Abstract
The COVID-19 pandemic exacerbated anti-Chinese sentiments, demonstrating an urgent need for effective prejudice reduction interventions. Study 1 tested the relationship between contact and prejudice and its mediators toward Chinese people in Australia ( N = 362). As predicted, positive contact was significantly associated with less prejudice and, reliably, intergroup anxiety had the strongest mediation effect in this relationship. In Study 2, European Australians ( N = 161) were randomly assigned to an intergroup or intragroup (control) contact interaction, delivered vicariously via audio or text. As hypothesized, vicarious intergroup contact significantly reduced prejudice and negative emotions toward Chinese people. Moreover, text-based vicarious contact led to significantly lower intergroup anxiety and negative intergroup emotions than audio-based contact. No interaction effects of vicarious contact and delivery mode were found. Overall, these studies support direct and vicarious text-based contact as effective strategies for reducing prejudice in a global crisis and improving attitudes toward Chinese people.
From the U.S. Chinese Exclusion Act (1882) to the White Australia policy (1901–1958), which disproportionately restricted Chinese migration, racial prejudice toward Chinese people has been ever present in Western history. Current anti-Chinese prejudice is underpinned by a myriad of perceived threats (Gao, 2022). For example, China is perceived as a geopolitical threat to the West, with ongoing trade wars and border disputes further enflaming anti-Chinese prejudice in the United States and Europe (Gao, 2022; Gries & Turcsányi, 2021; He & Xie, 2022; Li & Nicholson, 2021; Tan et al., 2022). More recently, the COVID-19 pandemic brought about an associated rhetoric labelling it the “China virus” or “kung flu,” implying that Chinese individuals also pose a health threat and, by extension, an economic (realistic) threat brought on by the pandemic (Borja et al., 2020; Budhwani & Sun, 2020; Dhanani & Franz, 2021; Yi, 2023). These sentiments, largely perpetuated by media channels, have been found to exacerbate entrenched implicit and explicit racial prejudice toward Chinese people (Darling-Hammond et al., 2020; Reny & Barreto, 2022; Shi et al., 2022; Song, 2023).
An analysis of the U.S. 15 largest cities revealed that hate crimes such as assault and vandalism targeting Asians, particularly those of Chinese descent, increased by nearly 170% during the COVID-19 pandemic (Levin, 2021). Similar increases in anti-Chinese prejudice since this time have also been reported in Australia (Biddle et al., 2020), Canada (Chinese Canadian National Council for Social Justice, 2022), and the United Kingdom (Tam, 2021). These intergroup frictions have resulted in physical assault, verbal harassment, and bullying toward Chinese individuals (Gries & Turcsányi, 2021; Gover et al., 2020). Furthermore, it is predicted that Western concerns about China’s economic power and the economic threat of the pandemic with the rising cost of living will continue to exacerbate anti-Chinese sentiments well beyond the pandemic (Ang & Colic-Peisker, 2022; Dhanani & Franz, 2021; Gao, 2022).
The vast evidence of anti-Chinese prejudice is concerning given that such marginalization is predicted to lead to various systemic issues, including poorer mental health, disadvantaged socioeconomic status, and, ultimately, reduced quality of life (Misra et al., 2020; T. T. Nguyen et al., 2020; Tessler et al., 2020; Wu et al., 2021). This is especially important in Australia, where people with Chinese ancestry comprises 5.5% of the total population, the largest racial minority group in Australia (Australian Bureau of Statistics, 2022; Hsu, 2023). To address this global and national issue, the present research aimed to shed light on whether and how intergroup contact may be a useful strategy to reduce anti-Chinese prejudice in Australia during the COVID-19 pandemic. Specifically, Study 1 explored the mechanisms that drive the contact–prejudice reduction relationship, and Study 2 experimentally examined the utility of vicarious contact as a potential solution, particularly the relative effectiveness of audio versus text modes of delivery in improving attitudes toward Chinese people.
The Role of Positive Contact During a Crisis
Intergroup contact is currently one of the most empirically tested and validated prejudice reduction strategies (Paluck et al., 2021). Current research has demonstrated promising evidence that direct contact is effective in improving attitudes toward Chinese people in Malaysia (Al Ramiah et al., 2014), Malawi (Gu et al., 2015), and the United Kingdom (Alston et al., 2022). However, much of this research occurred before the health and economic threats of COVID-19 were understood, and the subsequent surge in anti-Chinese prejudice and rhetoric became more widespread (Croucher et al., 2020; Pan & Gao, 2021; Ruiz et al., 2021; Silver et al., 2020). To date, there is barely any research that has examined contact’s effects during a global health crisis (e.g., COVID-19) toward the targeted outgroup (Adler et al., 2022). In support of this, a recent meta-analysis examining the benefits of contact in the face of threat outlined that more research is still needed to understand the role of unique historical and political (i.e., external) contexts as potential limiting conditions for contact (Van Assche et al., 2023). This is especially the case when health and financial threats due to the pandemic have been shown to predict unwillingness to help and the dehumanization of outgroups (Adler et al., 2022).
The few research studies that have examined contact in historical and political crises have been mixed. For example, some researchers have argued that threatening contexts via major crises might in fact reduce the effectiveness of contact in bringing positive attitude changes (Ditlmann & Samii, 2016; Mousa, 2020; Paluck & Clark, 2020; Stephan & Renfro, 2002). Specifically, a study testing Iraqi Christians’ attitudes toward Muslims during a political crisis found that it remained difficult for Christians to generalize improved attitudes toward the Muslim outgroup after contact (Mousa, 2020; Paluck & Clark, 2020), while another study found that positive contact had a negative effect on outgroup attitudes toward Jewish Israelis amongst Arab Palestinians (Ditlmann & Samii, 2016). On the other hand, Abrams et al. (2017) demonstrated that, after the 2005 London terrorist attack, positive contact was associated with lower prejudice toward Muslims in the UK despite an increase in prejudice.
A possible reason for this inconsistency could be due to the fact that Abrams et al. (2017) investigated attitudes toward Muslims as immigrants in a Western country (i.e., UK) while Mousa (2020) and Ditlmann and Samii (2016) investigated contact in threatening contexts driven by significant power imbalances over a long period of time. Similar to Abrams et al.’s (2017) study, our research examined attitudes toward Chinese people as immigrants in a Western country (i.e., Australia) during a health crisis over a shorter period of time, where it is plausible that contact remains an effective prejudice reduction strategy in such a context. By reexamining the contact–prejudice relationship, given the complex interplay of threats contributing to anti-Chinese prejudice today, we aim to contribute to a larger body of research investigating the factors and contexts that may act as boundary conditions to the efficacy of contact on prejudice. As it stands, the empirical question remains whether contact continues to be as effective in reducing anti-Chinese sentiments in light of a large-scale health crisis.
