Social status and anger expression: the cultural moderation hypothesis

Abstract

Individuals with lower social status have been reported to express more anger, but this evidence comes mostly from Western cultures. Here, we used representative samples of American and Japanese adults and tested the hypothesis that the association between social status and anger expression depends on whether anger serves primarily to vent frustration, as in the United States, or to display authority, as in Japan. Consistent with the assumption that lower social standing is associated with greater frustration stemming from life adversities and blocked goals, Americans with lower social status expressed more anger, with the relationship mediated by the extent of frustration. In contrast, consistent with the assumption that higher social standing affords a privilege to display anger, Japanese with higher social status expressed more anger, with the relationship mediated by decision-making authority. As expected, anger expression was predicted by subjective social status among Americans and by objective social status among Japanese. Implications for the dynamic construction of anger and anger expression are discussed.

Figure 1

The relationships between anger expression and social status (A: subjective, B: objective) for Americans (solid line) and Japanese (dashed line). Americans with lower subjective social status expressed more anger than Americans with higher subjective social status, b = −.22, t(2002) = −3.71, p < .001. In contrast, Japanese with higher objective social status expressed more anger than Japanese with lower objective social status, b = .64, t(2002) = 5.15, p < .001. Statistical significance is indicated by asterisks (***p < .001).

Figure 2

Results of mediation analyses examining the role of frustration (Panel A) and decision authority (Panel B) in mediating the relationship between social status and anger expression for Americans (N = 1035) and Japanese (N = 728), respectively. Unstandardized coefficients are shown. The values in parentheses show the relationship between subjective (or objective) social status and anger expression after controlling for frustration (or decision authority). The values in square brackets are 95% bias-corrected confidence intervals from a bootstrap test with 2000 replications; the mediation is significant if the confidence interval does not include zero. Note. *p < .05, **p < .01, ***p < .001.