Wang Shiyu, Hu Yiling, Wang Qi, Wu Bian, Shen Yawei, Carr Martha
Quantitative Methodology Program, Department of Educational Psychology, University of Georgia, Athens, GA, United States.
Department of Educational Information Technology, East China Normal University, Shanghai, China.
Front Psychol. 2020 Feb 26;11:305. doi: 10.3389/fpsyg.2020.00305. eCollection 2020.
This study reported on development and evaluation of a learning program that integrated a multidimensional diagnostic assessment with two different learning interventions with the aim to diagnose and improve three-dimensional mental rotation skills. The multidimensional assessment was built upon the Diagnostic Classification Model (DCM) framework that can report the binary mastery on each specific rotation skill. The two learning interventions were designed to train students to use a holistic rotation strategy and a combined analytic and holistic strategy, respectively. The program was evaluated through an experiment paired with multiple exploratory and confirmatory statistical analysis. Particularly, the recently proposed joint models for response times and response accuracy within dynamic DCM framework is applied to assess the effectiveness of the learning interventions. Compared with the traditional assessment on spatial skills, where the tests are timed and number correct is reported as a measure for test-takers' performances, the developed dynamic diagnostic assessment can provide an informative estimate of the learning trajectory for each participant in terms of the strengths and weaknesses in four fine-grained spatial rotation skills over time. Compared with an earlier study that provided initial evidence of the effectiveness of building a multidimensional diagnostic assessment with training tools, the present study improved the assessment and learning intervention design. Using both response times and response accuracy, thus current study additionally evaluated the newly developed program by investigating the effectiveness of two interventions across gender, country and rotation strategy.
本研究报告了一个学习项目的开发与评估，该项目将多维诊断评估与两种不同的学习干预相结合，旨在诊断和提高三维心理旋转技能。多维评估基于诊断分类模型（DCM）框架构建，该框架可以报告每个特定旋转技能的二元掌握情况。两种学习干预分别旨在训练学生使用整体旋转策略和分析与整体相结合的策略。该项目通过一项实验以及多项探索性和验证性统计分析进行评估。特别是，在动态DCM框架内最近提出的响应时间和响应准确性联合模型被用于评估学习干预的有效性。与传统的空间技能评估相比，传统评估是限时测试，并将正确答案数量作为考生表现的衡量标准，而开发的动态诊断评估可以根据每个参与者在四种细粒度空间旋转技能上随时间的优势和劣势，提供关于学习轨迹的信息性估计。与早期一项研究相比，早期研究为使用训练工具构建多维诊断评估的有效性提供了初步证据，本研究改进了评估和学习干预设计。通过使用响应时间和响应准确性，本研究还通过调查两种干预在性别、国家和旋转策略方面的有效性，对新开发的项目进行了评估。
