Description

Snap! [1] is a graphical programming language tailored to teaching programming at schools. Two overtaken projects in our group have been able to extend Snap! with blocks that allow to define contracts that can then be verified both dynamically [2]and statically via an encoding into the Boogie verifier [3]. Even though the graphical approach of Snap! has already been evaluated as an effective tool for teaching programming at initial levels of education, we still have no empirical study on the our approach of teaching verification with it.

We propose you to design and overtake an empirical study which would allow us to analyse how good is our approach for teaching verification via the Snap! extensions projects. We would like to know not only the quality of the approach but also what we can improve on our approach in order to maximize it's impact in education.

The current lock-down situation (by April 2020) suggests further challenges for an empirical evaluation of school students. We propose to define two approaches for it, one applicable inside a classroom in a normal teaching situation, and one that could be also applicable in a lock-down situation where classrooms are closed and teaching is done virtually. It would be interesting to compare results of both types of evaluations.

• [1] https://snap.berkeley.edu/
• [2] https://essay.utwente.nl/80736/
• [3] https://essay.utwente.nl/80849/