Spaccapanico Proietti, Giada ; Matteucci, Mariagiulia ; Mignani, Stefania ; Veldkamp, Bernard P.
(2020)
ChanceConstrained Automated Test Assembly.
[Preprint]
Full text available as:

Text(pdf) (Preprint article "ChanceConstrained Automated Test Assembly")
License: Creative Commons: Attribution 4.0 (CC BY 4.0) Download (438kB)  Preview 
Abstract
Classical automated test assembly (ATA) methods assume fixed and known parameters, an hypothesis that is not true for estimates of item response theory parameters which are key elements in test assembly. To account for uncertainty in ATA, we propose a chanceconstrained version of the MAXIMIN ATA model which allows to maximize the alphaquantile of the sampling distribution function of the test information function obtained by bootstrapping the item parameters. An heuristic based on the simulated annealing is proposed to solve the ATA model. The validity of the proposed approach is verified by simulated data and the applicability is verified by an application to real data.
Abstract