This chapter introduces multi-channel sequence analysis, a novel method that examines two or more synchronised sequences. While this approach is relatively new in social sciences, its relevance to educational research is growing as researchers gain access to diverse multimodal temporal data. Throughout this chapter, we describe multi-channel sequence analysis in detail, with an emphasis on how to detect patterns within the sequences, i.e., clusters —or trajectories— of multi-channel sequences that share similar temporal evolutions (or similar trajectories). To illustrate this method we present a step-by-step tutorial in R that analyses students’ sequences of online engagement and academic achievement, exploring their longitudinal association. We cover two approaches for clustering multi-channel sequences: one based on using distance-based algorithms, and the other employing mixture hidden Markov models inspired by recent research.