This is an open access article distributed under the terms of the Creative Commons
Attribution License (In remarks made in November, 2017, UN High Commissioner for Refugees Filippo Grandi criticized “wealthy states” for “closing their borders, restricting access to asylum and deterring entry” as the number of displaced refugees approached 66 million (UN, 2017). While there are certainly many examples worldwide of compassionate and charitable responses to the current refugee crisis, the reaction in the United States has gone from ambivalent to hostile.
In January of 2017, President Trump signed an executive order barring persons from certain Muslim majority countries from entering the United States temporarily and preventing Syrian refugees from entering the United States indefinitely. This order was later revised after the initial order was blocked by a federal court. Nationwide protests occurred condemning the order, frequently referred to as the “Muslim ban” or “travel ban.” However, support for refugees continues to decline among Americans. A survey conducted by Pew Research in March of 2018 (Hartig, 2018) found that 51% of Americans believe the United States has a responsibility to accept refugees, while 43% say America does not bear this responsibility. This is a slight change from February 2017 (Hartig, 2018), when 56% said it is America’s responsibility to accept refugees and 41% said it was not. This shift in support for refugee varies along party lines. In March of 2018, only 26% of Republicans stated American has a responsibility to accept refugees, down from 35% in the February 2017 survey. Belief in America’s responsibility to refugees increased in Democrats during this same time frame, from 71% in February of 2017 to 74% in March of 2018 (Hartig, 2018).
Similar negative attitudes towards refugees are common among Europeans. A survey of people from 10 European countries found that the majority believed that refugees increased the likelihood of terrorist attacks and posed an economic threat to host countries and that these fears of refugees were driven to some extent by anti-Muslim attitudes and prevalent beliefs that Muslims would not assimilate into host cultures (Wike, Stokes, & Simmons, 2016). The ideological divides on this issue seen in the U.S. are mirrored in European nations, with supporters of far-right political parties in France and the United Kingdom being far more likely to endorse anti-immigrant statements like “refugees are a burden because they take our jobs” and “refugees will increase the likelihood of terrorism” (Wike et al., 2016).
There are likely several political and psychological factors that influence individuals’ reactions to the refugee crisis. However, one seemingly important factor might be the extent to which people feel a sense of similarity or commonality to the refugees.
According to Social Identity Theory (Tajfel & Turner, 1979), there is a fundamental human tendency to divide the world into the group or groups to which one belongs and to which one does not belong. This tendency towards group-based categorizations is not just cognitive but emotional, as people are strongly motivated to enhance their self-esteem by protecting a positive perception of their in-group. This motive manifests itself in a tendency to maximize perceived differences between in-group and out-group members and to preferentially help in-group members over out-group members even when group membership is based on minimal or trivial domains (Tajfel, Billig, Bundy, & Flament, 1971).
Although people have the capacity to feel distress about the suffering of others, these feelings may be limited to in-group members and people may in fact be indifferent or even feel schadenfreude (pleasure derived at the misfortune of another) in response to the suffering of out-group members (Cikara, Bruneau, & Saxe, 2011). Consequently, it seems reasonable that people who define refugees as out-group members based on perceived religious, nationality, or ethnic differences would be less compassionate to their plight and less interested in providing aid, particularly aid in the form of offering them asylum.
According to Self-Categorization Theory (Turner, Hogg, Oakes, Reicher, & Wetherell, 1987), people can define their identity at different levels from the individual to various levels of inclusiveness within groups. The lowest, most specific level is to define one’s identity as a unique individual. At the intermediate level people categorize themselves based on group membership and focus on the characteristics that distinguish their in-group from out-group members. For example, a college professor could identify as a member of his or her department or at the more inclusive level as a member of the university as a whole. The highest and most inclusive level of self-categorization would seem to be as a member of humanity. While most group identities are divisive in that they distinguish between in-group and out-group, the exception is to define one’s identity based on membership in the category human being. Feeling a sense of shared human identity would seem to be antithetical to ethnocentrism, privileging the needs of some groups over others, or prejudicial treatment and differential helping based on group membership (McFarland, Webb, & Brown, 2012).
