...
首页> 外文期刊>Quality Control and Applied Statistics >Calibrating item families and summarizing the results using family expected response functions
【24h】

Calibrating item families and summarizing the results using family expected response functions

机译:使用系列预期响应函数校准项目系列并汇总结果

获取原文
获取原文并翻译 | 示例
           

摘要

The items produced from a single model constitute an item family as they are related to one another through the common generating model. Thus, use of models in real tests will require statistical models which can account for the dependence structure. The present work examines a model which assumes the item parameters of a three-parameter logistic (3PL) model that are normally distributed with a mean vector and a variance structure that depend on the family generating the item. It shows how to fit the hierarchical model using the MCMC algorithm. This implies for each family a FERF gives the probability of a correct response to a randomly generated item. A method to estimate the FERF and an approximate prediction limit for members of the family is also outlined. Two real data examples are included.
机译:由单个模型生产的物料构成了物料族,因为它们通过共同的生成模型相互关联。因此,在实际测试中使用模型将需要可以解释依赖性结构的统计模型。本工作研究了一个模型,该模型假定一个三参数逻辑模型(3PL)模型的项目参数,该参数通常以均值矢量和方差结构分布,该平均参数和方差结构取决于生成项目的族。它显示了如何使用MCMC算法拟合层次模型。这意味着对于每个家庭,FERF都会给出对随机生成的物品做出正确响应的概率。还概述了一种估计FERF的方法以及该家庭成员的近似预测极限。包括两个真实的数据示例。

著录项

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