This study aims to optimize the use of classroom assessment to improve 4A SD Negeri Belitung Selatan 5 mathematics learning outcomes through the trial by error method. The research employed a Classroom Action Research (CAR) approach conducted over one semester by utilizing the Mid-Semester Summative Assessment (STS) as baseline data and the End-of-Semester Summative Assessment (SAS) as final data. The research instruments consisted of a 25-item multiple-choice test administered in both STS and SAS to ensure measurement consistency, as well as non-test instruments in the form of classroom observations and reflective notes. The results indicated an increase in the class average score from 55,00 in STS to 61,00 in SAS. The percentage of students achieving mastery learning improved from 35,71% to 53,57%. The N-Gain value of 0,13 fell into the low category, indicating that the improvement in learning outcomes occurred gradually. Nevertheless, the increased number of students reaching the minimum mastery criterion demonstrated that optimizing assessment contributed to a more equitable distribution of learning achievement. Observational findings revealed that students became more responsive to assessment feedback, were able to correct their mistakes, and showed greater confidence in solving mathematical problems. This study concludes that the reflective use of classroom assessment through the trial by error method is effective in gradually improving students’ mathematics learning outcomes over one semester. Keywords: classroom assessment, trial by error, classroom action research, mathematics learning outcomes, elementary school.
Copyrights © 2026