Abstract
The aim of the present research was to investigate the psychological processes driving outgroup helping intentions among ethnic groups actually involved in natural disasters. We proposed that disaster exposure would be associated with different degrees of outgroup helping intentions depending on victim’s ethnic group of belonging. Specifically, based on an integration between the common ingroup identity model (Gaertner & Dovidio, 2000) and the integrated threat theory (Stephan & Stephan, 2000), we hypothesized that perceived disaster exposure would be negatively associated with ethnic majorities’ altruistic intentions via increased outgroup threat, and it would be positively associated with ethnic minorities’ altruistic intentions via enhanced one-group perception. The findings of our research conducted with Italian natives (i.e., majority group) and immigrants (i.e., minority group) struck by the 2012 Northern Italian earthquakes fully supported our predictions. Implications of these findings for developing “altruistic” communities in postdisaster contexts are discussed together with the importance of considering the combined contribution of different psychological theories.
Helping behaviors play a critical role in mitigating the adverse material and psychological consequences of dramatic events such as natural disasters. Accordingly, over the last decades research has devoted considerable attention to psychological processes involved in helping toward disaster victims. However, to date most empirical work has focused on altruistic behaviors enacted by individuals or social groups not actually involved in the natural disaster (e.g., Zagefka, Noor, Brown, Hophtrow, & de Moura, 2012), whereas considerably less literature has taken into account the mutual helping behaviors among members of the struck community (e.g., Bolin, 1982; Kaniasty & Norris, 1995, 2000). Moreover, this latter research has focused on communities that were homogenous along ethnic lines (or has neglected interethnic differences within communities), while there is a substantial lack of data on multiethnic contexts (Norris, Friedman, Watson, Byrne, & Kaniasty, 2002). We believe this is a clear gap in the literature. Indeed, when multiethnic settings are involved, the psychological processes fostering mutual helping may be particularly complex and differ depending on one’s ethnic group of belonging.
With the aim of extending the literature in this field, we explored the role of perceived exposure to a natural disaster as an antecedent of outgroup helping intentions among Italian (majority group) and immigrant (minority group) victims of the earthquakes that struck Northern Italy in 2012. Moreover, we investigated the processes driving the effects of perceived disaster exposure on outgroup helping, by proposing an integration between the common ingroup identity model (CIIM; Gaertner & Dovidio, 2000, 2012) and the integrated threat theory (ITT; Stephan & Stephan, 2000; Stephan, Ybarra, & Rios Morrison, 2009).
The Negative (and Positive) Consequences of Natural Disasters
A large number of clinical studies (for a review, see Raphael & Maguire, 2009) has documented the psychological consequences that disaster exposure has on individuals, such as posttraumatic stress disorders (e.g., Basoglu, Livanou, Salcioglu, & Kalender, 2003; Frankenberg, Nobles, & Sumantri, 2012; Norris, Murphy, Baker, & Perilla, 2004), depression (e.g., North et al., 1999; Shore, Tatum, & Vollmer, 1986), and increased anxiety (e.g., Bystritsky, Vapnik, Maidment, Pynoos, & Steinberg, 2000; Kar & Bastia, 2006; Norris, Perilla, Riad, Kaniasty, & Lavizzo, 1999). However, most psychological literature in this field has mainly focused on individual consequences of being exposed to natural disasters and has neglected the effects of such events at a broader level and, in particular, on social relations (Bonanno, Brewin, Kaniasty, & La Greca, 2010). Similarly to humanly caused events (e.g., wars, acts of terrorism), the consequences of natural disasters go beyond the individual level and deeply influence the struck community as a whole (Kaniasty & Norris, 2004). In particular, as suggested by different scholars (e.g., Bolin, 1989; Eranen & Liebkind, 1993), disaster exposure may tangentially have “therapeutic” effects (C. E. Fritz, 1961) on the community, by enhancing cohesion within its members and intensifying mutual altruistic behavior. Emerging perspectives in psychosocial research also hint at the possibility that these adverse life events may increase victims’ willingness to help others, including outgroup members. For instance, Staub (2003, 2005) introduced the term “altruism born of suffering” to indicate that individuals’ sufferings may enhance motivation to help others—not despite their traumatic experiences, but precisely because of them.
