<![CDATA[This study analyzes the implementation of Project-Based Learning (PjBL), Problem-Based Learning (PBL), and the Case Method among Accounting Department students at Politeknik Negeri Manado in terms of learning outcomes and student perceptions. Although active learning is widely promoted in Indonesian higher education, comparative evidence on these three models within vocational accounting programs remains limited. A quantitative descriptive-comparative approach was employed. Learning outcome data were obtained from the pre-tests and post-tests of 70 students (PjBL = 41, PBL = 17, Case Method = 12), while perception data were collected through a 21-item Likert-scale questionnaire completed by 70 respondents. Data were analyzed using descriptive statistics, N-Gain, paired t-tests and Wilcoxon tests, Kruskal-Wallis and Mann-Whitney tests, ANCOVA with the pre-test as a covariate, and instrument validity and reliability tests. The results show that all three models significantly improved learning outcomes (p < 0.001), with mean N-Gain in the medium category: PjBL 0.56, PBL 0.47, and Case Method 0.34. Differences between models remained significant after controlling for prior ability (ANCOVA, F = 5.80; p = 0.005). However, further analysis indicates that N-Gain was also influenced by student characteristics: prior ability was positively correlated with N-Gain (ρ = 0.60), and within the PjBL group, D4 Taxation Accounting students achieved considerably higher N-Gain than D4 Financial Accounting students. Student perceptions were very high (mean 4.23 out of 5; Cronbach’s Alpha = 0.945), with collaboration and teamwork as the main strength and critical thinking as the aspect most in need of reinforcement. These findings imply that learning models should be matched to students’ level of study and prior readiness.]]>