The Strongest Mediator of the Contact–Prejudice Relationship
Pettigrew and Tropp’s (2008) meta-analysis of 515 contact studies showed that contact successfully reduced prejudice via the mediating mechanism of lowering intergroup anxiety, the emotion felt when anticipating or engaging in intergroup contact, and increasing outgroup empathy, the ability and willingness to take the perspective of an outgroup member. More recently, Banas et al. (2020) supported that intergroup anxiety and outgroup empathy are significant mediators of both positive and negative indirect contact. Another potential mediating factor is interpersonal closeness (Stefaniak & Bilewicz, 2016; Turner et al., 2008). It is thought that experiencing positive contact with an outgroup member can reduce the cognitive distance between the individual and the outgroup, thus reducing prejudice (Turner et al., 2008). A recent study examining interpersonal closeness posits that major crises (e.g., the COVID-19 pandemic) can incite greater closeness in the ingroup toward the outgroup, resulting in reduced prejudice toward Chinese people (Valtorta et al., 2022). Extrapolating from this research, intergroup anxiety, outgroup empathy, and interpersonal closeness may be key drivers underpinning the contact–prejudice reduction relationship in a crisis.
While the mediators of the contact–prejudice reduction relationship have been extensively explored, research comparing their relative effectiveness across different dimensions (i.e., affective, cognitive, behavioral intentions) of prejudice remains limited. So far, research has shown that high intergroup anxiety is a consistent driver of prejudice toward ethnic groups perceived as dangerous or competitive (Hutchison & Rosenthal, 2011; Stephan, 2014). According to Stephan and Stephan (1985), intergroup anxiety is most prevalent in an intergroup context where there is a high perception of danger and threat from the outgroup. Indeed, with the growing anti-Chinese prejudice due to the threat of a global crisis, intergroup anxiety has already been highlighted as a key predictor of anti-Chinese sentiments at the onset of the COVID-19 pandemic (Alston et al., 2022; Croucher et al., 2020; Gao, 2022). As such, it is plausible that lowered intergroup anxiety will remain the strongest mediator of the contact–prejudice reduction relationship in today’s anti-Chinese prejudice landscape, regardless of the outcome variables’ dimensions of prejudice. Moreover, research has yet to compare the effect of interpersonal closeness in major crises on other known mediators (Valtorta et al., 2022). By comparing the relative mediating role of intergroup anxiety in the contact–prejudice relationship across different dimensions of prejudice, especially in the context of a global crisis, we aim to identify key mechanisms that can better inform effective future contact interventions to be used in crises and beyond.
Study 1
Aims and Hypotheses
The aims of Study 1 were to establish whether contact continued to be a relevant factor amid the levels of anti-Chinese sentiment during the COVID-19 pandemic, and investigate the relative strengths of intergroup anxiety, outgroup empathy, and interpersonal closeness as mediators of the contact–prejudice relationship. It was predicted that among European Australian participants:
Method
Participants
The final study sample, collected in late 2021 and early 2022 during the COVID-19 pandemic, consisted of 173 psychology Australian undergraduates and 189 Australian community members (N = 362) after excluding participants that failed the attention check (n = 24) and the suspicion check (n = 9). The ages ranged from 18 to 49 (Mage = 21.36 years, SDage = 4.05). Two hundred and twelve participants (58.6%) identified as female, 147 as male, and three as nonbinary. Two hundred and twenty participants (58%) identified as not religious, 128 as Christian, 15 as Jewish, and nine identified with other religions. All participants identified as European Australian.
Measures
Positive contact
Hayward et al.’s (2017) intergroup contact measure was adapted for participants to indicate whether they had had a previous interaction with a Chinese person (0 = no, 1 = yes), how frequently they experienced interactions with Chinese individuals on a 7-point Likert scale (1 = extremely rarely, 7 = extremely frequently), and how positively they would rate these interactions (1 = not at all positive, 7 = very positive). These scores were combined by multiplying the quantity and quality of contact, as per Hayward et al. (2017) and Voci and Hewstone (2003), to make a composite overall score of positive contact, with higher scores indicating greater positive contact (Cronbach’s α = .76).
Outgroup attitudes
Outgroup attitudes were assessed using Burke et al.’s (2017) single-item feeling thermometer, “How warm do you feel toward Chinese people?” (0 = extremely cold, 100 = extremely warm). Higher scores indicated a more positive affective attitude.
Realistic threat
Maddux et al.’s (2006) modified nine-item measure was used to assess realistic threat. Participants were asked to rate the extent to which they agreed that “Chinese people have more economic power than they deserve,” on a 7-point Likert scale (1 = strongly disagree, 7 = strongly agree). Higher scores indicated greater threat (Cronbach’s α = .71).
Social distance
Crawford et al.’s (2017, Study 2) seven-item social distance measure was administered to determine participants’ behavioral intentions (e.g., “I would be happy to have a Chinese person living in Australia [as a roommate/marry into the family]”). Each item was rated on a 7-point Likert scale (1 = strongly disagree, 7 = strongly agree). Higher values indicated greater prejudice (Cronbach’s α = .92).
Modern racism
Son Hing et al.’s (2008) nine-item Asian Modern Racism Scale was adapted for participants to rate items such as “Discrimination against Chinese people is no longer a problem in Australia,” on a 7-point Likert scale (1 = strongly disagree, 7 = strongly agree). Higher scores reflected greater prejudice (Cronbach’s α = .93).
Interpersonal closeness
Interpersonal closeness was measured using Aron et al.’s (1992) single-item pictorial Inclusion of the Other in the Self. Participants selected the picture that best represented their perception of themselves in relation to the outgroup from a set of Venn diagrams, each representing different degree of overlap from 1 to 7, with higher scores denoting closeness.
Outgroup empathy
Pedersen et al.’s (2004) five-item empathy measure was adapted to assess participants’ empathy levels across a 7-point Likert scale (1 = strongly disagree, 7 = strongly agree); for example, whether they “often feel sympathy” toward the outgroup. Higher scores were coded to indicate more empathy (Cronbach’s α = .75).
Intergroup anxiety
An adapted version of Stephan and Stephan’s (1985) six-item Intergroup Anxiety Scale was used to gauge participants’ affect (e.g., “anxious,” “comfortable”) when interacting with the target group. Participants rated their anxiety toward Chinese people on a 7-point Likert scale (1 = strongly disagree, 7 = strongly agree), with higher scores indicating greater anxiety, after reverse-scoring relevant items (Cronbach’s α = .89).