Applying Self-Categorization Theory to reactions to refugees, categorization at the intermediate level would be expected to reduce interest in helping because it fosters focus on group-based-identity that would likely reduce perceived commonality and therefore empathy. In contrast, categorization at the highest level of shared human identity would be expected to increase interest in helping because it fosters perceived commonality, empathy, and concern.
Similarly, the common in-group identity model (Gaertner & Dovidio, 2000) suggests that when subgroup members are re-conceptualized as common-group members based on their membership in a superordinate category, then reactions to members of this group should become more favorable. From this perspective, the tendency to think of refugees as common-group members (fellow human beings) rather than sub-group members (people of different ethnicity, nationality, or religion) should increase positive sentiments towards refugees and consequentially increase support for offering them aid and asylum (Nickerson & Louis, 2008).
Another reason to expect that thinking about the shared humanity of refugees would result in more favorable attitudes and a greater likelihood of offering aid comes from research on dehumanization. Dehumanization involves thinking about groups of people in animalistic or infrahuman ways as a means of justifying mistreatment, marginalization, or violence against them on the rationale that they do not warrant the same moral rights and considerations as those granted fully human status. American participants blatantly dehumanized Muslims and Arabs by rating these groups as significantly less evolved than Americans and this blatant dehumanization was associated with less support for allowing Arab immigrants into the U.S. (Kteily, Bruneau, Waytz, & Cotterill, 2015). Among American participants, blatant dehumanization of Muslims was associated with more anti-Muslim attitudes, more support for policies restricting Muslims from gaining access to the U.S., and increased willingness to sign a petition asking congress to refuse to accept any refugees from predominantly Muslim countries (Kteily & Bruneau, 2017).
Among people in potential host nations, individual differences in the stable tendency to feel a shared sense of humanity with all people should predict increased willingness to help refugees because it facilitates a focus on shared social identity and counters the tendency towards dehumanization. In contrast, the tendency among people in potential host nations to identify more strongly with their community and their country should be associated with decreased willingness to help refugees because it facilitates a focus on a non-shared social identity.
Identification With All Humanity
The notion of shared humanity is one that evolved slowly over time (McFarland, 2011). Western philosophical questions about humanity arose in the 16th century in response to the exploration of the new world. The initial ethnocentric assumptions about the infrahuman status of indigenous peoples that justified their enslavement slowly over several centuries gave rise to a sense of shared humanity and human rights. The notion of shared humanity was further shaped by cultural debates over slavery, women’s liberation, and the civil rights movement. The extension of humanity and human rights was also a reaction to horror in response to the atrocities of war, ethnic cleansing, and the holocaust. This emergent construct, which McFarland labels Identification with all Humanity (IWAH), involves feeling a shared common identity with people of ethnic, cultural, and religious variability, and promotes extending empathy and concern for others beyond those in one’s in-group.
McFarland and colleagues (2012) developed a scale to measure individual differences in IWAH. Although the human capacity to feel concern about the welfare of out-group members may be increasing from a historical vantage point, it still seems that people feel a stronger identification with their in-group than with all humanity. A summary of the research employing the IWAH scale, suggested that in college student and adult samples IWAH is moderate for most respondents (McFarland, Brown, & Webb 2013). On the scale from 1 (not at all) to 5 (very much) the average level of IWAH is roughly 3 (somewhat). Further, 80% of people fell between 2 (just a little) and 4 (quite a bit). Less than 15% of people identified as strongly with all humanity as they did with the comparison groups of Americans and people in their community. On average, American respondents score about half a point on the scale higher for identifying with Americans and people in their community than for identifying with all humanity.
IWAH has been demonstrated to be predictive of several humanitarian attitudes and behaviors (McFarland et al., 2012). For example, IWAH was associated with placing higher value on global human rights relative to American national interest (Study 1), greater knowledge of global humanitarian challenges and existing efforts to face these challenges (Study 8), and with greater reported willingness to give more money to charity and help international victims of natural disaster (Study 10).
Goals and Hypotheses of the Current Research
The goal of the present research was to examine the extent to which individual differences in IWAH were predictive of support for providing aid and asylum to refugees. We expected IWAH to be associated with more support for refugees because people scoring high on IWAH are likely to view refugees as fellow human beings in need of assistance and this sense of shared identity should be translated into greater compassion and a greater perceived moral obligation to help. In contrast, people low on IWAH are more likely to privilege the concerns of their own group (potential economic and national security threat) over the needs of refugees who they perceive as members of an outgroup and potentially as less deserving of full moral considerations. We expected identification with all humanity to predict more support for refugees over and beyond the effects of identification with one’s community and America. Moreover, we expected these effects to hold even when taking into account more established predictors of intergroup attitudes such as social dominance orientation (SDO), right-wing authoritarianism (RWA), political conservatism, and religiosity.
