<![CDATA[This study was motivated by the low problem-solving ability and self-confidence of fifth-grade elementary school students in mathematics learning, as shown by Indonesia's low PISA achievement and the results of preliminary observations and interviews in several elementary schools in Tulungagung Regency. This study aims to determine the effect of differentiated learning based on visual, auditory, and kinesthetic learning styles on problem-solving ability, self-confidence, and both variables simultaneously in mathematics. The study used a quantitative approach with a quasi-experimental method and a non-equivalent control group design. The research sample consisted of fifth-grade students in class VA (N=36) as the experimental class, who received differentiated learning based on learning styles, and class VB (N=35) as the control class, who received conventional learning, at SDN 1 Kedungwaru, selected using purposive sampling. The research instruments consisted of eight essay questions to measure problem-solving ability and fifteen questionnaire items to measure self-confidence, both of which were tested for construct validity using exploratory factor analysis and for reliability using Cronbach's Alpha. Data were analyzed using Multivariate Analysis of Variance (MANOVA) with the assistance of the Jamovi application. The results showed that the experimental class obtained an average problem-solving ability score of 91.2 and self-confidence of 91.1, higher than the control class, which obtained only 68.0 and 73.1, respectively. The MANOVA test results showed a Pillai's Trace value of 0.878 with p < 0.001 on all test criteria, meaning that differentiated learning based on learning styles had a significant effect on students' problem-solving ability and self-confidence, both separately (F=299, p<0.001 and F=280, p<0.001) and simultaneously, with a very large effect size (η²=0.813 and η²=0.802).]]>