Demographic questionnaire
The questionnaire asked participants about their gender, age, residency status, ethnic background, and religion.
Attention check
One item tested participants’ attention during the study. Participants were asked to select “disagree” on a 7-point Likert scale (1 = strongly disagree, 7 = strongly agree).
Suspicion check
To reinforce the cover story and probe for suspicion, participants were asked whether they would like to comment on anything about the “two studies” and whether they noticed anything unusual.
Procedure
Advertisements of the study were posted on the University of Sydney’s online recruitment research portal and on social media (e.g., Facebook). All measures were completed online via Qualtrics. All participants took part in the study voluntarily. Once consent was given, participants completed the demographic questionnaire. Participants were then asked to complete a fake random selection task to determine which social group out of five they would be asked questions on. Unbeknownst to participants, they were always completing questionnaires assessing their attitude toward Chinese people living in Australia. The order of all outcome measures in the questionnaire was randomized, with the attention check item randomly placed amongst the measures. At the end of the questionnaire, participants were given the opportunity to express any suspicions on the true nature of the study, and were then debriefed with the true aims of the study.
Results
Preliminary analyses
A Pearson’s correlation analysis was conducted to determine whether the relationships between the main variables were theoretically and conceptually consistent. As reported in Table 1, most variables were significantly correlated in the expected direction, except for interpersonal closeness with realistic threat and modern racism.
Means, standard deviations, and Pearson’s correlations of main variables.
Note. Higher scores for social distance, intergroup anxiety, realistic threat, and modern racism indicate greater racial prejudice. Higher scores for outgroup attitudes, outgroup empathy, and interpersonal closeness indicate lower racial prejudice.
*p < .05. ** p < .01 (two-tailed).
Main analyses
Linear regressions, using SPSS, were conducted on the mediator and outcome variables (see Table 1 for means and standard deviations). In support of H1, positive contact with Chinese individuals was associated with less prejudice on all of the outcomes variables: outgroup attitude: b = 0.31, SEb = 0.07, t(360) = 4.16, p < .001; interpersonal closeness: b = 0.03, SEb = 0.01, t(360) = 5.83, p < .001; social distance: b = −0.02, SEb = 0.00, t(360) = −4.74, p < .001; outgroup empathy: b = 0.01, SEb = 0.00, t(360) = 3.96, p < .001; intergroup anxiety: b = −0.13, SEb = 0.00, t(360) = −3.32, p < .001; and modern racism: b = −0.01, SEb = 0.00, t(360) = −2.11, p = .036. And positive contact was not found to be associated with realistic threat, b = −0.00, SEb = 0.00, t(360) = −1.20, p = .232.
To test H2 and H3, mediation path analyses using Hayes’s (2018) PROCESS macro (Model 4) were conducted on the mediator variables intergroup anxiety, interpersonal closeness, and outgroup empathy, and on their effects on the contact–prejudice relationship (i.e., outgroup attitude, social distance, modern racism, and realistic threat). As per the typical statistical procedure (see Rosenthal & DiMatteo, 2001), standardized β was used to compare the relative effect of the mediators on the four outcome variables.
Consistent with H2 and H3, the indirect mediation effect of positive contact on outgroup attitudes through intergroup anxiety, β = .08, SEβ = 0.03, 95% CI [0.03, 0.13]; interpersonal closeness, β = .06, SEβ = 0.02, 95% CI [0.03, 0.09]; and outgroup empathy, β = .03, SEβ = 0.02, 95% CI [0.00, 0.06], was significant, with intergroup anxiety having the strongest mediation effect (see Figure 1a).
Parallel multiple mediation models with standardized regression coefficients for the positive contact–outcome relationship as mediated by intergroup anxiety, outgroup empathy, and closeness for (a) outgroup attitudes, (b) social distance, (c) modern racism, and (d) realistic threat. *p < .05. **p < .01. ***p < .001.
Consistent with H2 and H3, the indirect mediation effect of positive contact on social distance through intergroup anxiety, β = −.07, SEβ = 0.02, 95% CI [−0.12, −0.02]; interpersonal closeness, β = −.03, SEβ = 0.01, 95% CI [−0.06, −0.01]; and outgroup empathy, β = −.03, SEβ = 0.02, 95% CI [−0.07, −0.01], was significant, with intergroup anxiety having the strongest mediation effect (see Figure 1b).
The indirect mediation effect of positive contact on modern racism through intergroup anxiety, β = −.06, SEβ = 0.02, 95% CI [−0.11, −0.02], was significant. However, outgroup empathy, β = −.01, SEβ = 0.01, 95% CI [−0.04, 0.01], and interpersonal closeness, β = −.01, SEβ = 0.02, 95% CI [−0.04, 0.03], were not significant for modern racism. Thus, H2 was partially supported for modern racism. Nevertheless, supporting H3, intergroup anxiety was found to have the strongest mediation effect on modern racism (see Figure 1c).
Finally, the indirect mediation effect of positive contact on realistic threat through intergroup anxiety, β = −.07, SEβ = 0.02, 95% CI [−0.12, −0.02], and outgroup empathy, β = −.03, SEβ = 0.01, 95% CI [−0.06, −0.01], was significant. However, interpersonal closeness, β = .03, SEβ = 0.02, 95% CI [−0.01, 0.06], was not significant for realistic threat. Thus, H2 was partially supported for realistic threat. Consistent with H3, intergroup anxiety had the highest mediation effect on realistic threat (see Figure 1d).
Overall, H1 and H2 were mostly supported, and H3 was fully supported.
Discussion
Considering the COVID-19-related threat posed by the Chinese diaspora during the pandemic, the contact–prejudice relationship and its mediators were reinvestigated. Study 1 results supported the predictions that greater positive contact with Chinese individuals would be associated with lower levels of prejudice, and that this relationship would be mediated by intergroup anxiety and partially mediated by outgroup empathy and interpersonal closeness. This finding suggests that a health crisis may not compromise the positive effects of contact as much, compared to other crises that have yielded mixed results (Ditlmann & Samii, 2016; Mousa, 2020). Only realistic threat was not significantly associated with positive contact. Realistic threats are those that endanger the physical or material well-being of the individual (Kachanoff et al., 2020). This latter finding may be explained by the salience of existing health and economic threats toward the well-being of participants during the pandemic (Adler et al., 2022), ultimately making it difficult for them to perceive Chinese people as less threatening, even with more positive contact (Ditlmann & Samii, 2016; Mousa, 2020; Paluck & Clark, 2020).
