The purpose of this study are to find out whether increasing students 'mathematical connection ability with learning using the SAVI approach is better with conventional learning, to find out how students' mathematical connection skills with learning the SAVI approach, and to find out a positive relationship between students 'attitudes towards learning using the SAVI approach with increasing students' mathematical connection abilities. This research method is a quasi experiment that tests the use of SAVI approach to the experimental class and conventional learning in the control class. Sampling using purposive sampling. The selected sample is X MIA 1 as the experimental class and X MIA 4 as the control class. The instrument used was a test instrument for mathematical connection ability and a non-attitude scale questionnaire test. The material in this study was Trigonometry. The results of this study are: (1) improvement of students' mathematical connection ability with the SAVI approach is better than with conventional learning, (2) students' mathematical connection skills with learning with the SAVI approach are good, (3) there is no positive relationship between students 'attitudes towards learning with the SAVI approach with increasing students' mathematical connection abilities.

No abstract