高级检索
顾晗昕, 张玉, 昝子晴, 王焕倩, 白媛媛, 张羽欧, 尤莉莉, 刘远立. 小学生营养健康素养问卷编制及Rasch模型测评[J]. 中国公共卫生, 2023, 39(2): 206-211. DOI: 10.11847/zgggws1138966
引用本文: 顾晗昕, 张玉, 昝子晴, 王焕倩, 白媛媛, 张羽欧, 尤莉莉, 刘远立. 小学生营养健康素养问卷编制及Rasch模型测评[J]. 中国公共卫生, 2023, 39(2): 206-211. DOI: 10.11847/zgggws1138966
GU Han-xin, ZHANG Yu, ZAN Zi-qing, . Grade-specific nutrition health literacy questionnaire for pupils: development and Rasch model-based evaluation[J]. Chinese Journal of Public Health, 2023, 39(2): 206-211. DOI: 10.11847/zgggws1138966
Citation: GU Han-xin, ZHANG Yu, ZAN Zi-qing, . Grade-specific nutrition health literacy questionnaire for pupils: development and Rasch model-based evaluation[J]. Chinese Journal of Public Health, 2023, 39(2): 206-211. DOI: 10.11847/zgggws1138966

小学生营养健康素养问卷编制及Rasch模型测评

Grade-specific nutrition health literacy questionnaire for pupils: development and Rasch model-based evaluation

  • 摘要:
      目的  基于项目反应理论开发不同学段版本小学生营养健康素养问卷,并运用Rasch模型进行质量评价。
      方法  在参考国内外有关量表及相关政策文件的基础上,通过专家咨询、组内讨论拟定我国小学生营养健康素养水平测评指标体系,构建低年级(1~2年级)、中年级(3~4年级)、高年级(5~6年级)三版小学生营养健康素养问卷,于2021年6 — 7月采用方便抽样选取北京市延庆区、青海省西宁市、山西省太原市3地区共计8所城乡小学,采用分层整群随机抽样抽取2530名在校小学生,对低、中、高年级学生分别进行测评;采用Rasch模型中的项目区分度、项目信度、单维性检验、怀特图、项目拟合和误差统计进行问卷质量分析。
      结果  3个年级段问卷均符合单维性假设,大多数条目拟合情况良好;低、中、高年级问卷项目区分度分别为9.57、6.03、13.68,项目信度分别为0.99、0.97、0.99;低年级问卷难度偏低,中高年级问卷难度略高,项目难度分别为2.72、 – 0.18、 – 0.42。
      结论  小学生营养健康素养问卷与Rasch模型拟合良好,具有较好的信度和区分度,能够较好地测量小学生营养健康素养,修改条目表述及考察内容等方式进一步提高问卷质量,为小学生营养健康素养的测量提供科学有效的工具。

     

    Abstract:
      Objective  To develop a nutrition health literacy questionnaire for pupils of different grades based on item response theory and to validate the questionnaire with the Rasch model.
      Methods  An index system for evaluating nutrition health literacy among pupils was established through literature study on researches and reports published at home and aboard and expert consultation for the complication of nutrition health literacy questionnaires for pupils of grade 1 – 2, 3 – 4, and 5 – 6. The developed questionnaires were then assessed among 1 074, 1 045 and 553 pupils of the three grade groups recruited with stratified random cluster sampling at 8 elementary schools in Beijing municipality, Xining city of Qinghai province and Taiyuan city of Shanxi province during June – July 2021. Based on Rasch model, the questionnaires' item discrimination, item reliability and item fitting were validated with unidimensional test, Wright map and other indicators using Winsteps software.
      Results  Valid responses were collected from 1 066, 921 and 543 pupils of the three grade groups, with an overall efficient rate of 94.7%. The questionnaires for the three grade groups were all consistent with the one-dimensional hypothesis and most of the items of the questions were of high goodness of fit. For the questionnaire towards pupils of grade 1 – 2, 3 – 4, and 5 – 6, the item discrimination indexes were 9.57, 6.03, and 13.68; the item reliability indexes were 0.99, 0.97, and 0.99; and the difficulty indexes were 2.72, – 0.18, and – 0.42, respectively, indicating a higher difficulty of the questionnaire for pupils of grade 1 – 2 and a lower difficulty of questionnaires for pupils of grade 3 – 4 and 5 – 6.
      Conclusion  The developed nutrition health literacy questionnaire for pupils of different grades fits with the Rasch model well and is of good reliability and discriminability; the questionnaire could be used in relevant surveys in pupils.

     

/

返回文章
返回