Intergroup anxiety was also found to be the most significant mediator in the contact–prejudice relationship across all three dimensions of prejudice (i.e., affective, cognitive, and behavioral intentions), whereas interpersonal closeness had the weakest mediating effect on the contact–prejudice relationship. Together, these findings support (a) the continuing theoretical and empirical relevance of the contact–prejudice relationship, (b) that a global health crisis does not limit the positive effects of contact, although a more enduring intervention might be needed to reduce any perceived realistic threats, and (c) that intergroup anxiety is a key mediator in the contact–prejudice relationship, regardless of the dimension of prejudice in question.
These findings should, however, be interpreted within the context of the limitations of Study 1. First, contact was self-reported and not manipulated, limiting causal inferences about the relationship between contact and anti-Chinese prejudice. Second, the study did not measure each of Allport’s (1954) facilitating conditions for contact, a critical design issue identified by experts in the field (Paluck et al., 2019; Pettigrew & Tropp, 2006). Finally, Study 1 only measured direct positive contact with Chinese individuals, despite direct contact being difficult to achieve due to crisis-related distancing and in general (White, Borinca, et al., 2020). Thus, investigation of indirect contact, such as vicarious contact, is needed. Indirect contact strategies might be especially relevant to reduce prejudice in crises that encourage avoiding social interactions. For example, in our research context, since the outbreak of COVID-19, social distancing was actively encouraged due to health concerns and, thus, research predicted direct contact could potentially exacerbate hostility toward Chinese people, who were stigmatized as being those spreading the virus (Barrero et al., 2022; Tsai et al., 2020).
To this end, Study 2 was conducted to overcome the limitations of self-reported contact by experimentally manipulating vicarious contact and exploring the efficacy of different modes of this type of contact (text vs. audio) as a potential strategy to reduce intergroup anxiety and increase outgroup empathy as mechanisms to reduce anti-Chinese prejudice.
Study 2
Vicarious Contact and Prejudice Reduction
One way to overcome the psychological (i.e., anxiety) and physical (i.e., segregation) barriers to direct contact is via indirect contact (White, Borinca, et al., 2020). A growing body of research has shown indirect contact to be effective in improving outgroup attitudes (Lemmer & Wagner, 2015; White, Borinca, et al., 2020). Moreover, extant literature has demonstrated that a potent factor related to the rise in negative attitudes toward Chinese people is the media’s stigmatizing language and racial bias (Croucher et al., 2020; Song, 2023; Tessler et al., 2020). With the anti-Chinese prejudice widely spread by mass media (Song, 2023; Tessler et al., 2020), it is important to consider a strategy whose processes match such media dissemination.
One such strategy is vicarious contact. Vicarious contact refers to merely observing a positive cross-group interaction, which can reduce intergroup prejudice (Brown & Paterson, 2016; Wright et al., 1997). Vicarious contact is asynchronous and passive and, thus, can easily occur either by listening to the radio (Paluck, 2009) and/or reading texts (Vezzali et al., 2014; Wojcieszak & Warner, 2020). Currently, there is a plethora of evidence demonstrating the effectiveness of vicarious contact in improving attitudes toward various outgroups (Ortiz & Harwood, 2007; Walker & Scior, 2013), including individuals of Chinese descent (Mazziotta et al., 2011). Given that the media has been shown to play a crucial role in the formation of negative attitudes toward Chinese people, and considering that vicarious contact can be operationalized in a similar format to media presentations and also widely disseminated through mass media channels, vicarious contact has the potential to be a genuine strategy for improving attitudes toward Chinese people (Tessler et al., 2020, Vezzali and Stathi, 2020; White, Borinca, et al., 2020).
Additionally, with anxiety being a key predictor of anti-Chinese sentiments (see Study 1), vicarious contact offers a more distal form of contact in the continuum of contact framework, which in turn may induce less anxiety compared to more “active” and synchronous strategies such as E-contact (White, Borinca, et al., 2020; White, Harvey, & Abu-Rayya, 2015). Nevertheless, empirical questions remain about the most efficacious mode (audio or text) of vicarious contact for reducing prejudice, and whether the mediating factors identified in Study 1 would contribute to the effectiveness of different vicarious contact modalities.
Another consideration when implementing vicarious contact is the inclusion of Allport’s (1954) four optimal conditions. There is evidence, for instance, that the intentional inclusion of cooperation and authority support significantly increases the positive effects of vicarious contact (Gómez & Huici, 2008). Despite such evidence, however, there is currently a dearth of contact research fully operationalizing all four conditions (Paluck et al., 2019).
Modes of Vicarious Contact
One major advantage of vicarious contact is its ability to be widely disseminated through different modes of delivery (Gómez & Huici, 2008; White, Borinca, et al., 2020). While most laboratory studies have used visual media (e.g., film, TV; Preuß & Steffens, 2021), adopting a broader definition of observation has allowed vicarious contact to be operationalized either via a text-based format or an audio-based format (Vezzali et al., 2014).
Text-based vicarious contact facilitates the observation of positive cross-group interactions through words (Vezzali et al., 2014). It has generally been tested in schools using specially created stories (Di Bernardo et al., 2017; Husnu et al., 2018; Tercan et al., 2021), but it has also been tested in adult populations via mass media communications. For instance, research showed that reading about a positive interaction between a progressive and a conservative individual in a newspaper reduced social distancing and improved perceptions of the political outgroup (Wojcieszak & Warner, 2020).
A less explored mode is audio-delivered vicarious contact. Only a limited number of studies have investigated the effectiveness of an audio-based approach (Bilali & Vollhardt, 2013; Bilali et al., 2016; Paluck, 2009, 2010). Using an educational radio soap opera featuring friendships between two rival ethnic groups in postgenocidal Rwanda, Paluck (2009) induced long-term improvement in listeners’ outgroup empathy, social beliefs, and cooperative intergroup actions. Further, this radio soap opera also positively influenced postconflictual Congolese participants’ levels of outgroup mistrust, historical perspective-taking (i.e., willingness to examine the outgroup’s point of view), and competitive victimhood (i.e., competing for victim status; see Bilali & Vollhardt, 2013). Indeed, current evidence suggests that both text-based and audio-based vicarious types of contact have their merits in reducing intergroup tensions.