However, we argue that in multiethnic communities specific group processes may affect such enhanced mutual helping, especially when ethnic majority and minority groups coexist within the same victimized context.
Intergroup Helping and Natural Disasters
A growing body of research has emphasized the key role of intergroup biases in shaping people’s intentions to help outgroup members (e.g., Nadler, 2002; van Leeuwen & Täuber, 2011). First, there is evidence that people are more prone to help ingroup than outgroup members (e.g., Dovidio, Piliavin, Schroeder, & Penner, 2006; Levine, Prosser, Evans, & Reicher, 2005; Yzerbyt, Dumont, Wigboldus, & Gordijn, 2003). However, this tendency is not always present and is stimulated by different processes, which also depend on the intergroup setting considered. For example, when interethnic settings are taken into account, bias in intergroup helping is especially frequent when the context affords individuals more opportunities to justify withholding help with nonracist explanations (see Saucier, Miller, & Doucet, 2005, for a meta-analysis). This could be the case for the sample considered in the present study, where we argue that the threat posed by the low-status group “justifies” the high-status members’ intentions to not help outgroup members, depending on their degree of exposure to the natural disaster. Moreover, the motivations behind the willingness to help ingroup or outgroup members are different. While empathy is an important factor that motivates the intention to help ingroup members (Siem & Stürmer, 2012; Stürmer, Snyder, Kropp, & Siem, 2006; Stürmer, Snyder, & Omoto, 2005), feelings of interpersonal attraction can motivate individuals to help outgroup members (Stürmer et al., 2005). However, as is the case of our study, these feelings of attraction may be less relevant when individuals express their intentions to help unknown outgroup members, who are nonetheless victims as they are of the disaster.
Another line of research (Nadler, 2002; Nadler & Halabi, 2006) has further highlighted the complexity involved in intergroup helping, especially when asymmetric status relations are involved. For example, when the group advantage is perceived as illegitimate, high-status groups may use intergroup helping as a method of maintaining social inequality, by providing low-status groups dependency-oriented forms of help. Extending these assumptions, we argue that in emergency situations the high-status group may even deny helping to low-status group members based on realistic threat considerations (see following lines).
Within the context of natural disasters, there are now various studies documenting the pervasive influence of different types of prejudice on the desire to help disaster victims. Sun, Zagefka, and Goodwin (2013) have shown that endorsing a stereotypical view of Japan predicted Chinese respondents’ emotions toward Japanese and, in turn, their concern for the victims of the 2011 Japanese earthquake. Cuddy, Rock, and Norton’s (2007) results revealed that the subtle denial of uniquely human emotions to outgroup members (i.e., infrahumanization; see Leyens, Demoulin, Vaes, Gaunt, & Paladino, 2007) reduced the intentions to help the Hurricane Katrina outgroup victims. Extending this work, Andrighetto, Baldissarri, Lattanzio, Loughnan, and Volpato (2014) reported that the denial of uniquely human attributes (e.g., rationality; see Haslam & Loughnan, 2014) to Haitians decreased Italians’ willingness to help victims of the 2010 Haitian earthquake, whereas the denial of human nature attributes (e.g., warmth) to Japanese people decreased Italians’ willingness to help victims of the 2011 Japanese earthquake.
Together, the reviewed studies suggest that diverse intergroup biases play a crucial role in determining altruistic behaviors following natural disasters. These biases may clearly contrast the “altruism born of suffering” process. However, as mentioned before, most of these studies explored helping in bystander groups. We aimed at investigating group processes affecting helping among victim groups actually involved in the disaster.
Our aim was to provide first evidence that exposure to natural disasters may trigger different psychological processes and have different consequences on outgroup helping, depending on one’s ethnic group of belonging. Next we provide the rationale for our hypotheses.
