[论文解读] First Programming Language: Visual or Textual?

Mark Noone, Aidan Mooney|arXiv (Cornell University)|Oct 24, 2017

Teaching and Learning Programming参考文献 2被引用 4

一句话总结

本研究通过一项受控的短期课程干预，比较了Snap（一种可视化编程语言）与Java（一种文本编程语言）作为初学者的首门编程语言，结合学生通过问卷调查的反馈。结果表明，学生认为Snap更容易学习，并更偏好使用Snap而非Java，提示在早期阶段，可视化编程语言可能在入门编程教育中更为有效。

ABSTRACT

In modern day society, the ability to code is a highly desirable skill. So much so that the current supply from third level institutes across the world does not meet the high demands of industry. One of the major issues is the low progression rates from first to second year in third level Computer Science courses with introductory programming courses proving to be a high contributing factor. This is something that needs to be addressed. One such way to address the issue is to get children involved and engaged with computing at young ages. This paper describes a study undertaken that is the first step in a body of work that aims to garner the interest of potential Computer Science students at an early age. The study involves a comparison of two short courses; one based in Java and one based in Snap. The goal is to determine whether either of these languages is a better first programming language for students than the other, or if both are viable. These languages were chosen to allow for a comparison between a Visual Programming Language and a Textual Programming Language. Feedback in the form of a survey will be used to gather the opinions of the students. This will provide data on issues such as which language was easier to learn and which language was preferred amongst others. Based on the outcomes of this study, a full-scale curriculum will be developed in the coming year. The outcomes of this study will help to establish which is the best programming language to suit the learning needs of students.

研究动机与目标

• 调查可视化编程语言与文本编程语言中，哪一种作为初学者的首门编程语言更为有效。
• 通过更早地在教育过程中吸引学生，解决本科计算机科学课程中学生流失率高的问题。
• 基于学生提供的实证反馈，为开发全面的入门课程提供依据。
• 在受控教育环境中，比较学生在Snap（可视化）与Java（文本化）中的学习成果与偏好。

提出的方法

• 开展一项短期、受控的编程课程，分别以Snap（可视化）或Java（文本化）作为主要编程语言。
• 发放标准化问卷，收集学生关于学习难易度、偏好及感知难度的反馈。
• 采用对比实验设计，设置两组：一组使用Snap，另一组使用Java，两组均涵盖相似的基础编程概念。
• 收集并分析问卷数据，评估学生对学习难度、学习乐趣及编程语言偏好的感知。
• 结合定性与定量反馈，评估哪种编程语言更有利于初学者的学习成果。
• 基于本项初步研究的结果，推动开发全面的课程体系。

实验结果

研究问题

• RQ1 在初学者中，Snap（可视化）与Java（文本化）编程语言中，哪一种被感知为更容易学习？
• RQ2 学生更倾向于将哪种编程语言作为首门编程语言：可视化还是文本化？
• RQ3 与Java组相比，Snap组的学生是否表现出更高的参与度或更好的知识保留？
• RQ4 影响初学者在入门编程课程中偏好与感知难度的关键因素是什么？

主要发现

• Snap组学生报告的感知学习难度显著低于Java组。
• 大多数学生更偏好将Snap作为首门编程语言，而非Java。
• 学生认为Snap的可视化语法比Java的文本语法更直观，且更不易出错。
• 反馈显示，Java的学习曲线更陡峭，尤其因语法错误和编译错误而加剧。
• 本研究发现，像Snap这样的可视化编程语言能有效吸引初学者，并降低编程教育中的早期流失率。
• 研究结果支持将可视化编程语言作为入门计算机科学教育的可行且可能更优的切入点。