Surprisingly, experimental comparisons of different modes of delivery of vicarious contact are scarce. Yet, in our information-rich mass media society, information is constantly disseminated through various media modalities. Apart from two correlational studies (Vezzali et al., 2015, Studies 2 and 3), only one study to date has experimentally compared video- and text-based interventions (Cocco et al., 2021). Thus, there is a genuine need to explore the relative effectiveness of audio- and text-based contact to better determine the best delivery method for future mass media vicarious contact strategies to improve outgroup attitudes (Vezzali & Stathi, 2020).
It is possible that audio-based vicarious contact, including the accents of outgroup members, may be more anxiety-provoking than text-based interactions (White, Harvey, & Abu-Rayya, 2015). Previous research has shown that communication barriers, such as accents, are associated with negative affective reactions (e.g., frustration), which in turn are associated with intergroup anxiety (Spencer-Rodgers & McGovern, 2002). Further, auditory information has been found to play a central role in helping people achieve social categorizations (Giles & Coupland, 1991), with research demonstrating the robust effect of hearing accents on making group membership salient (Rakić et al., 2011). Consequently, interactions that are seen as intergroup in nature are more anxiety-provoking than interpersonal ones (Amichai-Hamburger & McKenna, 2006). In one study, American participants listening to a Nigerian English accent experienced higher intergroup anxiety and a lower perception of the speaker’s social status than listening to an American English accent (Acheme, 2022). As concluded in Study 1, intergroup anxiety was a particularly powerful mediator of the contact–prejudice relationship toward Chinese people during the pandemic. Increased intergroup anxiety is positively associated with negative intergroup outcomes (Stephan, 2014), such as a higher likelihood to apply outgroup stereotypes (Bodenhausen & Wyer, 1985) and resistance to changing prior negative outgroup beliefs (Amichai-Hamburger & McKenna, 2006; Wilder & Shapiro, 1989).
Furthermore, much like electronic contact (E-contact; White, Maunder, & Verrelli, 2020), text-based vicarious contact can conceal outgroup features (physical or vocal) that may be too threatening for some individuals. It also allows ingroup members to read and experience the outgroup interaction at a speed that is comfortable for them, potentially leading to greater contact quality and therefore greater improvement in intergroup attitudes (Dovidio et al., 2017). To the extent that reduced anxiety and quality of contact predict more positive intergroup relations (Dovidio et al., 2017; Voci & Hewstone, 2003), text-based vicarious contact could be less anxiety-provoking and therefore provide a more robust approach to prejudice reduction than an audio-based intervention. Thus, by evaluating the differences between modes of delivery, we will be able to unpack the technical features of audio- versus text-based vicarious contact that might influence the effectiveness of intergroup contact in improving racial attitudes.
Anxiety and Empathy as Mediators of Vicarious Contact
In Study 1, we demonstrated that intergroup anxiety and outgroup empathy are strong mediators of the direct contact–prejudice relationship, with anxiety being the strongest mediator. Whereas previous research has shown mixed findings when it comes to anxiety as a mediator in the vicarious contact–prejudice relationship (e.g., Vezzali et al., 2019), recent research has shown that affective factors such as empathy and anxiety do mediate the vicarious contact–prejudice relationship, mirroring the results for direct contact (Preuß & Steffens, 2021). In fact, vicarious contact, compared to direct contact, has been shown to produce greater reductions in anxiety for future contact (Ioannou et al., 2018). Clearly, more robust empirical evidence is needed to continue to investigate the role of affective mediators, especially during experimentally manipulated vicarious contact in the reduction of anti-Chinese prejudice (Alston et al., 2022; Croucher et al., 2020).
Aims and Hypotheses
There has been a dearth of literature exploring the relative effectiveness of the different modes of vicarious contact (Cocco et al., 2021). Study 2 will be the first to address this whilst systematically implementing all of Allport’s optimal conditions to significantly increase the positive effects of contact (Pettigrew & Tropp, 2006). Further, Study 2 also extends the findings of Study 1 by examining the role of the two strongest mediators—intergroup anxiety and outgroup empathy—in the relationship vicarious contact–prejudice toward Chinese individuals. It was hypothesized that:
Method
Participants
After excluding participants that failed the attention check (n = 11) and the suspicion check (n = 5), the final study sample, collected in 2022 during the COVID-19 pandemic, consisted of 161 Australian participants, including 112 community members and 49 undergraduate students, all aged over 18 years. 1 One hundred participants identified as women. One hundred and five participants identified as having Anglo-Celtic background, 54 identified as having European background, and two identified as Hispanic. One hundred and one participants identified as not religious, 52 identified as Christian, three as Buddhist, and five as having other religions. Finally, 105 participants supported left-wing political parties, 28 supported right-wing parties, and 28 supported other parties or preferred not to respond.
An a priori G*Power analysis and a Monte Carlo power analysis for the mediation effect were conducted (Faul et al., 2007; Schoemann et al., 2017) using a medium effect (f = .25), as indicated by a recent vicarious contact meta-analysis (Banas et al., 2020). The analysis showed that for a two-way ANOVA with four groups, 128 participants were needed to detect a medium effect size, and 151 participants were needed to detect the predicted indirect effects, allowing a power of 0.80 at α = .05. Given the final sample included 161 participants, sufficient statistical power was obtained.
Overall, in the intergroup conditions (n = 64), 33 participants were in the audio condition and 31 were in the text condition. In the intragroup conditions (n = 97), 53 were in the audio condition and 44 in the text condition.
Design
Participants were randomly allocated to one condition of a 2 (contact: intergroup vs. intragroup) x 2 (mode of delivery: audio vs. text) between-subjects experimental design.
Materials and measures
Audio versus text vicarious contact manipulation
Two interaction scripts were developed for the intergroup and intragroup conditions. Consistent with the relatively neutral discussion themes (e.g., work–life balance in Australia) used in previous E-contact materials (Boccanfuso et al., 2021; Maunder et al., 2019), the scripts depicted two characters discussing strategies for sustainable living as solutions to tackle climate change in a talk show setting (see supplemental material). Apart from the absence of racial identity disclosure in the intragroup condition, the intergroup and intragroup interactions were identical. Moreover, in the intergroup condition, the Chinese character’s ethnic background was revealed after the initial self-introduction phase. Here, the names Wuchen and John were used for the Chinese and ingroup member, respectively. In the intragroup condition, the characters’ White identity was signaled using prototypical European Australian male names, Tom and Lucas, and the fact that they were both born in Australia.
The scripts also incorporated all four of Allport’s (1954) optimal conditions, which have been neglected in previous vicarious contact research despite their ability to enhance the positive effects of contact (Paluck et al., 2019; Pettigrew & Tropp, 2006). These conditions were operationalized by having a “talk show host” moderate the interaction (authority support), establishing both characters as Australian residents (equal status), and having the characters collaboratively devise a sustainable living strategy together (cooperation and common goal).