Given that concerns about national security and terrorism might be sources of opposition to offering asylum to refugees, especially those from predominantly Muslim countries, we also examined how individual differences in IWAH would be related to support for extreme measures in the war on terrorism that were likely to increase the risk of civilian casualties. American tolerance of civilian war casualties tends to be high. In a Gallup (2011) survey of people in 131 countries conducted between 2008 and 2010, Americans were the most tolerant of military attacks targeting civilians, with 49% of Americans indicating that it is sometimes justified for the military to target and kill civilians. These attitudes may be a product of dehumanization. Among American participants, blatant dehumanization of Arabs and Muslims was associated with increased support for extreme counter-terrorism measures including torture, targeting civilians, and bombing entire countries that might be harboring terrorists (Kteily et al., 2015, Study 3).
If tolerance of civilian causalities is increased by a tendency to dehumanize these casualties, then feeling a shared sense of identity and focusing on the humanness of at-risk civilians should be associated with less tolerance of civilian casualties. Consequently, we hypothesized that IWAH would be associated with less support for extreme counter-terrorism measures likely to result in civilian casualties. The topics of support for refugees and tolerating civilian casualties in combating terrorism were studied together because they share some conceptual similarities. Being motivated to offer assistance to refugees seeking asylum and being concerned about avoiding civilian casualties both seem to require a sense of concern for the suffering of those affected as well as a sense of moral responsibility or obligation to help other people. The people affected in both cases are primarily from Middle Eastern countries where Islam is the predominant religion. Consequently, ethnic, religious, and nationality differences could lead to both reduced concern and lowered perceived moral obligation because of a tendency to categorize them as out-group members (Nickerson & Louis, 2008). Inversely, IWAH should be associated with thinking about the shared humanity with refugees and potential civilian casualties rather than focusing on group differences. This focus on shared humanity should in turn translate into a greater sense of concern with alleviating or preventing suffering and a stronger perceived moral obligation to offer help.
In addition to IWAH, we also examined how attitudes about refugees and extreme counter-terrorism methods were related to social dominance orientation (SDO), right-wing authoritarianism (RWA), political conservatism, and religiosity. McFarland (2011) conceptualized ethnocentrism as the opposite of IWAH and described ethnocentrism as being influenced by authoritarianism (a rigid adherence to social and cultural traditions and resistance and hostility towards progressives who challenge these traditional values) and social dominance orientation (support for social inequality and status hierarchies particularly in the form of an in-group’s aim to protect their superiority and domination over out-group members afforded by the status quo). Consistent with this conceptualization, IWAH was negatively correlated with both SDO and RWA (McFarland et al., 2012). In addition, SDO and RWA are consistently strong predictors of generalized prejudice against multiple outgroups (McFarland, 2010) and are associated with less commitment to human rights (McFarland & Mathews, 2005). Further, research has shown that SDO and RWA were associated with less support for asylum seekers (Lyall & Thorsteinsson, 2007) and that SDO and RWA were associated with more support for military action in the Middle East as part of the war on terrorism (Crowson, 2009).
Political conservatism is associated with less support for human rights (McFarland & Mathews, 2005) and more negative attitudes towards immigrants (Leon McDaniel, Nooruddin, & Faith Shortle, 2011). Further, political conservatives report stronger concerns about terrorism and less support for allowing Syrian refugees into the U.S. (Cox & Jones, 2015). Political conservatism is also associated with less IWAH (McFarland et al., 2012).
Religion could be conceptualized as a factor promoting support of refugees given the ethical focus of many religious teachings on helping the vulnerable and needy. However, research seems to support an alternative view that religion is a barrier to helping, because refugees are perceived as a cultural threat due to their different religious and ethnic out-group status. Leon McDaniel et al. (2011) found that religious Americans had more negative attitudes towards immigrants than did non-religious Americans. Further, among American Catholic university students, activating their religious social identity, by priming them to think about religious social interactions, increased support for anti-immigrant policies but only in reference to immigrants who were dissimilar in terms of ethnicity and religion (Bloom, Arikan, & Courtemanche, 2015). This tendency for religion to promote out-group bias is also evidenced in the fact that more frequent religious attendance has been shown to predict greater hostility towards religious out-groups (Ginges, Hansen, & Norenzayan, 2009).
