PROJECT TITLE :

Recommender Systems and Scratch An integrated approach for enhancing computer programming learning

ABSTRACT:

Learning computer programming is a difficult task. Visual programming languages (VPLs), such as Scratch, have demonstrated to be particularly promising for novices among the current efforts to solving this difficulty. Surprisingly, several colleges and universities have begun to use VPLs to teach basic programming concepts, primarily in CS1 classes. However, one significant concern with Scratch's use in higher education is that students may become demotivated when confronted with programming tasks that do not meet their unique needs. We offer CARAMBA, a Scratch extension with an exercise recommender system, to try to overcome this hurdle. CARAMBA is able to tailor student learning with Scratch by appropriately proposing tasks for students based on variables such as taste and complexity. An in-depth analysis of the effects of our approach on both the learning of basic CS1 concepts and students' overall performance was done. With 88 college students from an Ecuadorian university, we used an analogous pretest-posttest design. The results show that recommending Scratch tasks had a good impact on students' programming learning abilities, as measured by pass rates. Overall, our approach obtained a pass rate of over 52%, which is 8% higher than the rate reached during a previous experience using solely Scratch (without advice) and 21% higher than the historical outcomes of traditional instruction (without Scratch). Furthermore, we examined the extent to which students exploited CARAMBA in order to demonstrate two facts: students did utilize CARAMBA, and there was a substantial, positive link between CARAMBA use and the students' grades.