Intergroup anxiety
Using Stephan and Stephan’s (1985) six-item Intergroup Anxiety Scale, participants rated how, for example, “happy” and “defensive” they would be if they were the only non-Chinese person in a group of Chinese people, on a 5-point Likert scale (1 = strongly disagree, 5 = strongly agree). After reverse-scoring relevant items, higher scores indicated higher levels of intergroup anxiety (Cronbach’s α = .85).
Outgroup empathy
The same scale used in Study 1 was used to measure outgroup empathy (Cronbach’s α = .67).
Prejudice
White and Abu-Rayya’s (2012) Cultural Issues Scale (CIS) required participants to rate the seriousness of 12 events involving Chinese individuals, such as “Having several Chinese families move into your neighbourhood,” across a 7-point Likert scale (1 = not at all serious, 7 = extremely serious). After reverse-scoring relevant items, higher scores indicated higher levels of prejudice (Cronbach’s α = .90).
Negative intergroup emotions
Binder et al.’s (2009) adapted six-item measure assessed participants’ level of negative emotions toward Chinese individuals. Participants rated the extent to which they agreed with statements describing general feelings (e.g., “irritated”) toward Chinese people on a 7-point Likert scale (1 = not at all, 7 = extremely). After reverse-scoring relevant items, higher values indicated higher levels of negative intergroup emotions (Cronbach’s α = .82).
Desire for social distance
Link et al.’s (1987) adapted six-item Desire for Social Distance Scale assessed participants’ willingness to engage with Chinese individuals (e.g., “Having a Chinese person as a neighbor”) on a 4-point Likert scale (1 = definitely willing, 4 = definitely unwilling). Higher scores indicated a greater desire for social distance (Cronbach’s α = .90).
Manipulation checks
Allport’s optimal conditions check
Four items checked whether Allport’s optimal contact conditions (Allport’s, 1954) were successfully implemented. Participants rated the extent to which the interaction was positive and cooperative, supported by an authority figure, and involving individuals of equal status and a common goal across a 5-point Likert scale (1 = strongly disagree, 5 = strongly agree). Higher scores indicated greater effective operationalization in the contact scripts (Cronbach’s α = .80).
Intergroup and intragroup identification checks
Four items assessed whether participants identified with the characters in the contact interactions on the following aspects: Australian identity, European Australian versus Chinese ethnicity, and the issue of climate change. Participants scored their level of agreement with statements such as “I strongly identify with being Australian,” on a 5-point Likert scale (1 = strongly disagree, 5 = strongly agree). Higher scores indicated greater identification. For the intergroup check: Cronbach’s α = .69; for the intragroup check: Cronbach’s α = .72.
Experimental check
One item assessed if participants correctly recognized the ethnicity of the individuals in their contact conditions. If participants could identify the individuals’ ethnicities, they were asked to specify it.
Suspicion check
Three items asked participants whether they perceived the talk show to be useful in promoting discussions about climate change, to comment on their general experience during the talk show, and whether they noticed anything unusual during the “two studies.”
Attention check
Two items identified inattentive participants. The first item asked them to select “strongly disagree” from a 7-point Likert scale (1 = strongly disagree, 7 = strongly agree). The second item asked participants to select the month after April (i.e., May).
Demographic questionnaire
Eight items were used to collect information about participants’ gender, age, residency status, ethnicity, religion, education, political orientation, and disability status.
Procedure
This research was conducted online using Qualtrics. Participants first read the consent form and were given information about two “unrelated” studies to disguise the true intent of Study 2. After random allocation, participants in the audio condition listened to a recording of the stimulus material. The voice actors in the recording all had an Australian accent except for the Chinese character in the intergroup condition, who distinctively had a Chinese accent. Participants in the text condition read a transcript of their assigned contact interaction. The audio and the text conditions lasted, on average, 5 and 3 minutes, respectively. After contact, participants in the intergroup contact condition answered two discussion questions intended to reinforce the effects of contact, whereas participants in the intragroup condition answered two filler discussion questions (see supplemental material). All participants then completed manipulation checks and proceeded to the “second study.” As in Study 1, participants completed a fake random selection task and only answered outcome measures involving Chinese individuals in which the order of all outcome measures was randomized. Two attention check items were randomly placed amongst the measures, and two suspicion checks assessed participants’ awareness of the study’s true purpose. Finally, participants were all debriefed in regard to the true aims of the study.
Results
A one-sample t test with 3 as the midpoint and test statistic showed that participants rated all items in the Allport’s conditions check significantly higher than the scale’s midpoint (see Table 2), overall indicating a successful operationalization of Allport’s conditions.
Means, standard deviations, and t-test results of ratings of Allport’s conditions check.
Note. Scores ranged from 1 ( strongly disagree ) to 5 ( strongly agree ), with 3 as the midpoint and test statistic. Higher scores indicate greater agreement with Allport’s conditions.
Additionally, independent samples t tests showed that participants in the intergroup condition (M = 4.59, SD = 0.58) rated the interaction as significantly more cooperative than participants in the intragroup condition did (M = 4.34, SD = 0.76), t(159) = 2.26, p = .025, but did not differ in their ratings of all other contact conditions. No significant differences in item ratings were found between the audio and text conditions.
In vicarious contact scenarios, identification with the ingroup member (i.e., ingroup identification) is an essential precondition for the effectiveness of the strategy, as it is important for the participants to identify with the ingroup character so that the positive effect of an intergroup contact is also relatable to them (Banas et al., 2020; Ortiz & Harwood, 2007; Vezzali et al., 2015). A one-sample t test found that intergroup participants did not significantly identify with the Chinese character (M = 3.09, SD = 0.97), t(160) = 0.77, p = .443, though they identified significantly with being Australian, the European Australian character, and the issue of climate change (all ps < .001). Similarly, the intragroup condition participants identified with the Australian identity, the Caucasian characters Tom and Lucas, and the issue of climate change significantly more than the midpoint of 3 (all ps < .050).
Table 3 shows the means, standard deviations, and Pearson’s correlations for the dependent variables. Specifically, intergroup anxiety and outgroup empathy significantly correlated with all three outcome measures in the expected directions. Consistent with prior evidence (Pettigrew & Tropp, 2008), intergroup anxiety was also negatively correlated with outgroup empathy.
Means, standard deviations, and Pearson’s correlations of mediator and outcome variables.
