• 主题
高级搜索  >
历史搜索 清空历史
首页> 外文学位 >Diagnosing examinees' attributes-mastery using the Bayesian inference for binomial proportion: A new method for cognitive diagnostic assessment.
【24h】

Diagnosing examinees' attributes-mastery using the Bayesian inference for binomial proportion: A new method for cognitive diagnostic assessment.

机译：使用二项式比例的贝叶斯推断来诊断应试者的属性掌握：一种用于认知诊断评估的新方法。

获取原文
获取原文并翻译 | 示例
页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

Cognitive diagnostic assessment (CDA) is a new theoretical framework for psychological and educational testing that is designed to provide detailed information about examinees' strengths and weaknesses in specific knowledge structures and processing skills. During the last three decades, more than a dozen psychometric models have been developed for CDA, which are also called cognitive diagnosis models (CDM). Although they have successfully provided useful diagnostic information about the examinee, most CDMs are complex due to a large number of parameters in proportion to the number of skills (attributes) to be measured in an item. The large number of parameters causes heavy computational demands for the estimation. Also, a variety of specific software applications is needed depending on the chosen models.;Purpose of this study was to propose a simple and effective method for CDA without heavy computational demand using a user-friendly software application. Bayesian inference for binomial proportion (BIBP) was applied to CDA because of the following fact: When we have binomial observations such as item responses (right/wrong), using a beta distribution as a prior of a parameter to estimate (i.e., attribute-mastery probability) makes it very simple to find the beta posterior of the parameter without any integration. The application of BIBP to CDA can be flexible depending on the test item-attribute design and examinees' attribute-mastery patterns. In this study, effective ways of applying the BIBP method was explored using real data studies and simulation studies. Also, other preexisting diagnosis models such as DINA and LCDM were compared to the BIBP method in their diagnosis results.;In real data studies, the BIBP method was applied to a test data using two different item designs: four and ten attributes. Also, the BIBP method was compared with DINA and LCDM in their diagnosis result using the same four-attribute data set. There were slight differences in the attribute mastery probability estimate ( p&d14;k ) among the three model (DINA, LCDM, BIBP), which could result in different diagnosis results for attribute mastery pattern (alphak). Simulation studies were conducted to (1) evaluate general accuracy of the BIBP parameter estimation, (2) examinee the impact of various factors such as attribute correlation (no, low, medium, and high), attribute difficulty (easy, medium, and hard) and sample size (100, 300, and 500) on the consistency of the parameter estimation of BIBP, and (3) compare the BIBP method with the DINA model in the accuracy of recovering true parameters. It was found that the general accuracy of the BIBP method in the true parameter estimation was relatively high. The DINA estimation showed slightly higher overall correct classification rate but the bigger overall biases and estimation errors than the BIBP estimation. The three simulation variables (Attribute Correlation, Attribute Difficulty, and Sample Size) showed significant impacts on the parameter estimations of both models. However, they affected differently the two models: Harder attributes showed the higher accuracy of attribute mastery classification in the BIBP estimation whereas easier attributes were associated with the higher accuracy of the DINA estimation. In conclusion, BIBP appears an effective method for CDA with the advantage of easy and fast computation and a relatively high accuracy of parameter estimation.
机译：认知诊断评估（CDA）是心理和教育测试的新理论框架，旨在提供有关应试者在特定知识结构和处理技能方面的优缺点的详细信息。在过去的三十年中，已经为CDA开发了十多种心理测量模型，也称为认知诊断模型（CDM）。尽管他们已经成功地提供了有关应试者的有用诊断信息，但是大多数CDM都是复杂的，因为与要测量的项目的技能（属性）数量成比例的参数很多。大量参数导致估计的大量计算需求。此外，取决于所选模型，需要各种特定的软件应用程序。本研究的目的是使用一种用户友好的软件应用程序，为CDA提出一种简单有效的方法，而无需大量的计算需求。由于以下事实，将二项式比例的贝叶斯推断（BIBP）应用于CDA：当我们有二项式观测值（例如项目响应（对/错））时，使用beta分布作为参数估计的先验值（即，属性-精通概率），无需任何积分即可轻松找到参数的beta后验。 BIBP在CDA上的应用可以灵活，这取决于测试项目属性设计和考生的属性掌握模式。在这项研究中，通过实际数据研究和模拟研究探索了应用BIBP方法的有效方法。此外，将其他已有的诊断模型（如DINA和LCDM）与BIBP方法的诊断结果进行了比较。；在实际数据研究中，BIBP方法通过两种不同的项目设计应用于测试数据：四个属性和十个属性。此外，使用相同的四属性数据集将BIBP方法与DINA和LCDM的诊断结果进行了比较。三种模型（DINA，LCDM，BIBP）之间的属性掌握概率估计值（p＆d14; k）略有不同，这可能导致属性掌握模式（alphak）的诊断结果不同。进行了仿真研究，以（1）评估BIBP参数估计的总体准确性，（2）检查各种因素的影响，例如属性相关性（无，低，中和高），属性难度（易，中和硬） ）和样本量（100、300和500）上BIBP参数估计的一致性，以及（3）将BIBP方法与DINA模型进行比较，以恢复真实参数的准确性。结果发现，BIBP方法在真实参数估计中的一般精度较高。与BIBP估计相比，DINA估计显示总体正确分类率略高，但总体偏差和估计误差更大。三个模拟变量（属性相关性，属性难度和样本大小）对两个模型的参数估计均显示出显着影响。但是，它们对两个模型的影响不同：较难的属性在BIBP估计中显示出较高的属性掌握分类准确度，而较容易的属性与DINA估计的较高准确度相关。综上所述，BIBP以其简便快捷的计算和相对较高的参数估计精度而成为一种有效的CDA方法。

著录项

  • 作者

    Kim, Hyun Seok John.;

  • 作者单位

    Georgia Institute of Technology.;

  • 授予单位 Georgia Institute of Technology.;
  • 学科 Psychology Psychometrics.;Education Tests and Measurements.
  • 学位 Ph.D.
  • 年度 2011
  • 页码 149 p.
  • 总页数 149
  • 原文格式 PDF
  • 正文语种 eng
  • 中图分类
  • 关键词

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱：kefu@zhangqiaokeyan.com

京公网安备：11010802029741号 ICP备案号：京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有

  • 客服微信

  • 服务号