Time series models and methods typically require data to be stationary or free from a unit root to ensure valid inference. As a result, several tests have been developed to check for the presence of a unit root in time series data. However, these tests vary in statistical power and size, making it difficult for practitioners to select the most suitable one. This study investigates the robustness of five commonly used tests for detecting unit roots in univariate time series data, using Monte Carlo simulations with different time series model specifications. The tests considered include the ADF, PP, ERS, NGP, and KPSS tests. The robustness of each test was evaluated based on its size and power. Results from the Monte Carlo experiment under various model specifications suggest that, in most cases, the Augmented Dickey-Fuller test is the most powerful for sample sizes above 100.