In this study, we aim to assess the effectiveness of Project-Based Learning (PjBL) based on Multiple Intelligences (MI) theory on students' critical thinking and mathematical communication skills. This research is a quasi-experiment using non-equivalent (pre-test and post-test) control-group design. The population in this study were VIII grade students in a public junior high school in Yogyakarta and two sample classes were selected based on recommendations from teachers and had mathematical abilities that could be considered equal. The experimental class applied PjBL based on MI theory and the control class applied PjBL learning. Data collection was conducted using a test instrument in the form of description questions. This instrument is valid and reliable. Learning is said to be effective if: (1) the mean test score of critical thinking and communication skills is at least 70 (from a score range of 0-100); (2) the mean of post-test of critical thinking and communication skills is higher than the mean of pre-test; (3) the mean final score of students' critical thinking and mathematical communication skills in the experimental class is higher than the control class. Data analysis used one-sample t-test, paired-sample t-test, and T2 Hotteling test. The results of the study using the significance level α = 5% showed that (1) PjBL learning based on MI theory is effective in terms of students' critical thinking and mathematical communication skills; (2) PjBL based on MI theory learning is more effective than PjBL learning in terms of communication, but not more effective in terms of critical thinking skills. Although the data from the sample showed the acquisition of mean scores and the increase in the mean value of critical thinking learning in PjBL based on MI theory was better than PjBL learning in terms of critical thinking skill, but statistically both did not show a significant difference.