Integrating leadership theory, role theory, social interaction theory, and cross-cultural psychology theory as the theoretical background, the present study proposes four role models that capture both cultural-common and cultural-specific aspects of the perception of managerial role behavior in three managerial activities: interaction management, decision making, and social influence. A survey instrument, "Superior-Subordinate Behavioral Differentials," was developed and used to examine perceptions of the four role models among managers in two cultures--that is, the United States and Hong Kong. Four models of the perception of managerial role behavior were found in both the U.S. group and the Hong Kong group: (1) "economic exchange/authoritarianism"; (2) "communal sharing"; (3) "authority ranking"; and (4) "equality." These were considered as cultural-common aspects of the perception of managerial role behavior. It was also found that, in each model, both the U.S. group and the Hong Kong group had their own different components which, as expected, reflected cultural-specific aspects of the perception of managerial role behavior. Comparing the perception of managerial role behavior, the Hong Kong managers were perceived as both more authoritarian and more communally sharing than the U.S. managers. Moreover, the Hong Kong sample scored high in "collectivism" and "Confucian work dynamics" (i.e., authoritarianism), whereas the U.S. sample scored high in "individualism." Across the four models of managerial role behavior, a significant country effect on the perception of these models was found repeatedly. Moreover, Hofstede's "power distance" appeared as a strong predictor for the variance of the perception of managerial role behavior across the cultural groups. The Chinese Value Survey, however, demonstrated a very limited predictive power. This study's findings have several implications for cross-cultural management studies and for cross-cultural management training and selection.

The person-situation congruence has been the subject of much psychological research. In this dissertation, a distinction was made between latent and manifest person-situation relations, a distinction which had never been addressed in the congruence literature. The dissertation investigated the psychological processes through which latent person-situation relations were transformed to their manifest form. Also examined was the relative importance of those two modes of person-situation relations in determining situational satisfaction and attachment. A heuristic model of supervisor-subordinate fit was proposed where the supervisor represented a situation for the subordinate. Latent and manifest modes of congruence were modeled in terms of interpersonal and work relations between the supervisor and the subordinate. Self-monitoring, expectancy, and situational force (self-efficacy, outcome expectation, ability conception, and situational controllability) were included in the model as mediators between latent and manifest interpersonal (work) relations. The interaction subjects consisted of 142 professor-graduate assistant pairs affiliated with a major eastern research university. Hierarchical regression analysis revealed that those subordinates having the sensitivity to social appropriateness (high self-monitors), those interacting with authoritarian supervisors (strong situations), and those perceiving positive incentives for having or continuing to maintain good interpersonal relations with their supervisors (high expectancy) modified their genotypical personality traits so that they fit their supervisors' personality requirements. The data also indicated that subordinates who were highly self-efficacious (high self-efficacy), those who perceived positive incentives to fulfill their work roles (high outcome expectancy), and those who viewed ability as an enhanceable process (incremental--as opposed to fixed--view of ability) exerted greater efforts to attain better fit with their work situations. Path analysis was used to examine the consequences of latent and manifest person-situation relations. The data indicated that latent person-situation relations affected situational attachment both directly and indirectly through situational satisfaction; for manifest person-situation relations, only the indirect path was confirmed. Variance decomposition analysis showed that latent person-situation relations predicted situational attachment significantly better than did manifest person-situation relations.