| dc.description.abstract | Guided by the theories of social identity (Tajfel & Turner, 1979; Turner, Hogg, Oakes, Re- icher, Wetherell, 1987) and intergroup contact (Allport, 1954; McIntyre, Paolini, Hewstone, 2016; Pettigrew, 1998; Pettigrew & Tropp, 2006; Williams, 1947), this study examines stereotype jokes about Asian Americans and stereotype agreement from Caucasian Americans’ perspectives. Sit- uated in the context of standup comedy, two factors were experimentally manipulated using a 2 × 2 within-subjects design: the racial group membership of the comedian (Asian/White) and the content valence of the joke (negative/positive stereotype). Four written scripts based on standup comedy routines and mass media messages were used for the experimental manipulations. These scripts are also in line with common stereotypes identified in prior empirical studies about Asian sojourners and immigrants in the United States (i.e., success driven and bad drivers). Data collection was conducted using Qualtrics online survey system. Following participants exposure to each of the four joke scenarios (randomly ordered), five dependent variables were measured: funniness and offensiveness of the joke, interpersonal anxiety toward the comedian, intergroup anxiety toward Asian Americans as a group, and level of agreement with the stereotype presented in the joke. Responses from 227 Amazon Mechanical Turk participants were collected. Following data screening, the final sample for data analysis was 220 (54 with missing values; 166 with complete responses). Analysis results from an univariate multilevel modeling approach (Bates, Mächler, Bolker, & Walker, 2015) showed that Caucasian American participants’ least preferred (most offensive) sce- nario was a White comedian telling a negative-stereotype (bad drivers) joke about Asian Americans (βmarginal = 4.421, p < .001). The most positively rated (least offensive) scenario was Asian Amer- ican comedian telling the positive-stereotype (success driven) joke (βmarginal = 2.825, p < .001). Furthermore, the participants rated Asian comedian telling a negative-stereotype joke to be the funniest (βmarginal = 4.513, p < .001) comparing to White comedian telling a negative-stereotype joke (least funny; βmarginal = 3.596, p < .001). Further analyses using a Bayesian multivariate multilevel approach (Bürkner, 2017; Carpenter et al., 2017) were conducted to examine potential mediation processes in between comedian group membership and stereotype agreement and in between content valence and stereotype agreement. Analysis results showed first-stage mediation effects of joke offensiveness (negative) and inter- personal anxiety (positive) between comedian group membership and stereotype agreement. In addition, two two-stage mediation processes were found: 1) from comedian group membership to stereotype agreement through interpersonal anxiety (positive) and intergroup anxiety (positive); and 2) from comedian group membership to stereotype agreement through joke offensiveness (pos- itive) and interpersonal anxiety (positive). For the potential mediation processes between content valence and stereotype agreement, anal- ysis results showed first-stage partial mediation effect of joke funniness (positive) between the two. A two-stage partial mediation process was found from content valence to stereotype agree- ment through joke offensiveness (positive) and interpersonal anxiety (positive). While accounting for the mediation paths, content valence still had a direct effect on stereotype agreement that the participants showed a higher level of agreement with the positive stereotype (Asian Americans are success driven) than with the negative stereotype (Asian Americans are bad drivers). The study findings are discussed in light of social identity, intergroup contact, racial discourse norms, and standup comedy as a means for stereotype agreement reduction. Theoretical implica- tions and contextual interpretations are addressed. Future research directions on stereotype humor in standup comedy and other communicative contexts are offered. | |
