Mental models and programming algorithms in computer science technical high school students

This paper analyzes the association between the typology of algorithmic programming structures and the mental models constructed for resolution. The study is part of the quantitative paradigm with a quasi-experimental design and included the participation of ninety-five students of eleventh grade of computer science technical high school. The methodology involved evaluating the processes developed by the students, through the models obtained in an algorithmic problem-solving test; the results obtained, were submitted to a descriptive analysis, in order to establish statistically significant differences between the levels of consistency of the models associated with the resolution process and the types of structures. The above, was analyzed in terms of Chi-square likelihood, supplemented by the magnitude of the Cramer V effect; the Dwass-Steel-Critchlow-Fligner (DSCF) test, was also applied to make comparisons between structures. The results indicated more consistent models in the sequential structure, in contrast to inconsistent and ambiguous models prevalent in conditional, cyclical, and nested structures. The main conclusions show a meaningful relationship between mental models and the typology of algorithmic programming structures, demonstrating the need to focus didactic strategies on problem solving.