Common Ingroup Identity Model and Integrated Threat Theory
Vollhardt (2009) theorized that the “altruism born of suffering” (Staub, 2003, 2005) could be explained by the fact that disaster exposure may create the perception of a common fate among individuals who suffered. Extending this assumption to the realm of intergroup relations, we argue that the traumatic experience of being hit by a natural disaster will create the perception that ingroup and outgroup members belong to the same superordinate group, that is, the group of disaster victims. This prediction is consistent with the basic tenets of the CIIM (Gaertner & Dovidio, 2000, 2012), stating that perceptions of belonging to a common ingroup have positive effects on intergroup relations. A vast amount of experimental (e.g., Gaertner, Mann, Murrell, & Dovidio, 1989), longitudinal (e.g., Levin, Sinclair, Sidanius, & van Laar, 2009), and cross-sectional (e.g., Capozza, Trifiletti, Vezzali, & Favara, 2013) studies has consistently shown that emphasizing the salience of a common identity including both ingroup and outgroup has a wide range of beneficial effects on intergroup relations. Some research also provided initial evidence that the development of a common ingroup identity enhances helping and altruistic behaviors toward outgroup members (Dovidio et al., 1997; Levine & Thompson, 2004; Nier, Gaertner, Dovidio, Banker, & Ward, 2001; Vezzali, Stathi, et al., in press, Study 1), also within emergency contexts (Vezzali, Cadamuro, Versari, Giovannini, & Trifiletti, 2015). For instance, Levine and Thompson’s results (2004) revealed that enhancing the salience of the superordinate identity of EU citizens increased British citizens’ willingness to help European national groups hit by natural disasters.
According to the CIIM (Gaertner & Dovidio, 2000, 2012), different factors promote the development of a common ingroup identity. Of particular interest here, a recent work (Dovidio et al., 2004, Study 1) has revealed that a common ingroup identity (i.e., Americans) elicited by the threat of a humanly caused disaster (i.e., a terrorist attack) directed at all Americans decreased White Americans’ prejudice toward African Americans. Similarly, Davies, Steele, and Markus (2008) reported that U.S. citizens primed with 9/11 terrorist attacks displayed heightened national identity and subsequent greater support for inclusion of ethnic groups associated with American common national identity.
Building from these studies, we aimed at extending literature on the CIIM (Gaertner & Dovidio, 2000, 2012) by empirically exploring the link between perceived natural disaster exposure and common ingroup perception within a multiethnic community. We hypothesized that, among minority group victims (i.e., immigrants), the degree of perceived disaster exposure—in terms of material and psychological consequences—would be positively related with minority victims’ perception of belonging to a common ingroup that includes both ethnic groups. In turn, such increased one-group perception was expected to be positively associated with altruistic behavioral intentions toward outgroup (majority) victims.
Instead, we hypothesized that perceived disaster exposure would have opposite effects for majority group members (i.e., Italian natives), leading to a decrease in their willingness to help outgroup members through higher perceptions of outgroup threat.
According to ITT (Stephan & Stephan, 2000; Stephan et al., 2009), different types of threat may foster negative outgroup attitudes, thus undermining the intentions to help threatening outgroups: realistic threat, symbolic threat, intergroup anxiety, and outgroup stereotypes. We believe that realistic threat is particularly important within natural disaster contexts. Indeed, perceived disaster exposure is generally associated with a loss of social, economic, and environmental resources (Hobfoll, 2001), that could lead individuals to perceive outgroup victims as a threat for obtaining the remaining resources. The idea of realistic threat is also rooted in realistic group conflict theory (LeVine & Campbell, 1972; Sherif, 1966), which posits that competition for limited resources increases negative attitudes and hostility between groups (e.g., Bobo, 1988; Quillian, 1995). Thus, this threat becomes relevant when ingroup members perceive the outgroup as a potential competitor for resources, threatening the ingroup’s physical or material well-being (e.g., jobs, health, or territory). Despite a great deal of research has shown that intergroup threat negatively affects outgroup attitudes (for a meta-analytic review, see Riek, Mania, & Gaertner, 2006), some characteristics of the threatening outgroup seem to moderate these effects. In particular, intergroup threat is a significant predictor of more negative outgroup attitudes when it is felt by majorities toward minorities. This effect is likely due to the majorities’ perception that minority members usurp their material and economical resources (Riek et al., 2006). For example, if immigrants (minority group) are seen as threatening job and material resources of natives (majority group), this may lead to negative attitudes or behavioral intentions toward minority group members. Supporting this assumption, McLaren (2003) found in a cross-European research that realistic threat held by European natives toward immigrants was positively related to a preference for their expulsion from the host country.