Given these findings, in the present study, we expected support for refugees to be associated with less SDO, RWA, political conservatism, and religiosity. We expected support for extreme counter-terrorism measures to be associated with more SDO, RWA, political conservatism, and religiosity. We also expected that IWAH would predict support for refugees and lack of support for extreme counter-terrorism measures over and beyond the influence of SDO, RWA, political conservatism, and religion. In Study 1, these predictions were tested in an online survey.
Study 1
The goal of Study 1 was to examine the ability of IWAH to predict greater support for refugees and less support for extreme counter-terrorism measures. We also aimed to demonstrate the predictive ability of IWAH even when controlling for other relevant variables, including identification with community, identification with America, SDO, RWA, political orientation, and religiosity.
Method
Participants
Participants were 202 people who completed an online survey between October 27 and November 18, 2015. Participants were recruited through email using a university student list serve as well as through social media by posting links to the study on the authors' Facebook pages. Of the participants, 146 were women, 54 were men, and 2 did not indicate their gender. Of the participants, 164 indicated their racial/ethnic background as White, 25 as Black, 6 as Hispanic, 1 as Filipino, and 6 as other or unspecified. The majority of participants were from South Carolina (79.2%). Participants ranged in age from 18-71 years (M = 26.62, SD = 12.40).
Materials
Support for extreme counter-terrorism measures
Participants rated their agreement, on a 5-point scale anchored at 1 (strongly disagree) and 5 (strongly agree), with four statements about support for the use of pre-emptive and extreme force in the war on terrorism. A representative item was “If we could capture or kill leaders of terrorist organizations that threaten the United States we should do it, even if thousands of civilians are injured or killed in the process.” These items were used in previous research by Pyszczynski et al. (2006).
Support for refugees
Participants read the brief passage below that was created for the purpose of the present study and described the refugee crisis in Europe and the Middle East.
There is currently a refugee crisis in Europe and the Middle East. Several million people have been displaced from their homes in countries like Syria and Afghanistan. The refugees are being displaced by years of brutal civil war and from fears of attacks from radical groups such as the Islamic State (ISIS). Many of these distressed and desperate people are seeking asylum in neighboring countries such as Jordan, Turkey, and Lebanon, while many more hope to make their way into wealthier countries further away such as Germany. Because the number of refugees far outstrips the ability of neighboring countries to offer assistance, many are living in limbo without basic needs and services.
Participants rated their agreement on a scale from 1 (disagree strongly) to 5 (agree strongly) with five statements about the extent to which they thought the U.S. should offer assistance and asylum to these refugees. Representative items were “The U.S. has an obligation to provide financial resources and humanitarian aid to help these refugees” and “The U.S. should accept more of these refugees into our country.” Factor analysis revealed one factor with an eigenvalue of 3.36 that accounted for 66.7% of the variance. All five items had strong loadings on this factor ranging from .78-.89. All other factors had eigenvalues of less than one. Consequently, the average rating of the five items was treated as a single measure of support for refugees.
Identification with all humanity
Participants then completed the Identification with all Humanity Scale (McFarland et al., 2012). On this scale, participants made rating on a 5-point scale in response to nine questions. They answered each question in reference to people in their community, Americans, and people all over the world. A representative item was “How much do you identify with (that is, feel a part of, feel love toward, have concern for) each of the following (people in my community, Americans, all humans everywhere)?” Responses were made on a scale from 1 (not at all) to 5 (very much). Responses to the nine items were then averaged to yield three separate scores showing the extent to which people identified with their community, America, and all of humanity respectively.
Social dominance orientation
Participants responded to 13 items from the Social Dominance Orientation Scale (Pratto, Sidanius, Stallworth, & Malle, 1994).i Each item presented an idea, and participants rated how they felt about the idea on a scale from 1 (very negative) to 7 (very positive). Seven of the statements were worded such that more positive feelings toward these items reflected acceptance of group-based inequality. Six of the items were worded such that more positive feelings toward these items reflected resistance to group-based inequality. These later six items were reverse scored before averaging them with responses to the first seven items. Higher resulting scores reflected more social dominance orientation.
