Creating algorithms by accounting for features of the solution: the case of pursuing maximum happiness.

Research shows that though some students can easily find the correct solutions to the problem(s) at hand, the algorithms that these students create are not always ones that would, when implemented, produce the correct solution(s). Towards shedding light on this phenomenon, the present study explicates a mechanism—accounting for features of the solution—by which students create algorithms. Via this mechanism, students notice particular features of the solution they found, then create specific rules (instructions) within their algorithm, which guarantee that the algorithm outputs an object that possesses the noticed features. Thus, an algorithm can produce the solution only if the object it outputs has all the features of the solution. In this paper, I explicate the accounting for mechanism within the collaborative work of two groups of students on a contextualised graph theory task, which invited the groups to create an algorithm for finding an optimal seating arrangement. Both groups found an optimal seating arrangement and seemingly employed the accounting for mechanism to create their respective algorithms. However, only one group's algorithm could actually produce the optimal arrangement. The questions explored in this study are: (1) What sorts of features of their solutions did the two groups account for? and (2) What differences (if any) between the features accounted for by the respective groups, might explain why only one group's algorithm could produce the optimal arrangement? [ABSTRACT FROM AUTHOR]