In the aftermath of a natural disaster, majority group members may perceive ethnic minority members as especially threatening. In such situation, material resources allocated by institutions to the victims may not be exactly defined, making the victims uncertain on the extent to which they will have access to these resources. This should be especially a problem for the majority, who may feel that immigrants are exploiting resources that ought to be primarily assigned to the hosting group. For these reasons, we hypothesized that the degree of perceived disaster exposure would be associated with increased majority members’ materialistic threat toward outgroup minority victims, because they would be perceived as potential competitors of the postdisaster resources allocated by institutions. In turn, such increased threat should be negatively associated with willingness to help outgroup victims.
The Present Research
The aim of our study was to examine the processes driving outgroup helping intentions within ethnically heterogeneous communities struck by a natural disaster. The study took place immediately after the 2012 Emilia Romagna (a region of Northern Italy) earthquakes. These two earthquakes have been routinely referred to as one of the most disastrous natural events that hit Italy in the last decades. Twenty-seven people died, 300 were injured, and approximately 19,000 families were displaced for a long time. Affected were 59 towns, and 14,000 buildings were seriously damaged (“Dossier of Emilia Romagna,” 2013). Since Emilia Romagna is one of the Italian regions with the largest percentage of immigrants (“Demography in Numbers,” 2013), it follows that the earthquakes struck a heterogeneous community composed by both Italian natives and immigrant residents. Somehow complicating this picture, relations between Italian natives and immigrants in this context are problematic, and Italian natives in particular have been shown to hold negative attitudes toward immigrants at both explicit and implicit levels (Giovannini & Vezzali, 2012; Vezzali & Giovannini, 2011).
Unlike most psychosocial research in this field, we decided to focus on actual victims, rather than on bystanders, because fostering reciprocal helping should be a primary aim for communities trying to recover from disastrous events (Kaniasty & Norris, 2004). Based on the literature reviewed, we hypothesized that the degree to which participants perceived having been exposed to the earthquake would be related with willingness to help outgroup victims, but that this relation would vary as a function of one’s group of belonging. Specifically, greater perceived disaster exposure should be associated among Italian victims with increased threat that immigrants exploit material aids for the consequences of the disaster, resulting in reduced willingness to help outgroup (minority) victims. On the other hand, immigrants have more to gain from the allocation of material resources than Italian natives; thus, for immigrants, the creation of a superordinate group including majority members would be a beneficial strategy, because belonging to a common ingroup would allow them to have access to the postearthquake material aids. Thus, for this group, a greater degree of perceived disaster exposure should be associated with stronger perception of belonging to the common group of victims, resulting in increased willingness to help outgroup (majority) victims.
Specifically, we expected a pattern of moderated mediation (Preacher, Rucker, & Hayes, 2007), such that, for Italians, the path from perceived disaster exposure to decreased helping intentions should be mediated by higher outgroup threat perceptions; for immigrants, the path from perceived disaster exposure to increased helping intentions should be mediated by higher common ingroup identity perceptions.
Method
Participants and Procedure
Participants were 162 earthquake victims, of which 70 were Italian natives (39 males, 31 females) and 92 were immigrants (50 males, 41 females, 1 missing value) living in Emilia Romagna towns struck by the earthquake. The distinction between Italian natives and immigrants was based on participants’ nationality, also taking into account whether participants had foreign origins, that is, only participants with Italian nationality and an Italian background were considered as Italians. Participants ranged in age from 15 to 88 years (M = 37.60; SD = 14.22). At the time of the earthquake, most participants lived in San Felice sul Panaro (30.9%), Cavezzo (27.8%), and Mirandola (25.9%). Immigrants came from Asia (50.0%), South America (19.6%), Africa (16.2%), and Eastern Europe (14.2%). During data collection, all participants were temporarily living in the tent cities. 1 A sample of 162 yields sufficient power (.80) to detect small-sized effects on standardized regression coefficients (b* = 0.22) in multiple regressions with three predictors (see Aberson, 2010) and for the analyses in which bias-corrected bootstrapped estimates for detecting mediated effects were calculated (Fritz & MacKinnon, 2007).
