Han Yuting, Ji Feng, Wang Pujue, Liu Hongyun
Cognitive Science and Allied Health School, Beijing Language and Culture University, Beijing, China.
Institute of Life and Health Sciences, Beijing Language and Culture University, Beijing, China.
Behav Res Methods. 2025 Apr 22;57(5):152. doi: 10.3758/s13428-025-02658-7.
With the advent of computer-based assessment (CBA), process data have assumed an increasingly pivotal role in estimating examinees' latent abilities by capturing detailed records of their response processes. This study introduces the Multidimensional sequential response model (MSRM), a novel model for assessing multiple abilities through process data in computer-based cognitive and psychological assessments. A Bayesian estimation method for the MSRM is proposed and examined through a Monte Carlo simulation study across varying conditions. The results suggest that the MSRM's parameter estimation demonstrates adequate accuracy and computational efficiency, with estimation quality improving as sample sizes and sequence lengths increase. We demonstrate the practical utility of MSRM through two empirical studies, showing that it can be effectively applied in various contexts. This methodology provides valuable insights for tailored instruction by offering detailed assessments of ability mastery across multiple dimensions, thereby supporting more targeted educational interventions.
随着基于计算机的评估（CBA）的出现，过程数据通过记录考生答题过程的详细记录，在估计考生潜在能力方面发挥着越来越关键的作用。本研究介绍了多维序列反应模型（MSRM），这是一种用于在基于计算机的认知和心理评估中通过过程数据评估多种能力的新型模型。提出了一种用于MSRM的贝叶斯估计方法，并通过在不同条件下的蒙特卡罗模拟研究进行了检验。结果表明，MSRM的参数估计具有足够的准确性和计算效率，随着样本量和序列长度的增加，估计质量会提高。我们通过两项实证研究证明了MSRM的实际效用，表明它可以有效地应用于各种情境。这种方法通过对多个维度的能力掌握情况进行详细评估，为量身定制的教学提供了有价值的见解，从而支持更有针对性的教育干预。
