For young children, imitation serves both a learning role to gain knowledge and skills, and a social role to connect with people. Research on young children's imitative behavior has presented a dilemma: On one hand infants and young children engage in goaldirected imitation, in which they selectively copy the model's goal, or actions that are necessary for achieving the goal. On the other hand they engage in faithful imitation or "overimitation", in which they also copy the exact manner of actions, or actions that are apparently unnecessary for achieving the goal. In this dissertation I approach this dilemma from an individual difference perspective. Two cohorts of children (N = 48) visited the lab on 3 occasions. Each time, children were tested on two types of imitation tasks, as well as tasks measuring other aspects of development. Parents filled out questionnaires about their children. Results from children's three visit showed stable individual differences in children's imitative behavior both within and between different types of tasks. Correlations between imitation measurements revealed two factors: one for goal-directed imitation and one for means-directed imitation. These two factors are correlated but also distinctive: goal-directed imitation is associated with children's general developmental level, Theory-of-Mind and prosocial behavior at 24 and 30 months; means-directed imitation is associated with children's executive functioning and normative reasoning at 36 months. In terms of developmental trajectory, there is a significant increase in means-directed imitation between 30 and 36 months of age, which coincide with an increase in children's normative reasoning. These results are discussed in terms of their implications for understanding children's social learning mechanisms, and also in terms of continuity in individual differences among infants and young children.