基于试题难度的测试成绩计算方法——以建筑结构设计课程为例
CSTR:
作者:
作者单位:

西南交通大学 土木工程学院,四川 成都 610031

作者简介:

林拥军(1974—),西南交通大学土木工程学院副教授,博士,主要从事结构工程研究,(E-mail)scsmith@126.com。

通讯作者:

中图分类号:

G642

基金项目:

西南交通大学2021年本科教育教学研究与改革项目(教学实践类项目)“混合式教学背景下土木工程专业课实施效果评价方法及优化策略” (2103008)


Calculation method of test scores based on the difficulty of test questions: taking the course of architectural structure design as an example
Author:
Affiliation:

School of Civil Engineering, Southwest Jiaotong University, Chengdu 610031, P. R. China

Fund Project:

  • 摘要
  • |
  • 图/表
  • |
  • 访问统计
  • |
  • 参考文献
  • |
  • 相似文献
  • |
  • 引证文献
  • |
  • 资源附件
  • |
  • 文章评论
    摘要:

    测试成绩作为教学效果评价的重要指标,其终极目标是能合理反映被试者的能力水平,但传统意义上的卷面得分并不能完全反映被试者测试得分和能力水平之间的关系。为此,提出根据试题难易程度,对卷面分数进行调整的测试成绩计算方法,并以土木工程专业课建筑结构设计为例,分析其具体应用。结果表明:试题客观难度系数取值对卷面高分段影响小,对低分段影响大;分数变化率最大绝对值在卷面最低分和最高分之间,偏向最低分一侧。当卷面分数越接近最高分或最低分时,其受影响程度越小。即当客观难度系数取0时,分数变化率在-55.4%~-15.0%之间;取0.5时,变化率在-11.7%~-3.9%之间;取0.7时,变化率在-5.7%~2.5%之间。客观难度系数越小,计算出的分数与实际卷面分数偏差就越大,分数分布越分散。不同教学班的试卷难度系数差别越大,教学班单独计算和多班合并计算分数差异就越大。当客观难度系数取0.5时,教学班单独计算和多班合并计算分数偏差在-10%~+3%以内;取0.7时,偏差在-4%~+3%以内。因此,考虑试题难度的测试成绩计算方法能合理反映分数和被测试者能力水平间的关系,客观难度系数建议取值在0.5~0.7之间。同一门土木工程专业课程,若存在多个教学班时,当客观难度系数不大于0.6时,建议多班合并计算;大于0.6时,各教学班可单独计算。

    Abstract:

    As an essential indicator of teaching effect evaluation, test scores can reasonably reflect the ability level of the examinees. However, the paper scores in the traditional sense cannot fully reflect the relationship between the test scores and the ability level of the examinees. Therefore, this paper proposes a test score calculation method that adjusts the test scores according to the difficulty of the test questions and analyzes its specific application by taking the civil engineering course architectural structure design as an example. The results show that the value of the objective difficulty coefficient of the test questions has little influence on the high-score section of the test paper but has a significant influence on the low-score section; The maximum absolute value of the score change rate is between the lowest score and the highest score, which is biased toward the lowest score side. The closer the score is to the highest or lowest score, the less it will be affected. When the objective difficulty coefficient is taken as 0, the score change rate is between - 55.4% to - 15.0%; When it is taken as 0.5, the rate is between - 11.7% to - 3.9%; When it is taken as 0.7, the rate is between - 5.7% to 2.5%. The smaller the objective difficulty coefficient is, the greater the deviation between the calculated score and the paper score. The more dispersed the score is, the greater the difference between the test paper difficulty coefficients of different teaching classes is, and the more significant the difference between the scores calculated by the teaching class alone and by the combined calculation of multiple classes is. When the objective difficulty coefficient is 0.5, the deviation between the scores calculated by the teaching class alone and by the combined calculation of multiple classes is within - 10% to + 3%, and when it is 0.7, the deviation is within - 4% to + 3%. The test score calculation method considering the difficulty of the test questions can reasonably reflect the relationship between the scores and the ability level of the examinees. The objective difficulty coefficient is recommended to be 0.5 to 0.7. When there are multiple teaching classes in the same civil engineering professional course, when the objective difficulty coefficient is not greater than 0.6, it is recommended that multiple classes be calculated together. Each teaching class can be calculated separately when it is greater than 0.6.

    参考文献
    相似文献
    引证文献
引用本文

林拥军,张曾鹏,张晶,卢立恒.基于试题难度的测试成绩计算方法——以建筑结构设计课程为例[J].高等建筑教育,2025,(4):143-158.

复制
分享
相关视频

文章指标
  • 点击次数:
  • 下载次数:
  • HTML阅读次数:
  • 引用次数:
历史
  • 收稿日期:
  • 最后修改日期:2022-09-16
  • 录用日期:
  • 在线发布日期: 2025-07-19
  • 出版日期:
文章二维码