Communication Differences: Koreans and American Cultures

Overall Logistic Regression Model

H1 stated that high-context Korean citizens would be more likely to use indirect communication strategies than their low-context US American counterparts. The overall model with country as the independent variable and indirect communication as thedependent variable was highly significant (p < .003). The main effect for country was significant with Nagelkerke R² = .05. Results showed that the odds of someone using indirect communication were 2.21 times higher if the person is Korean as opposed to a US American.

H2 stated that low-context US Americans are more likely to employ direct communication strategies than their high-context Korean counterparts. The overall model with country as the independent variable and direct communication as the dependent variable was also highly significant (p < .0001). Main effects for country were significant with Nagelkerke R² = .09.Results showed that the odds of someone using direct communication were .70 times lower if the person is Korean as opposed to a US American.

MANOVA

Results

Results showed that multivariate analysis was warranted because the multivariate main effect for country was significant (Wilk's l= .78, F(3, 233) = 21.99, p < .0001, h2= .22). There were significant univariate effects for intercultural communication apprehension (F(1,235) = 57.77, p < .0001, h2 = .20),and intercultural nonverbal immediacy (F(1, 235) =22.36, p < .0001, = .09). Verbal aggression results were not significant. Internal consistency reliabilities obtained for both the US and Korea together were acceptable for ICAS (r = .94), NIS (r = .82), and VAS (r = .79). The exact means and standard deviations can be found in Table 1.

In general, the ICAS mean in Korea was higher than the mean in the US. This result substantiated H3, that Korean citizens are more likely to be more communicatively apprehensive that their US American counterparts. H4, that US Americans will be more nonverbally immediate than Koreans was also supported because the mean for in NIS was lower for Koreans than the mean for US Americans.

VAS results were not significant; therefore, H5, that Koreans are more likely to use verbal aggression than their US American counterparts, was not supported. It was clear that the VAS Scale was reliable because the VAS demonstrated acceptable reliabilities with alpha coefficients of .79 for the combined US and Korean sample (n = 242), .75 for the Korean sample (n =147), and .83 (n= 118) for the US sample. The VAS Scale also had adequate power and an adequate sample size for this test because when this scale was factor analyzed,all samples warranted multivariate analysis given that all of the tested samples had KMO's greater than the usual criterion of .66 (i.e., US/Korea KMO = .844; Korea KMO = .743; and the US KMO = .836)