Note. Higher scores indicate greater intergroup anxiety, outgroup empathy, prejudice, negative outgroup emotions, and more desire for social distance. Hyphens represent variables’ correlations with themselves. NIE = negative intergroup emotions; DSD = desire for social distance.
**p < .01 (two-tailed).
To test the main hypotheses, a series of two-way between-subjects ANOVAs tested the effects of the experimental manipulation on all dependent measures. Refer to Table 4 for all relevant statistics. In partial support of H1, participants in the intergroup vicarious contact condition reported lower levels of prejudice toward Chinese people (M = 19.58, SD = 8.85) than those in the intragroup vicarious contact condition (M = 23.49, SD = 10.59), and significantly lower negative intergroup emotions (M = 13.61, SD = 5.07) than those the intragroup contact condition (M = 15.46, SD = 5.79). There was no significant main effect of contact on the remaining dependent measures. In partial support of H2, participants in the text-based vicarious contact condition reported significantly lower intergroup anxiety (M = 14.75, SD = 4.61) than participants in the audio-based condition did (M = 16.27, SD = 5.01), and significantly lower negative intergroup emotions (M = 13.53, SD = 5.61) than participants in the audio-based condition did (M = 15.77, SD = 5.36). There was no significant main effect of mode of delivery on reported outgroup empathy, prejudice, and desire for social distance. There was no support for H3, which predicted an interaction effect between contact and mode of delivery on any of the dependent measures.
F values, p values, and effect sizes for contact and mode of delivery on dependent variables.
Note . The interaction between the two independent variables is indicated by IV1 × IV2. NIE = negative intergroup emotions; DSD = desire for social distance. Significant results (p < .05) are boldfaced.
The moderated mediation model predicted in H4 was analyzed using Model 8 of Hayes’s (2018) PROCESS macro. Whilst intergroup anxiety was found to be positively associated with all three outcome variables: prejudice, b = 0.55, SEb = 0.16, 95% CI [0.24, 0.85]; negative intergroup emotions, b = 0.36, SEb = 0.07, 95% CI [0.22, 0.50]; and desire for social distance, b = 0.11, SEb = 0.04, 95% CI [0.03, 0.19]; and outgroup empathy was negatively associated with prejudice, b = −0.52, SEb = 0.15, 95% CI [−0.82, −0.23]; negative intergroup emotions, b = −0.35, SEb = 0.07, 95% CI [−0.48, −0.21]; and desire for social distance, b = −0.11, SEb = 0.04, 95% CI [−0.18, −0.03], the analyses revealed that there were no significant direct effects of contact on any of the dependent variables (ps > .050). Therefore, there was no support for H4.
Discussion
The finding that vicarious contact reduced prejudice and negative intergroup emotions is consistent with previous evidence (Castelli et al., 2012; Liebkind & McAlister, 1999; Pahlke et al., 2012). However, contrasting with previous findings and Study 1 (Koc & Anderson, 2018; E. Nguyen et al., 2012; Pagotto & Voci, 2013), intergroup vicarious contact did not reduce anxiety or desire for social distance. One potential explanation for the null finding on social distance is that the current sample reflected a substantial ceiling effect, due to the fact that Chinese people make up the largest racial minority group in Australia (Australian Bureau of Statistics, 2022), and thus that the opportunity for prior contact, measured in Study 1, was relatively high. This social context limits the ability for manipulated vicarious contact to promote additional positive changes on social distance measures (Hodson, 2011; White, Borinca, et al., 2020; White, Verrelli, et al., 2019). Alternatively, there may have been an insufficient level of stereotypic information in the vicarious contact condition, which is necessary for reducing anxiety about the outgroup (Maunder et al., 2019; Pettigrew, 1998; Walch et al., 2012). In the COVID-19 context, Chinese individuals are stereotyped as dirty and disease carriers (Roberto et al., 2020; Xu et al., 2021). However, the current vicarious contact scripts did not focus on this topic, and instead centered on the less stereotypic topic of sustainable living. Future vicarious contact interventions may be more effective by directly challenging specific stereotypic information about Chinese people.
Despite only partial support for our second hypothesis, Study 2 yielded the first experimental evidence demonstrating the benefits of text-based (over audio-based) vicarious contact in reducing intergroup anxiety and negative intergroup emotions. These results are supported by previous findings (Acheme, 2022) that accented speech indicating outgroup membership induces higher intergroup anxiety than accented speech indicating ingroup membership. Further, these results align with research showing that hearing different accents is associated with a broad range of negative affective responses (e.g., annoyance, frustration; Spencer-Rodgers & McGovern, 2002). These results also indirectly support Bilali and Vollhardt’s (2013) finding that an audio-based vicarious contact intervention was effective in improving outgroup attitudes when the outgroup membership of the characters could not be identified via their accent. Overall, this finding poses an interesting theoretical question for future research on what features of text- versus audio-based contact (e.g., accents) can elicit greater anxiety and/or outgroup bias (Stephan, 2014).
Nevertheless, it is also worth noting that a methodological feature of the current text-based vicarious contact condition may have impacted the results. Most past research investigating text-based vicarious contact entailed narration of ad hoc created texts by an extenal and neutral voice, usually the researcher, 2 (e.g., Cameron & Rutland, 2006) due to the young age of the children involved. Participants in the present research, however, read the interaction scripts by themselves. Previous research has associated reading with inner speech (Filik & Barber, 2011), the psychological experience of covertly articulating words in one’s mental space (Alderson-Day & Fernyhough, 2015). As this has been closely associated with the vocal features of one’s ingroup (e.g., regional accents, age; Filik & Barber, 2011), participants may have interpreted the intergroup interaction as an intragroup experience. Compared to participants in the audio-delivered condition who experienced either intergroup or intragroup contact, such “absence” of intergroup contact in the text-delivered condition may also explain our current findings.
Finally, there was no significant difference between the two delivery mode conditions across the remaining outcome variables (i.e., prejudice, outgroup empathy, and desire for social distance), and no significant interaction effect between contact and mode of delivery. One possible explanation for these findings is the lack of adequate self-disclosure by the outgroup character. Previous theoretical (Brewer & Miller, 1984) and experimental research (Davies et al., 2011; Ensari & Miller, 2005) has suggested that self-disclosure during contact is necessary for promoting positive outgroup attitudes. This is because self-disclosure (i.e., information about oneself) facilitates contact on an interpersonal basis, thereby personalizing the interaction and placing emphasis on the outgroup member’s individual (rather than group) features. Contrasting the extended self-disclosure manipulations in previous studies (Ensari & Miller, 2002), the outgroup character here only briefly disclosed personal information through sharing hobbies, and this limited self-disclosure may have contributed to our null findings (see Bagci et al., 2021).
