Random Intercept Latent Transition Analysis(RI-LTA)——Separating the Between-subject Variation from the Within-subject Variation Abstract: The traditional latent transition analysis belongs to single-level model, but it can also be viewed as a two-level model from a multi-level perspective. In 2020, Muthén and Asparouhov proposed a so-called random intercept latent transition analysis (RI-LTA) model separating the between-subject variation from the within-subject variation. By integrating a random intercept factor, the latent class transitions are represented on the within level whereas the between level captures the variability across subjects, which avoids the overestimation on the transition probabilities of staying in the original class and allows a clearer interpretation of the data. This article resumes the proposition and principles of the random intercept latent transition analysis, uses a prestigious research-oriented university undergraduates’ academic motivation longitudinal data to demonstrate how to implement random intercept latent transition analysis in Mplus software. By comparing the results of traditional LTA and RI-LTA, we find out that the transition probability of staying in the second positive class is overestimated. This proves that RI-LTA model is better than traditional LTA model in this situation. We also include the covariate “enrolment way to university” in the model to accomplish a multiple-group analysis purpose and investigate the interaction effects between the covariate and c1 on c2. At the end of this article, the advantages and future explorations of this new technique are discussed. Key words: latent transition analysis; random intercept latent transition analysis; single-level analysis; multi-level analysis; monte carlo simulation study; The mplus syntax files have already been uploaded to this project. You can regard them as references. We hope that these syntax can help your study.