This study was conducted with the aim of tracking the mathematics attitudes of primary school students from the year they started school until they graduated from primary school and transitioned to middle school, covering a four-year period. The research involved a longitudinal method and was carried out with the participation of primary school students in Istanbul, Turkey. The data collection instruments utilized in this study were the "Student Information Form" prepared by the researchers and the "Mathematics Attitude Scale" developed by Askar (1989). Statistical analysis of the data was performed using SPSS 25.0, incorporating the Mann Whitney U test, Kruskal Wallis H test, and Spearman Correlation data analysis methods. In the study, it was observed that students had a 'moderate' mathematics attitude in the first grade, a 'very high' attitude in the second grade, and an overall 'high' attitude in the third and fourth grades. As students progressed through the grades, their mathematics attitudes decreased. Although there was a trend favoring males in scores over the years based on gender, this difference was not statistically significant. Finally, a significant relationship was found between mathematics attitude and students' mathematics grades in both the first and fourth grades.