This paper reviews the use of bootstrap methods for multiple comparison procedures in the statistical software Sisvar. Sisvar is a widely used system for statistical analysis and scientific research, known for its accuracy, precision, simplicity, and robustness. While Sisvar offers many statistical analysis options, the use of bootstrap-based multiple comparison procedures is less common. This paper aims to review this topic and highlight the advantages of using Sisvar for such analyses, particularly in comparing treatment means. Tests such as Dunnett, Tukey, Student-Newman-Keuls, and Scott-Knott are performed using bootstrap methods, which show greater power and better control of experimentwise type I error rates under non-normal, asymmetric, platykurtic, or leptokurtic distributions. The paper discusses the use of bootstrap methods in Sisvar for multiple comparisons, including the determination of p-values for each hypothesis using several methods. The adjusted p-values are considered the best performance. The paper also evaluates the performance of these methods through Monte Carlo simulations, considering experimentwise and comparisonwise type I error rates and power. The results show that bootstrap tests have greater power than original procedures in several cases, making them suitable for non-normality. The bootstrap Tukey test is considered the best for multiple comparisons, as it properly controls experimentwise type I error rates under normal and non-normal models and shows high power under the alternative hypothesis. The paper also acknowledges the financial support from CNPq, CAPES, and FAPEMIG, and highlights the collaboration with Bryan Frederick John Manly and Clarice Garcia Borges Demétrio in the development of the multiple comparison procedures.This paper reviews the use of bootstrap methods for multiple comparison procedures in the statistical software Sisvar. Sisvar is a widely used system for statistical analysis and scientific research, known for its accuracy, precision, simplicity, and robustness. While Sisvar offers many statistical analysis options, the use of bootstrap-based multiple comparison procedures is less common. This paper aims to review this topic and highlight the advantages of using Sisvar for such analyses, particularly in comparing treatment means. Tests such as Dunnett, Tukey, Student-Newman-Keuls, and Scott-Knott are performed using bootstrap methods, which show greater power and better control of experimentwise type I error rates under non-normal, asymmetric, platykurtic, or leptokurtic distributions. The paper discusses the use of bootstrap methods in Sisvar for multiple comparisons, including the determination of p-values for each hypothesis using several methods. The adjusted p-values are considered the best performance. The paper also evaluates the performance of these methods through Monte Carlo simulations, considering experimentwise and comparisonwise type I error rates and power. The results show that bootstrap tests have greater power than original procedures in several cases, making them suitable for non-normality. The bootstrap Tukey test is considered the best for multiple comparisons, as it properly controls experimentwise type I error rates under normal and non-normal models and shows high power under the alternative hypothesis. The paper also acknowledges the financial support from CNPq, CAPES, and FAPEMIG, and highlights the collaboration with Bryan Frederick John Manly and Clarice Garcia Borges Demétrio in the development of the multiple comparison procedures.