Latent growth curve modeling provides a powerful and flexible tool for researchers to study individual differences in change as well as the correlates and predictors of change. Recent developments in estimation and hypothesis testing procedures are largely based on confirmatory structural equation approaches. In this article, an alternative exploratory approach is proposed for the analysis of growth and change using multidimensional scaling (MDS). When applied to growth data, it is a growth pattern recognition technique that partitions individual differences into initial level and growth pattern components. When applied to other longitudinal data, it can be used to study change patterns. A math achievement data set is used to illustrate the growth modeling method and a mood variable is used to illustrate change modeling. The strengths and limitations of the MDS growth profile analysis are discussed.