Title: Automated top-down heuristic assembly of a classification multistage test

Authors: Yi Zheng; Yuki Nozawa; Rongchun Zhu; Xiaohong Gao

Addresses: Mary Lou Fulton Teachers College and School of Mathematical and Statistical Science, Arizona State University, 1050 S. Forest Mall, Tempe, AZ 85281, USA ' Benesse Educational Research and Development Institute, 1-34 Ochiai, Tama City, Tokyo, 206-0033, Japan ' ACT, Inc., 500 ACT Dr., Iowa City, Iowa 52243, USA ' ACT, Inc., 500 ACT Dr., Iowa City, Iowa 52243, USA

Abstract: This paper discusses automated top-down heuristic assembly of multistage testing (MST). MST has gained increasing popularity in recent years. Automated test assembly (ATA) is one of the most important components for developing MST tests. In this study, we developed a top-down, heuristic design to assemble a classification MST test. Heuristic ATA methods have great application potential due to simplicity and feasibility. Top-down assembly is also a useful strategy for assembling short MST tests or migrating existing linear tests to MST. The full detail of the proposed assembly method is given in this paper. A simulation study was conducted to evaluate the assembly results under conditions crossed by the following five factors: 1) information target source; 2) assembly order among stages; 3) level of item overlap control; 4) number of stages; 5) stage length assignment. This paper could be used to guide practitioners and researchers on implementing this method in practice.

Keywords: multistage testing; MST; automated test assembly; heuristic test assembly; simulation; information target source; assembly order; item overlap control; stages; stage length assignment.

DOI: 10.1504/IJQRE.2016.10003201

International Journal of Quantitative Research in Education, 2016 Vol.3 No.4, pp.242 - 265

Received: 18 Oct 2015
Accepted: 12 Apr 2016

Published online: 22 Feb 2017 *

Full-text access for editors Access for subscribers Purchase this article Comment on this article