Online Program Home
  My Program

All Times EDT

Abstract Details

Activity Number: 424 - Recent Advances in Educational and Psychological Data Analysis
Type: Invited
Date/Time: Thursday, August 6, 2020 : 10:00 AM to 11:50 AM
Sponsor: Journal of Educational and Behavioral Statistics
Abstract #309252
Title: Compound Sequential Detection of Compromised Items
Author(s): Xiaoou Li* and Yunxiao Chen
Companies: University of Minnesota and London school of economics and political science
Keywords: test security; sequential analysis; change-point detection; continuous testing; compound sequential decisions
Abstract:

We propose a compound sequential change detection method for monitoring the item pool in an educational test. Taking a Bayesian change point model, we propose a compound detection method that maximizes the utilization of the items while controlling the proportion of compromised items among the ones being used at all the time points. We provide theoretical results regarding the uniform optimality of the proposed approach among all compound sequential detection rules and illustrate its use and performance through numerical examples.


Authors who are presenting talks have a * after their name.

Back to the full JSM 2020 program