This study aims to determine the types of errors students face when solving math problems. First, review the error types based on Watson's criteria and student learning styles. This study uses an explanatory sequential design model. Data analysis techniques include Quantitative data analysis techniques, descriptive statistics, and inferential statistics. At the same time, the qualitative data analysis techniques include; data reduction, data presentation, and conclusion. The results showed that the errors based on Watson's criteria and learning styles obtained were based on student responses, including; Inappropriate Data 40%, Omitted Conclusion 20%, Response Level Conflict 40%, and Above Other 40%. Furthermore, the types of errors obtained in classroom learning include Inappropriate Data 25%, Omitted Conclusion 50%, and Skill Hierarchy Problems 25%. In the visual learning pattern, there are five error classes: Inappropriate Data 9%, Omitted Conclusion 55%, Response Level Conflict 18%, Skill Hierarchy Problem 27%, and Above Other 18%. Several factors influence the error, including; the decline in students' motivation, interest, and thinking skills. Efforts can be made to minimize student errors, including increasing practice on story questions and eliminating students' negative mindset towards learning mathematics