Approximately one month after the earthquakes of May 2012, two researchers administered individually to each participant a questionnaire, presented as a research on the consequences of the earthquake. Before fulfilling the scales described next, participants were asked some demographics, including age, sex, and country of origin. At the conclusion of the study, participants were thanked and fully debriefed.
The Questionnaire
Responses to all items were given on a 4-step scale, ranging from 1 (not at all) to 4 (very much).
Perceived disaster exposure
Participants rated perceived disaster exposure with six items (e.g., “The eartquake has damaged my house and/or my other properties”; “I am suffering psychological traumas following the earthquake”) adapted from previous works (e.g., Andrighetto, Mari, Volpato, & Behluli, 2012; Myers, Hewstone, & Cairns, 2009). Scores were combined to yield an overall perceived disaster exposure score (α = .71); higher scores denote greater perceived disaster exposure.
Common ingroup identity
In line with research on the CIIM (Gaertner & Dovidio, 2000, 2012), common ingroup perception was assessed with a single-item measure: “To what degree do Italian and immigrant victims of the earthquake belong to a single group?”
Outgroup threat
Perceived threat toward outgroup members was measured with three items based on previous literature (see Riek et al., 2006): “In the tent cities, immigrant [Italian] victims are taking up areas that should be assigned to Italian natives [immigrants]”; “In the tent cities, immigrant [Italian] victims are consuming meals that should be primarily given to Italian natives [immigrants]”; “Financial aids to Italian natives [immigrants] will be reduced, because part of them will also be assigned to immigrants [Italian natives].” Scores were combined across the three items to yield a composite index (α = .65); higher scores indicate greater perceived outgroup threat.
Outgroup helping intentions
The willingness to help outgroup members was assessed with the following three items: “If I had the chance, I would not hesitate to help immigrant [Italian] victims of the earthquake”; “I believe that we should give a helping hand to immigrant [Italian] victims of the earthquake”; “Regardless of my economic position, I would be glad to donate some money to immigrant [Italian] victims of the earthquake.” The three items were collapsed into a single index (α = .93); higher scores denote greater willingness to help outgroup members.
Results
Introductory Analyses
Table 1 presents means and standard deviations for each composite variable by group of belonging. As shown in Table 1, immigrants exhibited more positive attitudes than Italian natives. In fact, despite perceived disaster exposure did not differ between groups, immigrants perceived the two groups as belonging to a common ingroup more than Italian natives, t(160) = 3.69, p < .001, d = 0.59. Further, compared with Italian natives, immigrants perceived outgroup members as less threatening, t(160) = 6.10, p < .001, d = 0.95, and displayed more willingness to help outgroup victims, t(160) = 12.56, p < .001, d = 1.94.
Means and standard deviations (in parentheses) for each composite variable by group of belonging.
Note. The response scale for all items ranged from 1 (not at all) to 4 (very much). Values with different subscripts on the same row are significantly different at p < .001.
Correlations among variables for Italian natives and immigrants are reported in Table 2. With regard to Italian natives, perceived disaster exposure was positively correlated with outgroup threat, whereas it was unrelated with common ingroup perception. Further, perceived disaster exposure and outgroup threat were negatively correlated with outgroup helping intentions. With regard to immigrants, perceived disaster exposure was positively correlated with common ingroup perception and outgroup helping intentions; perceived disaster exposure and outgroup threat were not significantly associated. Further, common ingroup perception was positively correlated with outgroup helping intentions.
Correlations between composite variables.
Note. Correlations for Italian natives are presented below the diagonal; correlations for immigrants are presented above the diagonal.
p < .05. **p < .01.***p < .001.
Moderated Mediation Analysis
To examine our predictions, we conducted a moderated mediation analysis by using Hayes’s (2012) PROCESS macro (Model 7). The conceptual framework of the proposed model is shown in Figure 1. In this model, perceived disaster exposure (centered) was entered as the independent variable, participants’ ethnic group (1 = Italian vs. −1 = immigrants) as the moderator, outgroup threat and common ingroup perception as the mediators and helping intentions as the dependent variable.

Theoretical moderated mediation model.
