Data collected from India, Argentina and the U.S. served to arbitrate between competing predictions concerning the amount of agreement between nationals in depicting other nations. Amount of agreement was hypothesized to vary with friendliness, familiarity and type of adjective used to depict the " target " nation. Several findings were obtained in each of the three cultures, including: (a) a higher level of agreement for friendly (or familiar) than for unfriendly (unfamiliar) nations on evaluative traits; (b) considerably more agreement for friendly (or familiar) nations on evaluative than on descriptive traits, and (c) somewhat more agreement for unfriendly (or unfamiliar) than for friendly (familiar) nations on descriptive traits. These results reaffirm and extend previous findings. Other findings were culture-specific, including: (a) the relative salience of familiarity (more important in Argentina and the US.) and friendliness (more important in India) and (b) the higher levcl of agreement for unfriendly than for fricndly nations in Argentina. The first result clarifies interpretations of findings obtained in previous studies wbcre these dimensions were confounded while the second suggests that the ethnocentrism theory prediction for friendliness may be culturespecific.This study was designed to provide additional clarification for competing predictions concerning the amount of agreement among nationals in characterizing other nations. Most treatments of intergroup stereotyping would agree that in order for a perceived characteristic of a group to qualify as a stereotype, a certain level of agreement between individuals should be observed. The level of agreement has been referred to as the amount of stereotypy " l(e.g., Freund, 7 950; Lambert and Klineberg, 1967) or " stereotypical uniformity " (e.g., Katz and Braly, 1933;Berry, 1969). Three factors, in particular, hypothesized to affect the level of agreement are familiarity, friendliness and the type of trait used to depict the other nation.