Right wing authoritarianism
Participants completed a 20-item version of the Right-Wing Authoritarianism Scale (Altemeyer, 2007). On this scale, participants rated their agreement with 20 statements using a 9-point scale anchored at -4 (strongly disagree), 0 (neutral), and 4 (strongly agree). Ten of the items were worded such that agreement reflects more RWA. The other 10 items were worded such that agreement reflects less RWA. The later 10 items were reverse scored prior to summing the responses. Higher resulting scores on this measure indicate more RWA.
Religion
Participants were asked to rate the importance of religion in their lives on a scale from 1 (not at all important) to 7 (extremely important). Participants were asked to indicate how frequently they attended religious services on a scale from 1 (never) to 7 (more than twice a week). The average religious attendance was 3.19 (SD = 1.93) and the average rated religious importance was 4.33 (SD = 2.27). The variables were strongly correlated (r = .75) so they were transformed into z scores before being averaged to form a composite measure of religiosity.
Political orientation
Participants responded to the question “How would you characterize your political orientation?” using a scale anchored at 1 (liberal), 4 (moderate), and 7 (conservative).
Procedure
First, participants provided demographic information including age, race, and gender, and answered the questions about political orientation and religion. Next, they completed the index of support for extreme counter-terrorism measures and then the measure of support for refugees. Then, participants completed the IWAH, SDO, and RWA.
Results
Differences in Levels of Identification Across IWAH Subscales
Scores on the three IWAH subscales (community, America, all humanity) were subjected to a within-subjects analysis of variance, which revealed an overall difference in levels of identification, F(2, 402) = 22.9, p < .001, η2p = .141. Pairwise comparisons were conducted using paired sample t-tests. Participants identified more strongly with America (M = 3.64, SD = 0.70) than with all humanity (M = 3.34, SD = 0.75), t(201) = -4.98, p < .001, d = .35. Participants identified more strongly with their community (M = 3.69, SD = 0.76) than with all humanity, t(201) = -5.63, p < .001, d = .40. Participants’ identification with their community was not significantly different from their identification with America, t(201) = 1.36, p = .175, d = .10.ii
Simple Order Correlations
Simple order correlations among all variables are presented in Table 1. Support for extreme counter-terrorism measures was associated with less IWAH, younger age, more identification with community, more identification with America, more political conservatism, more religiosity, higher RWA, and higher SDO. Support for refugees was associated with more IWAH, less political conservatism, less religiosity, less identification with America, less identification with community, lower RWA, and lower SDO.
Table 1
Simple Order Correlations Among All Variables in Study 1
| Variable | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 |
|---|---|---|---|---|---|---|---|---|---|---|
| 1. Age | -- | -.11 | -.09 | -.02 | -.18* | .02 | .10 | -.07 | -.19** | .07 |
| 2. PC | -- | -- | .48** | .37** | .74** | .28** | .25** | -.24** | .49** | -.47** |
| 3. REL | -- | -- | -- | .21** | .67** | .31** | .27** | .00 | .39** | -.26** |
| 4. SDO | -- | -- | -- | -- | .40** | .03 | .03 | -.32** | .37** | -.41** |
| 5. RWA | -- | -- | -- | -- | -- | .35** | .30** | -.20** | .60** | -.51** |
| 6. IC | -- | -- | -- | -- | -- | -- | .66** | .33** | .27** | -.17* |
| 7. IA | -- | -- | -- | -- | -- | -- | -- | .36** | .33** | -.18* |
| 8. IWAH | -- | -- | -- | -- | -- | -- | -- | -- | -.21* | .39** |
| 9. SFECT | -- | -- | -- | -- | -- | -- | -- | -- | -- | -.49** |
| 10. SFR | -- | -- | -- | -- | -- | -- | -- | -- | -- | -- |
| α | -- | -- | -- | .87 | .87 | .89 | .87 | .87 | .83 | .87 |
| M | 26.62 | 3.88 | 0.00 | 2.38 | 74.81 | 3.69 | 3.64 | 3.35 | 2.30 | 3.15 |
| SD | 12.40 | 1.72 | 0.93 | 1.05 | 33.92 | 0.76 | 0.70 | 0.75 | 1.02 | 1.03 |
Note. PC = political conservatism; REL = religiosity; SDO = social dominance orientation; RWA = right wing authoritarianism; IC = identification with community; IA = identification with America; IWAH = identification with all humanity; SFECT = support for extreme counter-terrorism measures; SFR = support for refugees.