For the moderated path from perceived disaster exposure to outgroup threat (see Table 3), results showed that both perceived disaster exposure and participants’ ethnic group were positively related with outgroup threat. These main effects were qualified by the expected two-way interaction Perceived Disaster Exposure × Participants’ Ethnic Group. As expected, decomposition of this interaction (see Figure 2) revealed that perceived disaster exposure was positively related with outgroup threat for Italian victims, b = 0.50, SE = 0.15, p = .001, whereas such relation was not significant among immigrant victims, b = 0.03, SE = 0.26, p = .792.
Regressions of perceived disaster exposure on helping intentions when outgroup threat and common ingroup identity are the mediators and participants’ ethnic group is the moderator.
Note. b = unstandardized beta weight; SE = standard error; sr2= squared semipartial correlation.
p < .07. *p < .05. **p < .01. ***p < .001.

Interactive effects of perceived disaster exposure and participants’ ethnic group on outgroup threat.
For the moderated path from perceived disaster exposure to common ingroup identity (see Table 3), results revealed a significant association between participants’ ethnic group and common ingroup perception; the association between perceived disaster exposure and common ingroup perception was not fully significant, b = 0.22, SE = 0.13, p = .097. Crucially, the interaction Perceived Disaster Exposure × Participants’ Ethnic Group (marginally) qualified these main effects. Consistent with our predictions, the decomposition of this interaction (see Figure 3) showed that, among immigrants, perceived disaster exposure was related with increased perceptions that Italian and immigrant victims belong to the same ingroup, b = 0.44, SE = 0.17, p = .01. Instead, this relations was not significant for Italians, b = −0.06, SE = 0.22, p = .777.

Interactive effects of perceived disaster exposure and participants’ ethnic group on common ingroup identity.
For the paths from the mediator variables to helping intentions (see Table 3), we found that outgroup threat was related with decreased outgroup helping intentions, whereas there was a positive association between common ingroup identity and outgroup helping intentions.
Most importantly, in line with predictions, the conditional indirect effect of perceived disaster exposure on reduced outgroup helping intentions through higher outgroup threat was significant for Italian respondents, a * b = −0.20, 95% CI [−0.038, −0.08], but not for immigrant respondents, a * b = −0.01, 95% CI [−0.10, 0.08]. In other words, among Italian natives—but not among immigrants—the path from perceived disaster exposure to decreased outgroup helping intentions was mediated by increased feelings of outgroup threat. Instead, the conditional indirect effect of perceived disaster exposure on improved outgroup helping intentions through higher common ingroup identity was significant for immigrants, a * b = 0.15, 95% CI [0.04, 0.32], but not for Italian natives a * b = −0.02, 95% CI [−0.19, 0.15]. That is, among immigrants—but not among Italian natives—the path from perceived disaster exposure to increased outgroup helping intentions was mediated by increased common ingroup perception.
Alternative Models
To provide further support for our hypotheses, we ruled out a series of alternative models. In the first, in line with previous research (see Riek, Mania, Gaertner, McDonald, & Lamoreaux, 2010), we examined the role of common ingroup identity as inhibitor of outgroup threat. Accordingly, it is plausible to imagine that (at least for immigrants) perceived disaster exposure would be related with helping intentions through an increase of common ingroup identity and a consequent decrease of outgroup threat. Based on this theoretical assumption, we tested a double mediation model (Hayes, 2012, Model 6) in which the cross-product term between perceived disaster exposure and participants’ ethnic group was the predictor variable, common ingroup identity the first-level mediator, outgroup threat the second-level mediator, and outgroup helping intentions the outcome variable. However, the indirect effect of the cross-product term on outgroup helping intentions via common ingroup identity (first-level mediator) and outgroup threat (second-level mediator) was not significant, a * b = 0.08, 95% CI [−0.0004, 0.04].