General Discussion
The results of Study 1 and Study 2 supported that both direct and indirect contact are viable strategies to improve outgroup attitudes toward Chinese people, especially during a global crisis (e.g., COVID-19 pandemic). While our results were obtained during a specific time frame, these findings can be generalized and contribute to the extant contact literature by demonstrating that within the context of a global (health) crisis, with greater intergroup hostility, the effectiveness of contact on prejudice reduction is not hindered (Adler et al., 2022; Van Assche et al., 2023). However, it should also be noted that positive contact did not yield an effect on realistic threat, suggesting that more enduring (i.e., longitudinal) strategies are still required to alleviate perceived threat from outgroups in challenging contexts.
Furthermore, Study 1 found that the association between direct contact and prejudice toward Chinese people is mediated by intergroup anxiety and partially mediated by outgroup empathy and interpersonal closeness, and that intergroup anxiety is the strongest mediator of the contact–prejudice relationship across the three dimensions (i.e., affective, cognitive, behavioral intentions) of prejudice. Stephan and Stephan’s (1985) contention that intergroup anxiety is most pervasive when the outgroup is perceived as a threat is particularly relevant to improving outgroup bias in crises.
Contrary to predictions, intergroup anxiety and outgroup empathy did not mediate the vicarious contact effects in Study 2. Nevertheless, in partial support of our hypothesis, Study 2 demonstrated that text-based vicarious contact resulted in more positive emotions and less intergroup anxiety in comparison to audio-based vicarious contact, supporting Study 1 findings that intergroup anxiety remains a key variable when improving attitudes toward Chinese people. Study 2 results provide preliminary evidence that audio-based vicarious contact, with its salient outgroup features (e.g., accents), can be negatively evaluated by ingroup members, and thus be less conducive to prejudice reduction.
Limitations and Future Directions
Long-term outcomes in reducing prejudice toward Chinese people were neglected in both studies. Moreover, the vicarious contact conditions tested in Study 2 were significantly shorter than previous vicarious contact involving 1-year-long or 6-week-long programs (Bilali & Vollhardt, 2013; Cameron et al., 2011; Paluck, 2009). These limitations are worth noting since intergroup contact over a longer period has been found to be more effective in improving attitudes (Dovidio et al., 2017; White & Abu-Rayya, 2012). Future research will benefit from employing a longitudinal design to explore the effects of audio- and text-based vicarious contact over time.
Although our research has demonstrated that a health crisis is not a limiting factor to the positive effects of intergroup contact toward the targeted outgroup, other historical and political crises might still reduce the effectiveness of contact in prejudice reduction (Ditlmann & Samii, 2016; Van Assche et al., 2023). Concurrently, even though our research measured realistic threat as an amalgamation of material and physical well-being (e.g., quality of life), we did not directly measure COVID-19 health threat as a standalone variable. Thus, whether health threats from a global crisis could have moderated the mediating effects of intergroup anxiety on the contact–prejudice relationship will need to be tested in future research. Moreover, by building on existing studies, future research could also consider examining the relative effectiveness of contact interventions during different crises (e.g., global warming, political).
Future research may also consider investigating the effects of contact on both minority and majority groups. While intergroup dynamics inherently involve bidirectional interactions, the contact literature has primarily taken a unidirectional approach prioritizing the attitudes of the majority (Shelton, 2000; White, Harvey, & Verrelli, 2015). This subsequently limits the extent to which intergroup harmony can be achieved (McGrane & White, 2007; White, Harvey, & Verrelli, 2015). Thus, given today’s landscape, it may be relevant for contact researchers to also consider the attitudes of Chinese people toward European Australians. Concurrently, features of audio-based contact, such as the accent of the majority outgroup, may not be as negatively salient in the contact situation for minority racial groups. Hence, future research should test whether the various features of audio-based contact compared to text-based contact will have a different impact on the minority’s attitudes toward the majority racial group.
Finally, although our audio-based vicarious contact yielded unfavorable results compared to its text-based counterpart, various researchers have demonstrated the benefits of audio-based contact in other contexts (Bilali & Vollhardt, 2013; Bilali et al., 2016; Paluck, 2009, 2010). More research is needed to delineate the conditions and exact features under which audio-based and text-based contact may bring about positive outgroup attitude change and, by extension, reduce anxiety in ingroup members. While our audio-based vicarious contact decidedly characterized the outgroup member with a nonstandard English accent, there are other group memberships that do not necessarily have a distinctive accent (e.g., sexuality, disability). Another future consideration is to compare vicarious contact to other indirect contact strategies (e.g., E-contact) to test their effectiveness in reducing prejudice and, in doing so, compare text- and audio-based versions of different indirect contact strategies (White, Maunder, & Verrelli, 2020).
Implications and Conclusions
Considering the current anti-Chinese sentiments exasperated by the COVID-19 pandemic, Studies 1 and 2 provided support for the utility of both direct and vicarious contact strategies in reducing prejudice toward Chinese people during a global crisis. This research suggests intergroup anxiety as an important variable to consider when devising future interventions aimed at reducing prejudice toward Chinese people, and provides a preliminary framework for exploring the conditions required to implement a successful vicarious contact intervention via different modes of delivery. Further, Study 2 suggests vicarious contact has the potential to provide real-world solutions to counteract widespread anti-Chinese sentiment incited via Western mass media channels (Borja et al., 2020; Wen et al., 2020), where such vicarious contact involves cooperation and collaboration (Gómez & Huici, 2008). The prospect of depicting a positive and cooperative intergroup interaction between European Australians and Chinese individuals is highly scalable through mass media. Thus, it is crucial that researchers build on the current research’s insights and continue to explore factors and external boundary conditions in which the positive effects of either direct or vicarious contact can be more readily realized. Ultimately, the practical and translational qualities of the current findings may assist policymakers and social psychology scholars in combining efforts to achieve a more harmonious society.
Supplemental Material
sj-docx-1-gpi-10.1177_13684302241231201 – Supplemental material for The power of words and voices: The role of direct and vicarious contact (text vs. audio modalities) in reducing anti-Chinese prejudice
Supplemental material, sj-docx-1-gpi-10.1177_13684302241231201 for The power of words and voices: The role of direct and vicarious contact (text vs. audio modalities) in reducing anti-Chinese prejudice by Roberta Chen, Fiona A. White, Michael Deng and Ellen Shi in Group Processes & Intergroup Relations
Footnotes
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
