Division of Restorative Dentistry, Department of oral health practice, University of Kentucky College of dentistry, Lexington, KY, USA.
Department of Statistics, University of Kentucky, Lexington, KY, USA.
Eur J Dent Educ. 2021 Aug;25(3):524-535. doi: 10.1111/eje.12628. Epub 2020 Dec 3.
Digital technology has the potential to provide a bias-free evaluation instrument for practical examination grading. E4D Compare software was designed for teaching institutions to allow comparison between scanned models prepared by students and master models prepared by instructors. The aim of this study was to determine the tolerance value for the E4D compare software (E4D Technologies LLC, Richardson, TX, USA) that provides scores comparable with faculty grades for wax-ups #23 and #19. Additionally, this study sought to assess the correlation between students' self-assessments, visual and digital grades and the reliability of the software.
Student wax-ups (n = 112) were uploaded into the software and then compared with faculty-generated master wax-ups. Digital grading was performed at tolerances 200-800 μm in 50 μm increments and was repeated twice.
A tolerance of 350 μm was the closest to faculty grades for #23 and 500 μm was the closest for #19. Visual and digital grades showed moderate to high correlation for both wax-ups. Correlations between students' self-assessments and visual and digital grades improved with #19 versus #23. A near-perfect correlation was found between grades at the first and second digital grading sessions.
The tolerance that closely matches faculty grades differs according to the tooth type. The software provided consistent grades and correlated well with faculty grades. Students' self-assessment skills improved as they proceeded throughout the course. Further studies are necessary to ascertain the role of the software in improving students' self-assessment skills.
数字技术有可能为实践考试评分提供一个无偏倚的评估工具。E4D Compare 软件专为教学机构设计，允许学生制作的扫描模型与教师制作的标准模型进行比较。本研究的目的是确定 E4D Compare 软件（美国德克萨斯州理查森的 E4D Technologies LLC 公司）的容差值，该软件可以为 #23 和 #19 蜡型提供与教师评分相媲美的分数。此外，本研究还旨在评估学生的自我评估、视觉和数字评分之间的相关性以及软件的可靠性。
将学生的蜡型（n=112）上传到软件中，然后与教师生成的标准蜡型进行比较。数字评分在 200-800μm 的容差值下进行，每隔 50μm 重复两次。
对于 #23，350μm 的容差值最接近教师评分，而对于 #19，500μm 的容差值最接近教师评分。对于这两种蜡型，视觉和数字评分之间都存在中等至高度的相关性。与视觉评分相比，学生的自我评估与数字评分之间的相关性随着蜡型 #19 提高而提高。在第一次和第二次数字评分之间，评分之间存在近乎完美的相关性。
与教师评分最匹配的容差值因牙齿类型而异。该软件提供了一致的评分，并与教师评分高度相关。随着课程的进行，学生的自我评估技能有所提高。需要进一步的研究来确定该软件在提高学生自我评估技能方面的作用。
