Search
Program Calendar
Browse By Day
Browse By Time
Browse By Person
Browse By Room
Browse By Unit
Browse By Session Type
Search Tips
Annual Meeting Registraion, Housing and Travel
Personal Schedule
Sign In
The utilization of response time has come to the fore with the increase of interest in recent years as testing technology develops. One important application is to detect examinees' aberrant testing behaviors, which impacts the accuracy of examinees' score and undermines the test validity. For example, the popularity of computer-based testing brings more exposed items, which could cause item preknowledge issue. Consequently, this results in higher probability of answering certain difficult items correctly in an extremely short time. This paper introduces an anomaly detection approach with four model-based nonparametric indices, utilizing both response accuracy and time information. Results show that the proposed method not only performs successful in detection rate and false alarm rate, but also presents computational efficiency.