Ethnocentrism is usually considered to be a strong and stable attitude. In most of the research it has been found that attitudes that are crystallized during preadult years will keep influencing the value patterns further on during life. Other researchers, however, doubt the persistence of political attitudes and argue that preadult experiences have little impact on adult perceptions. Previous research mainly focused on the evolution of ethnocentrism at an aggregate level. Context variables as unemployment rates and immigration flows are often used to explain changes in ethnocentrism. At the individual level, education and intergroup contact are theoretically relevant predictors of ethnocentrism. Higher educated individuals are less ethnocentric than lower educated individuals. Allport’s (1954) contact hypothesis suggests that intergroup contact reduces prejudice toward the outgroup. In this paper, we attempt to extend previous work on change in ethnocentrism theoretically and methodologically. Theoretically, by defining individual trajectories of change in ethnocentrism during preadulthood. At the aggregate level ethnocentrism can increase, decrease or stay the same over time. But within this general trend, individual trajectories may occur. If one wants to understand the mechanism of ethnocentrism it is important to define the causes of the different trajectories. We will investigate the effect of two important predictors on ethnocentric development: education and intergroup friendship. Methodologically, this paper wants to extend previous work by using advanced structural equation methods for longitudinal designs. The evaluation of individual change will be done using Latent Growth Curve Modeling. This technique allows to identify and explain variability in individual trajectories over time. Research questions - Can an adolescent’s attitude toward immigrants predict his/her attitude as a young adult? - Do adolescents experience different trajectories in the change of ethnocentrism? - Can these different developmental trajectories be explained by educational attainment and intergroup friendship? Data: In this paper we use the Belgian Political Panel Survey data for the period 2006-2011. In this panel, 3,000 adolescents were questioned about their social and political attitudes at three points in time. The first wave was conducted at the age of 16, the second wave at 18 and the third wave at 21.
