There can be many reasons why students fail to answer correctly to summative tests in advanced computer science courses: often the cause is a lack of prerequisites or misconceptions about topics presented in previous courses. One of the ITiCSE 2020 working groups investigated the possibility of designing assessments suitable for differentiating between fragilities in prerequisites (in particular, knowledge and skills related to introductory programming courses) and advanced topics. This paper reports on an empirical evaluation of an instrument focusing on data structures, among those proposed by the ITiCSE working group. The evaluation aimed at understanding what fragile knowledge and skills the instrument is actually able to detect and to what extent it is able to differentiate them. Our results support that the instrument is able to distinguish between some specific fragilities (e.g., value vs. reference semantics), but not all of those claimed in the original report. In addition, our findings highlight the role of relevant skills at a level between prerequisite and advanced skills, such as program comprehension and reasoning about constraints. We also suggest ways to improve the questions in the instrument, both by improving the distractors of the multiple choice questions, and by slightly changing the content or phrasing of the questions. We argue that these improvements will increase the effectiveness of the instrument in assessing prerequisites as a whole, but also to pinpoint specific fragilities.
The Bebras challenge offers pupils and teachers an engaging opportunity to discover informatics, by solving small tasks that aim at promoting computational thinking. Explanations and comments that reveal the computing concepts underlying the tasks are published after the contest, and teachers are encouraged to use this material in their school practice. In this paper we present an exploratory study aimed at investigating how teachers can make use of Bebras material; in particular our interest is understanding whether teachers are able to identify, comprehend, and apply the computing concepts implied by Bebras tasks, and how they can integrate them into their teaching practice. We qualitatively analyzed teaching projects developed by Italian teachers during a workshop on computing education and based on Bebras tasks; the analysis shows that teachers are in general able to build upon the tasks soundly, but it also raises some critical issues.