ISSN 1671-3710
CN 11-4766/R
主办:中国科学院心理研究所
出版:科学出版社

心理科学进展 ›› 2016, Vol. 24 ›› Issue (4): 654-664.doi: 10.3724/SP.J.1042.2016.00654

• 研究方法 • 上一篇    

可修改答案的计算机化自适应测验的方法

高旭亮;涂冬波;王芳;张龙;李雪莹   

  1. (江西师范大学心理学院, 南昌 330022)
  • 收稿日期:2015-05-11 出版日期:2016-04-15 发布日期:2016-04-15
  • 通讯作者: 涂冬波, E-mail: tudongbo@aliyun.com.cn
  • 基金资助:

    国家自然科学基金(31100756, 31300876, 31160203, 31360237), 江西省社会科学规划项目重点项(13JY01), 江西省教育科学规划项目(12YB088, 13YB029), 高等院校博士点基金项目(20123604120001), 江西师范大学青年英才培育资助计划, 东北师范大学应用统计教育部重点实验室开放课题(130026509)资助。

The methods of computerized adaptive testing that allows changing answers

GAO Xuliang; TU Dongbo; WANG Fang; ZHANG Long; LI Xueying   

  1. (College of Psychology, Jiangxi Normal University, Nanchang 330022, China)
  • Received:2015-05-11 Online:2016-04-15 Published:2016-04-15
  • Contact: TU Dongbo, E-mail: tudongbo@aliyun.com.cn

摘要:

与传统的纸笔测验(Paper And Pencil Based Test, P&P)相比计算机化自适应测验(Computerized Adaptive Testing, CAT)根据被试的作答反应自适应地选择题目, 它不仅缩短了测验长度, 还极大地提高了测验的准确性。然而, 目前绝大多数CAT不允许被试修改答案, 研究者主要担心修改答案会降低CAT的有效性。允许修改答案符合被试一贯的测验习惯, 修改之后的分数更能反映被试真实的水平, 从而能够进一步促进CAT在实际中的应用。现有的研究主要从三个方面提出了可修改答案CAT的控制方法:一是测验设计; 二是改进选题策略; 三是建构模型。未来的研究应进一步探讨这些方法之间的比较与结合, 以及对可修改答案认知诊断CAT (Cognitive Diagnostic CAT, CD-CAT)的研究。

关键词: 计算机化自适应测验, 检查答案, 测验效率, 修改答案

Abstract:

Based on each examinee’s answers of previous items, computerized adaptive testing (CAT) selects items and aims to measure examinee’s actual ability more efficiently and accurately than paper and pencil-based tests. While it is a natural thing for examinees to review their answers and possibly change them in paper and pencil-based tests, the same thing is less common to happen in most CATs since it could deteriorate the measurement efficiency and make tests vulnerable to manipulative test-taking strategies, as confirmed by many researchers. The absence of review opportunities on operational CATs creates a dilemma for test developers as examinees need to review and change answers during the test in order to achieve more accurate estimates of their true ability. Researchers on reviewable CAT mainly focus on three aspects, namely changing the test design, improving the item selection strategy and building models. Future studies could further explore the comparison and combination of different methods and pay more attention to review and change methods for cognitive diagnostic CAT.

Key words: computerized adaptive testing, answer review, measurement efficiency, answer change