Travel time estimation is a crucial component of intelligent transportation systems, affecting various applications such as navigation, ride-hailing, and route planning. Traditional methods for travel time estimation rely on subjective judgments, limited data sources, and straightforward modeling techniques. Owing to recent advances in data mining and machine learning, numerous data-driven methods are adopted to address the problem that occurred in traditional schemes, which demonstrate exceptional performance. In this paper, we present a comprehensive survey of data-driven methods for travel time estimation, encompassing application scenarios, spatial-temporal modeling approaches, and data representation learning techniques. To support and promote further research in this field, we provide a valuable list of open data sources and source codes, offering researchers a solid foundation for their future endeavors. Furthermore, this survey discusses emerging trends and key challenges faced by the research community, such as the integration of real-time data streams and the use of uncertainty estimation. We also explore the potential impact of these advancements on transportation systems, highlighting opportunities for improvement and innovation. To the best of our knowledge, this work is among the first to offer a comprehensive, in-depth review of data-driven methods for travel time estimation, providing researchers and practitioners with a valuable reference in the field.