We also ruled out an alternative moderated mediation model focusing on the potential moderating role of common ingroup identity. Indeed, it is plausible to expect that, for Italian victims, the pattern we found would only emerge among those who perceive a common ingroup identity to a lower extent. 2 Thus, we carried out a moderated mediation analysis (see Model 11, Hayes, 2012) in which both common ingroup identity and participants’ ethnic group were entered as moderators, perceived disaster exposure was the predictor, outgroup threat the mediator, and outgroup helping intentions the outcome variable. However, results of this model showed that neither the two-way interaction Common Ingroup Identity × Perceived Disaster Exposure, b = −0.09, SE = 0.08, p = .291, nor the three-way interaction Common Ingroup Identity × Perceived Disaster Exposure × Participants’ Ethnic Group, b = −0.05, SE = 0.08, p = .557, were significantly related with outgroup threat, thus failing to support the moderator role of common ingroup identity.
General Discussion
Whereas prior investigations have typically focused on antecedents of postdisaster helping within bystander groups (e.g., Zagefka, Noor, Brown, & de Moura, & Hopthrow, 2011), the present research extends this literature by exploring group processes driving mutual helping intentions in groups actually involved in the disaster. The results of our research conducted with Italian natives and immigrants struck by the 2012 Northern Italian earthquakes suggest that intergroup consequences of perceived disaster exposure radically vary as a function of the ethnic group of belonging. With regard to immigrant respondents (i.e., the ethnic minority), higher levels of perceived disaster exposure were positively related with prosocial intentions toward Italian victims through increased common ingroup perception. These effects are in line with previous investigations (for a review, see Vollhardt, 2009) showing that prosocial behaviors are not only rooted in positive experiences (see Staub, 2005), but may also rise following collective experiences of sufferings, whether they are caused by human intentionality (e.g., terrorist attacks; see Penner, Dovidio, Piliavin, & Schroeder, 2005) or not (e.g., natural disasters; Kaniasty & Norris, 1995).
These findings are also consistent with the CIIM (Gaertner & Dovidio, 2000, 2012; see also Dovidio et al., 1997): the salience of victimization due to the natural disaster, acting as a threat external to both ingroup and outgroup, enhanced among immigrants the perception of a shared group membership among victims, giving rise to increased motivations to help outgroup victims. Different reasons could underlie this enhanced immigrants’ common ingroup perception. On the one hand, being included in the superordinate victim group together with the majority group would improve immigrants’ status position (Tajfel & Turner, 1979), thus facilitating their access to postdisaster resources. On the other hand, this increased sense of “we-ness” may significantly help the process of resilience after the trauma, which may be particularly difficult due to the disadvantaged status of their group, generally implying more scarce social connections and social support compared with majority group members.
In contrast with findings obtained for immigrants, the impact of the disaster was associated with negative intergroup implications for Italian natives (i.e., the ethnic majority group). Specifically, perceived disaster exposure was positively related with Italians’ feelings of materialistic threat toward immigrant victims, who were presumably seen as competitors for postdisasters resources allocated by institutions. Coherently with the intergroup threat literature (Riek et al., 2006), such increased feelings of threat were negatively related with altruistic intentions toward immigrant victims. Thus, for Italian respondents, outgroup threat seemed to have a predominant role which nullified the positive effects of common ingroup perception and inhibited the willingness to help outgroup victims.
Although our findings seem to fully confirm our predictions, it is noteworthy that we were interested in examining the impact of exposure to dramatic events on prosocial intentions by focusing on the underlying processes consistent with ITT (Stephan & Stephan, 2000) and CIIM (Gaertner & Dovidio, 2000). However, we do not exclude that other psychological mechanisms concur with common ingroup identity and outgroup threat to explain helping responses toward the outgroup. For instance, we did not examine the role of mortality concerns that a natural disaster may raise. There is evidence that mortality salience increases helping intentions, but only when the beneficiaries are ingroup members (see Jonas, Schimel, Greenberg, & Pyszczynski, 2002). Thus, it is possible that (at least part of) the effect of perceived disaster exposure on outgroup helping intentions would be explained by increased morality concerns. Specifically, the exposure to the natural disaster may increase mortality salience which, in turn, could increase helping intentions toward ingroup members, at the same time decreasing intentions to help outgroup members.
