Generative AI has the potential to drastically improve the landscape of computing education by automatically generating personalized feedback and content. In particular, this potential lies in the advanced capabilitie...
详细信息
ISBN:
(纸本)9781450399753
Generative AI has the potential to drastically improve the landscape of computing education by automatically generating personalized feedback and content. In particular, this potential lies in the advanced capabilities of state-of-the-art deep generative and large language models such as OpenAI’s Codex [7], ChatGPT [11], and GPT-4 [12]. In our work, we seek to investigate the capabilities of these models in visualprogramming domains popularly used for K-8 programming education, including domains like Scratch [17], Hour of Code: Maze Challenge by *** [4, 5], and Karel [13]. Recent works have shown us sparks of advanced capabilities of such models for various education scenarios in introductory Python programming [2, 14, 18, 20]. In fact, a study in 2022 had ranked Codex in the top quartile w.r.t students in a large Python programming course [8]. However, all these works consider only text-based Python programming and leave open the question of how well these models would perform for visualprogramming. The main research question is: Do state-of-the-art neural generative models show advanced capabilities for visualprogramming on par with their capabilities on text-based Python programming?In our work, we evaluate these models for visualprogrammingbased on the following three settings designed to capture various generative and problem-solving capabilities: We conduct our evaluation based on 10 representative tasks from two visualprogramming domains: Hour of Code: Maze Challenge by *** [4, 5] and Intro to programming with Karel course by *** [3, 13]. As illustrative examples, Figures 1, 2, and 3 show the output of GPT-4 in three settings for Maze18 task. We will provide the detailed analysis and prompts used in a longer version of this poster. Our preliminary results for ChatGPT (based on GPT-3.5) and GPT-4 show that these models perform poorly and produce incorrect output the majority of the time. These results highlight that state-of-the-art neur
A proliferation of introductory visualprogramming language raises the question of how to introduce VPLs to more creators and how to improve the usability and learnability of the VPL platform. This paper compares two ...
详细信息
A proliferation of introductory visualprogramming language raises the question of how to introduce VPLs to more creators and how to improve the usability and learnability of the VPL platform. This paper compares two different teaching methods and visualprogramming paradigm software to observe the influence of different factors on the use of visualprogramming software by adult learners in the designer group. A more constructive teaching style using gamification between participants and a more behaviorist teaching style using small lecture and interaction were exposed to participants in different instruction group. Different visualprogramming platforms were also tested in each group. User experience scores based on performance score and self-reported scores were collected during and after participants operating on the visualprogramming software. The independent-sample t-tests were used to answer the research question that: is there a mean difference in scores for operating and self-rating between different instruction groups and different visualprogramming platform. The test result shows that there is a mean difference in the efficiency (performance score) between the behaviorist instruction approach and the constructivism instruction approach for using visualprogramming software. The performing scores in the constructive teaching group are statistically significantly higher than the performing scores in the behaviorist teaching group. In addition, designers who exposed to the imperative visualprogramming software also perform better than those exposed to declarative visualprogramming software. The study of constructive education in teaching visualprogramming language worth further exploration, in fact, under the trend of digital learning, constructive learning mechanism and the auxiliary of visualprogramming, a combination of both to learn programming, especially for programming beginners’ introductory courses, has a positive effect.
暂无评论