational Data Mining EDM 2014 : Proceedings
EditorsJohn Stamper, Zachary Pardos, Manolis Mavrikis, Bruce M. McLaren
Number of pages4
Place of PublicationLondon
Publication date2014
Pages221-224
ISBN (print)978-0-9839525-4-1
Publication statusPublished - 2014
Externally publishedYes
Event7th International Conference on Educational Data Mining - EDM 2014 : Big Data - Big Ben - Education Data Mining for Big Impact in Learning and Teaching - Institute of Education, London, UK, London, United Kingdom
Duration: 04.07.201407.07.2014
Conference number: 7
http://www.educationaldatamining.org/EDM2014/