Social identities, conceptualized as self-designations and measured by the TST, were examined for samples of high school adolescents in three societies: the United States, Puerto Rico, and Mexico. Four identities were explored in terms of salience, frequency, and valence: gender, religion, family, and peer. For both males and females in Latin and Anglo cultures gender emerged as the most prominent Religious IDs were more frequent for Catholic adolescents. The strongest cultural difference was found with respect to negative religious IDs: these were significantly more frequent for Anglo adolescents. Positive gender and family IDs were more frequent for Latin adolescents, while peer IDs were slightly more common self-designations for Anglos. These tendencies were generally in the expected direction. Social and cultural differences between these Anglo and Latin societies were considered as explanations for variations in adolescent identity structures. The concept of identity, as used by social psychologists of a sociological bent, refers to the locatioin of an individual in social space (Goffman, 1959; Stone, 1962; Strauss, 1959). This social space consists of the myriad of statuses and roles which a society provides to its members. Their selective internalization and integration form the structure of a person's self-concept. As Stone (1962:93) pointed out, identity establishes what and where the person is in social terms: When one has identity, he is situated-that is, cast in the shape of a social object by the acknowledgment of his participation or memberships in social relations. Similarly, Kuhn and McPartland (1954) viewed the self-concept as deriving its consistency and structure from social affiliations. From this perspective, a person may (and usually does) have many identities depending on the diversity of social relations in which he engages. Some identities are more important than others depending on the degree to which an individual is committed to them; the amount of the that is encompassed by them; and the number of social situations in which they are relevant. The structure of the self can be viewed as the hierarchical organization of a person's identities (cf. Gordon, 1968; Kuhn and McPartland, 1954; McCall and Simmons, 1966). Identity conceptualized in these terms is more accurately termed social identity, to distinguish it from personal identity (see McCall and Simmons, 1966; Sarbin, 1970, for this distinction). Social identities are derived from a person's group memberships and roles. They are expressed through broad social categories such as occupation, sex, religion, family, etc. Personal identity, on the other hand, typically refers to self-definitions in terms of unique characteristics.' This is similar to Kuhn and McPartland's (1954) distinction between consensual and subconsensual identities. * This research was supported by NSF Grant GS 2650; a Woodrow Wilson Dissertation Fellowship; and by Project 1743, Department of Rural Sociology, College of Agriculture, Washington State University as scientific paper No. 3905, Washington Agricultural Experiment Station. 1 Another usage of the concept of identity as a personal experience is exemplified in the work of Erik Erikson. Erikson (1956:60) views identity as a highly subjective, personal experience which the individual develops about who he is. He discusses its expression in at least four distinct ways: At one time . . . it will appear to refer to a conscious sense of individual identity; at another, to an unconscious striving for a continuity of personal character; at a third, as a criterion for the silent doings of ego synthesis; and finally, as maintenance of an inner solidarity with a group's ideals and identity.