We believe our work has several theoretical novelties. First, in contrast with most of previous research, it takes into account the role of group processes in the aftermath of a natural disaster, by focusing on majority and minority victims actually involved in the disaster. Our findings highlight the importance of considering the perspective of both majority and minority members within ethnically heterogeneous contexts. In light of our findings, previous investigations in which victims were considered as belonging to a homogeneous group along ethnic lines are likely to have obscured important differences. Second, this study reconciles the two important perspectives provided by CIIM (Gaertner & Dovidio, 2000, 2012) and ITT (Stephan & Stephan, 2000; Stephan et al., 2009). In particular, results showed that both theoretical accounts can help understanding the patterns guiding intergroup help following a natural disaster, depending on the ethnic group considered.
We believe the present results have noteworthy practical implications. Fostering reciprocal support is critical for helping victims to recover after a devastating natural disaster, regardless of the specific group of belonging. This may be especially complicated within ethnically heterogeneous contexts. In particular, the majority group should be targeted with reassurances about the fact that minority group members will not spoil postdisaster resources. Reducing this threat may favor the creation of a superordinate identity including both ingroup and outgroup, also among majority group members. Thus, an appropriate and quick decision-making by institutions on how and when postdisaster resources will be allocated may play an important role in reducing uncertainty and consequent increased hostility toward outgroup victims.
Limitations and Future Directions
There are some limitations that should be considered in interpreting our findings and in determining the direction of future research. First, although the causal link between our variables is consistent with CIIM (Gaertner & Dovidio, 2000, 2012) and ITT literature (Stephan & Stephan, 2000; Stephan et al., 2009), the correlational nature of our data does not allow to draw causal inferences. Second, concerning measures, we used 4-point scales for all items, which did not include a midpoint, and we sometimes used ad hoc created scales in order to account for the specific event under investigation; however, it would be important that future studies will replicate our findings by including a midpoint in the response scales and by using validated measures of the constructs that were considered.
Third, although our sample of victims was moderately large and included substantial proportions of ethnic minorities, immigrants were considered as a single group despite their heterogeneous origin. We believe this is not a problem in the context of the present study, as immigrants were asked to answer questions concerning the well-defined Italian outgroup, rather than their heterogeneous ingroup. Nonetheless, future studies should consider more homogeneous immigrant groups and test whether effects differ depending on the specific minority group taken into consideration. Fourth, the present results have been obtained with an adult sample, and may not be automatically generalized to other age groups or postdisaster contexts. For instance, majority children may be unaware of the realistic threat posed by immigrants and react to the disaster by thinking of themselves and the outgroup as a single group. This is precisely what Vezzali, Cadamuro, et al. (2015) found by examining a sample of elementary school children in the aftermath of the 2012 Northern Italian earthquakes.
On a theoretical level, results may also differ depending on the specific postdisaster context taken into consideration. For instance, majority members may react to the disaster with increased one-group perceptions and intentions to help the outgroup whenever they are certain to receive help from institutions. Also, minority members may react with increased outgroup threat and reduced intentions to help the outgroup if they know that resources will be allocated primarily (or almost exclusively) to the majority group. Future research should explore these possibilities in order to provide a clearer picture of how to foster outgroup helping after a natural disaster.
Finally, one may wonder how it is possible that victims of the disaster have the possibility to help other victims, thus raising concerns of our choice of measuring (outgroup) helping among actual victims. However, it should be noted that the disaster literature shows that mutual social support and cooperation are often present in the aftermath of disasters (e.g., Drury, Cocking, & Reicher, 2009; Oliver-Smith, 1999). Strengthening community ties and increasing mutual helping among victims after a disaster is crucial to return to “normal life,” also when considered in strategic terms, with the aim of reducing the costs of not helping (Piliavin, Dovidio, Gaertner, & Clark, 1981).
Conclusion
We believe that our findings shed light for the first time on psychological antecedents of mutual helping intentions when the multiethnic nature of victimized communities is considered. In particular, our work indicates the complexity of developing altruistic behaviors when ethnic majorities and minorities coexist within the same victimized context. If the “therapeutic features” (C. E. Fritz, 1961) of disasters seem to clearly emerge in ethnic minorities, this does not seem to occur for ethnic majorities, who feel threatened by ethnic minorities because of the scarcity of resources. Thus, we argue that the effects of ethnicity cannot be neglected if the aim is to provide a full analysis of postdisaster dynamics within a victimized ethnically heterogeneous community.
Footnotes
Funding
This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.
