Abstract
The article examines the development of Grade 10 students’ skill in constructing a solution model for olympiad mathematics problems. Existing research often reports overall problem-solving success, while changes in method selection, planning, checking conditions, and completing a proof are less often examined separately. The study aimed to trace changes in these components during systematic use of the partial-search method and formative assessment. The study was conducted as Action Research with one group of 24 Grade 10 students during the 2024–2025 academic year. Data were collected through initial and final diagnostic tasks, written solutions, questionnaires, semi-structured interviews, classroom observation, and a five-indicator diagnostic rubric. The research cycles used a solution framework, tasks with varying degrees of support, peer checking, group reconstruction of solution algorithms, and formative feedback. The mean rubric score increased from 6.4 to 11.9 out of 15. The proportion of students able to justify method selection rose from 29% to 71%, while the proportion checking conditions of applicability and equality cases rose from 17% to 63%. The proportion presenting a complete proof increased from 21% to 58%. Analysis of written work showed that method selection and planning improved more quickly than the completion of a rigorous proof. The findings describe positive changes within this group, but the absence of a control group does not allow the changes to be attributed solely to the pedagogical approach