This study was conducted to determine the effect of REACT strategy assisted by GeoGebra software on students' mathematical representation ability. This study was quasi-experimental with a quantitative approach. The design used in this study was The Nonequivalent Posttest Only Control Group Design. Four posttest questions were given to 25 students of Class VIII A and 25 students of Class VIII B at SMP Negeri Arun Lhokseumawe (public middle school). The data obtained from the posttest results were analyzed by data normality test and data homogeneity test, then hypotheses were tested using a t-test to see the effect of REACT strategy assisted by GeoGebra software on students' mathematical representation ability. The results of the data analysis of the posttest score tested using the t-test, the sig value was obtained. (2-tailed) of 0.000 > 0.05. This showed that there was a significant effect of using the REACT strategy assisted by GeoGebra software on students' mathematical representation ability.