当前位置: X-MOL 学术EPJ Data Sci. › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
Both sides of the story: comparing student-level data on reading performance from administrative registers to application generated data from a reading app
EPJ Data Science ( IF 3.0 ) Pub Date : 2021-08-19 , DOI: 10.1140/epjds/s13688-021-00300-y
Bent Sortkær 1 , Emil Smith 1 , David Reimer 1 , Stefan Oehmcke 2 , Ida Gran Andersen 1
Affiliation  

The use of various learning apps in school settings is growing and thus producing an increasing amount of usage generated data. However, this usage generated data has only to a very little extend been used for monitoring and promoting learning progress. We test if application usage generated data from a reading app holds potential for measuring reading ability, reading speed progress and for pointing out features in a school setting that promotes learning. We analyze new data from three different sources: (1) Usage generated data from a widely used reading app, (2) Data from a national reading ability test, and (3) Register data on student background and family characteristics. First, we find that reading app generated data to some degree tells the same story about reading ability as does the formal national reading ability test. Second, we find that the reading app data has the potential to monitor reading speed progress. Finally, we tested several models including machine learning models. Two of these were able to identify variables associated with reading speed progress with some degree of success and to point at certain conditions that promotes reading speed progress. We discuss the results and avenues for further research are presented.



中文翻译:


故事的两面:将管理寄存器中的学生级阅读表现数据与阅读应用程序中的应用程序生成的数据进行比较



学校环境中各种学习应用程序的使用不断增长,从而产生越来越多的使用生成数据。然而,这种使用生成的数据仅在很小的程度上用于监控和促进学习进度。我们测试从阅读应用程序生成的应用程序使用数据是否具有衡量阅读能力、阅读速度进度以及指出学校环境中促进学习的功能的潜力。我们分析来自三个不同来源的新数据:(1)来自广泛使用的阅读应用程序的使用生成数据,(2)来自全国阅读能力测试的数据,以及(3)有关学生背景和家庭特征的注册数据。首先,我们发现阅读应用程序生成的数据在某种程度上讲述了与正式的全国阅读能力测试相同的阅读能力故事。其次,我们发现阅读应用数据有潜力监控阅读速度进度。最后,我们测试了包括机器学习模型在内的多种模型。其中两个能够在一定程度上成功地识别与阅读速度进步相关的变量,并指出促进阅读速度进步的某些条件。我们讨论了结果并提出了进一步研究的途径。

更新日期:2021-08-19
down
wechat
bug