Ethnic diversity is often seen to be detrimental to peace and stability, particularly if ethnicity is the basis for political mobilization. Mobilization is assumed to increase the salience of ethnicity, and with it in-group cohesion, out-group animosity, and national instability; yet the mechanisms have rarely been studied empirically. This article argues that we need to study ethnicity as the outcome of political processes, focus on the attitudinal mechanisms underlying ethnic relation; and examine this phenomenon at the individual level. To this end, the article first disaggregates the term "ethnicity" into attributes, meanings, and actions. Referring to constructivism, it then argues that political science should focus on meanings. Building on the theory and findings of social psychology, this paper shows that political science must distinguish analytically between meanings regarding different in- and out-group processes. Doing so can help advance the study of ethnic relations and conflict-management practices.