Latent change score models (LCSMs) proposed by McArdle (McArdle, 2000, 2009; McArdle & Nesselroade, 1994) offer a powerful tool for longitudinal data analysis. They are becoming increasingly popular in social and behavioral research (e.g., Gerstorf et al., 2007; Ghisletta & Lindenberger, 2005; King et al., 2006; Raz et al., 2008). Although conducting both univariate and multivariate latent change score analysis is not a difficult task any more (e.g., Ghisletta & McArdle, 2012; Zhang et al., 2015), there is little discussion on the design issues such as sample size planning for LCSMs. To fill the gap, this study proposes a Monte Carlo based method to determine the required sample size and the number of measurement occasions for both univariate and bivariate LCSMs. The method can obtain the power for testing each individual parameter of the models such as the change rate and coupling parameters. The Monte Carlo procedure is implemented and provided in a free R package RAMpath (Zhang et al., 2015). Examples for sample size and measurement occasion planning for both univariate and bivariate LCSMs are provided. [This report was published in: E. Ferrer, S. M. Boker, and K. J. Grimm (Eds.) "Advances in Longitudinal Models for Multivariate Psychology: A Festschrift for Jack McArdle."]