This study was undertaken to determine the effects of the problem-based and activity-based approach to students’ achievement and retention scores in Introductory Probability and Statistics. A pre-test-post-testquasi experimental and control group design was used in the study to test the effectiveness of the three approaches of instruction. Three intact sections, with a total of 101 students of Saint Joseph Institute of Technology-Butuan City, were the subjects of this study. Two sections were randomly assigned as experimental groups and the other one, a control group. Experimental groups were exposed to problem-based and activity-based approach, while the control group was exposed to lecture-discussion. All groups were given the pre-testand post-test using the validated teacher-made questionnaire before and after the treatment respectively. The Analysis of Covariance (ANCOVA) of unequal n’s was used to analyze the effect of the three approaches of instruction on students’ achievement. Likewise, the Analysis of Variance (ANOVA) of unequal n’s was used to analyze the effect of the three methods on students’ retention scores. Pre-test was used as a covariate in the study. Based on the analysis, the study revealed the following results: (1) Students who were exposed to problem-based instruction were as good as those who were exposed to lecture-discussion as evidenced by their overall mean post-test scores. (2) There was a significant difference in the students’ achievement scores as influenced by the three methods of instruction. Students exposed in problem-based instruction performed better than the activity method. However, problem-based method was as good as the lecture-discussion approach and the activity approach was as good as the lecture-discussion approach.