It matters how you ask: how scenario and question wording influence problematic word problem solving

• We experimentally addressed ways to encourage realistic word problem solving. • A mathematics classroom context led to fewer realistic solution. • It further led to fewer expected preferences for realistic solutions. • Questions indicating an imminent action led to more realistic solutions. It has been repeatedly shown that students often provide unrealistic solutions to word problems. We investigated the impact of the problem scenario and the question wording on solving word problems that require realistic considerations (p-problems). We assumed that (1) a real-world scenario and (2) question wordings emphasising the need for an applicable solution by indicating imminent actions of persons in the problem statement would foster realistic responses. A total of N = 127 7th graders participated. A two-factorial experiment with the factors “scenario” (maths test vs. school trip) and “imminent action” (with vs. without) was implemented. Participants solved three p-problems and indicated whether they expected a person from the problem scenario to prefer a realistic or unrealistic solution. In the school trip scenario, students had statistically significant higher odds of providing realistic solutions ( OR = 2.38, p = 0.008) and of expecting a realistic solution to be preferred compared to the maths test scenario ( OR = 9.01, p < 0.001). Questions indicating imminent actions of persons in the problem statement led to significantly higher odds of providing realistic solutions ( OR = 2.80, p = 0.002), but did not significantly affect expectations regarding a third person’s preference ( OR = 1.50, p = 0.368), compared to standard questions. Results suggest a relevant impact of the mathematics classroom context, and introduce a way to effectively encourage realistic solutions.