Creating IRT-Based Parallel Test Forms Using the Genetic Algorithm Method

Sun, Koun-Tem; Chen, Yu-Jen; Tsai, Shu-Yen; Cheng, Chien-Fen

doi:10.1080/08957340801926151

Cited by 29 publications

(44 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The adapted CLONALG is evaluated by comparing it with LP and GA. LP is widely used in the test construction problem and GA has been reported an efficient method for solving a large and complex test construction problem in recent years [11,14,25,30]. The Premium Solver Platform with the Large-Scale LP Solver Engine is used as the LP solver tool [8].…”

Section: The Random Test Specification For the Absolute Tif Problemmentioning

confidence: 99%

“…According to Sun's experiment design [25], a virtual item bank with 4000 items is generated for constructing various tests. The parameter and attribute values are randomly generated according to a reasonable Table 1.…”

Section: Experiments and Evaluationmentioning

confidence: 99%

“…2 Concatenate multiple strings to regarded as parallel tests [16] Because the test construction is a combinatorial optimization problem the number of combinations grows exponentially with the item bank size. Many researchers have applied novel heuristic algorithms such as the neural network technique [24], the Monte Carlo random search [5], the tabu search algorithm [15], the particle swarm optimization algorithm [32], the immune algorithm [18], and GA [11,14,25] to determine a near-optimal test in recent years. These studies focused on the single test construction.…”

Section: Related Workmentioning

confidence: 99%

See 2 more Smart Citations

Simultaneously construct IRT-based parallel tests based on an adapted CLONALG algorithm

Chang

Shiu

2011

Appl Intell

View full text Add to dashboard Cite

The simultaneously construct IRT-based (Item Response Theory) parallel tests problem requires large numbers of variables and constraints, which leads to high computational complexities and now there is no polynomial time algorithm that exists for finding the optimal solution. This article proposes an adapted CLONALG algorithm to simultaneously construct IRT-based parallel tests. Based on the CLONALG features, the proposed scheme can use a single test construction model to simultaneously construct multiple parallel tests. At the same time, it avoids the inequality problem in the sequential construction and solves the drawback of larger numbers of variables and constraints in the simultaneous construction. The serial experiments show that the proposed scheme has a lower deviation in simultaneously constructing parallel tests than the Linear Programming (LP) and the Genetic Algorithm (GA). It is also able to construct parallel tests with identical test specifications from a large item bank.

show abstract

Section: The Random Test Specification For the Absolute Tif Problemmentioning

confidence: 99%

Section: Experiments and Evaluationmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Simultaneously construct IRT-based parallel tests based on an adapted CLONALG algorithm

Chang

Shiu

2011

Appl Intell

View full text Add to dashboard Cite

show abstract

“…van der Linden (2005) gave a general discussion of GAs in the ATA framework; Verschoor (2007) considered the topic in great depth, providing specific details on applying GAs to ATA with a unidimensional IRT model. Sun, Chen, Tsai, and Cheng (2008) also applied GAs alongside a unidimensional model, using them to assemble parallel test forms. Such works all illustrate the utility of local search algorithms as tools to solve measurement problems.…”

Section: The Ga For Cdmsmentioning

confidence: 99%

Automated Test Assembly for Cognitive Diagnosis Models Using a Genetic Algorithm

Finkelman

Kim

Roussos

2009

J Educational Measurement

View full text Add to dashboard Cite

Much recent psychometric literature has focused on cognitive diagnosis models (CDMs), a promising class of instruments used to measure the strengths and weaknesses of examinees. This article introduces a genetic algorithm to perform automated test assembly alongside CDMs. The algorithm is flexible in that it can be applied whether the goal is to minimize the average number of classification errors, minimize the maximum error rate across all attributes being measured, hit a target set of error rates, or optimize any other prescribed objective function. Under multiple simulation conditions, the algorithm compared favorably with a standard method of automated test assembly, successfully finding solutions that were appropriate for each stated goal.

show abstract

“…These problems appear in multiple areas, specifically in test design (van der Linder, 2005) both for edu-cation (Porter et al, 2013;Sun et al, 2008) and psychology studies (Veldkamp, 2005).…”

Section: Introductionmentioning

confidence: 99%

Variable neighborhood search heuristics for a test assembly design problem

Pereira

Vilà

2015

Expert Systems with Applications

View full text Add to dashboard Cite

Test assembly design problems appear in the areas of psychology and education, among others. The goal of these problems is to construct one or multiple tests to evaluate some criteria. This paper studies a recent formulation of the problem known as the one-dimensional minimax bin-packing problem with bin size constraints (MINIMAX BSC). In the MINIMAX BSC, items are initially divided into groups and multiple tests need to be constructed using a single item from each group, while minimizing the difference among the tests. We first show that the problem is NP-Hard, which remained an open question. Second, we propose three different local search neighborhoods derived from the exact resolution of special cases of the problem, and combine them into a Variable Neighborhood Search (VNS) metaheuristic. Finally, we test the proposed algorithm using real-life-based instances. The results show that the algorithm is able to obtain optimal or near-optimal solutions for instances with item pools with up to 60.000 items. Consequently, the algorithm is a viable option to design large-scale tests, as well as to provide tests for online small-sized situations such as those found in e-learning platforms.

show abstract

Creating IRT-Based Parallel Test Forms Using the Genetic Algorithm Method

Cited by 29 publications

References 18 publications

Simultaneously construct IRT-based parallel tests based on an adapted CLONALG algorithm

Simultaneously construct IRT-based parallel tests based on an adapted CLONALG algorithm

Automated Test Assembly for Cognitive Diagnosis Models Using a Genetic Algorithm

Variable neighborhood search heuristics for a test assembly design problem

Contact Info

Product

Resources

About