Classifying different types of network traffic, including encrypted content, guarantees a fair and efficient internet experience for all users. With the increasing trend in encrypting network traffic and the dynamic nature of internet applications, there is a need for more advanced and adaptive classification techniques. Machine learning algorithms have demonstrated promising results in real-time network analysis, but the nature of the encrypted network traffic makes these models prone to overfitting because the model might memorize details about the encrypted traffic instead of learning general patterns in the data. In this paper, we explore the use of ensemble methods for encrypted network traffic classification. Our results show that the ensemble methods can boost the performance of machine learning models, and thus are more feasible for encrypted network traffic classification. We also conducted a statistical assessment using Friedman test followed by post-hoc analysis to evaluate the performance of the different machine learning models across different encrypted datasets.