Page 79 - Getting the Picture Modeling and Simulation in Secondary Computer Science Education

Defining and Observing Modeling and Simulation in Computer Science
code it. During testing it appeared to be difficult to attribute unexpected behavior to a fundamental modeling mistake, a programming error, or unexpected (i.e. emergent) behavior that was characteristic for the phenomenon under scrutiny. Students tend to rely on an incremental trial-and-error strategy while implementing their simulation model. Only a few conducted systematic and well documented experiments. Most of these experiments, together with the analysis of the results, were intermingled with the construction of the models.
This incremental development is consistent with description of the modeling
practice, for example by Wilensky and Rand (2015). An ad hoc incremental 3 development (trial-and-error strategy) is typical for novices (Robins et al., 2003).
General remarks. Although this was a small study with a limited number of participants, we learned a lot about students’ understanding of modeling and simulation. Also, our findings indirectly informed us about the quality of the instruction, which leaves room for improvement. The instructional materials used were written with Artificial Intelligence (AI) and the role modeling could play in AI in mind. We feel they lacked specific focus and depth needed to teach modeling to a satisfactory degree.
Several students told us that through work on this project, they learned about the phenomena they modeled, which is in line with earlier findings (Blikstein & Wilensky, 2009; Taub et al., 2014). We often heard them laugh during their work and we observed that many students enjoyed working on this project. We saw that these CS students were able to utilize their CS/CT knowledge and skills to advance their learning in other disciplines.
In the subsequent phases of this research project, we will use these findings to explore CS teachers’ initial pedagogical content knowledge of Computational Science in our second study (see chapter 4), and to develop instructional materials which will be used for the study on the assessment instrument (third study, see chapter 5) and the study on students’ understanding and difficulties while working on Computational Science assignments (fourth study, see chapter 6).
Furthermore, we believe that the results of this research will contribute to the development of the CS curriculum in secondary education in the Netherlands, CS teacher training and CS education in general.
77