*p < .05. **p < .01.
Partial Correlations
The partial correlation of IWAH and support for extreme counter-terrorism measures was still statistically significant even when controlling for identification with America and identification with community, r(198) = -.39, p < .001. The partial correlation of IWAH and support for refugees was still statistically significant even when controlling for identification with America and identification with community, r(198) = .51, p < .001.
Regression Analyses
The support for extreme counter-terrorism measures scores and support for refugees scores were subjected to separate hierarchical multiple regression analyses. The grand mean centered continuous variables age, religiosity, political conservatism, RWA, and SDO were entered in Step 1. Identification with community and identification with America were entered in Step 2, and IWAH was entered in Step 3. As can be seen in Table 2, RWA, SDO, and Identification with America were all associated with more support for extreme counter-terrorism measures.
Table 2
Predictors of Support for Extreme Counter-Terrorism Measures and Support for Refugees in Study 1
| Predictor Variable | Step 1
|
Step 2
|
Step 3
|
|||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| B | SE | 95% CI | β | B | SE | 95% CI | β | B | SE | 95% CI | β | |
| SFECT
|
||||||||||||
| REL | .021 | .083 | -.144 – .185 | .019 | -.006 | .062 | -.167 – .155 | -.006 | .019 | .081 | -.140 – .178 | .018 |
| Age | -.008 | .005 | -.017 – .002 | -.094 | -.010 | .005 | -.020 – -.001 | -.127* | -.013 | .005 | -.022 – -.004 | -.160** |
| PC | .051 | .050 | -.047 – .149 | .086 | .042 | .049 | -.054 – .137 | .070 | .016 | .048 | -.079 – .112 | .027 |
| RWA | .013 | .003 | .007 – .019 | .440** | .012 | .003 | .006 – .018 | .399** | .010 | .003 | .004 – .016 | .346** |
| SDO | .160 | .060 | .041 – .279 | .164** | .180 | .059 | .063 – .297 | .185** | .143 | .059 | .026 – .260 | .147* |
| IC | -.054 | .100 | -.251 – .144 | -.040 | .006 | .100 | -.192 – .203 | .004 | ||||
| IA | .335 | .108 | .122 – .548 | .229** | .441 | .112 | .220 – .661 | .301** | ||||
| IWAH | -.277 | .093 | -.463 – -.093 | -.202** | ||||||||
| r2 | .393 | .430 | .455 | |||||||||
| SFR
|
||||||||||||
| REL | .125 | .088 | -.048 – .299 | .113 | .132 | .089 | -.043 – .307 | .119 | .086 | .084 | -.079 – .251 | .077 |
| Age | -.001 | .005 | -.010 – .009 | -.008 | .000 | .005 | -.010 – .010 | .000 | .005 | .005 | -.005 – .051 | .060 |
| PC | -.101 | .052 | -.204 – -.003 | -.169 | -.099 | .053 | -.202 – .005 | -.165 | -.053 | .050 | -.152 – .047 | -.088 |
| RWA | -.011 | .003 | -.018 – -.005 | -.377** | -.011 | .003 | -.018 – -.005 | -.368** | -.008 | .003 | -.015 – -.002 | -.275** |
| SDO | -.221 | .063 | -.346 – -.096 | -.225** | -.226 | .064 | -.352 – -.099 | -.230** | -.159 | .062 | -.281 – -.037 | -.162* |
| IC | .021 | .109 | -.193 – .235 | .016 | -.085 | .104 | -.291 – .120 | -.063 | ||||
| IA | -.089 | .117 | -.320 – .142 | -.060 | -.279 | .116 | -.508 – -.049 | -.189* | ||||
| IWAH | .497 | .097 | .305 – .688 | .360** | ||||||||
| r2 | .338 | .340 | .420 | |||||||||
Note. SFECT = support for extreme counter-terrorism measures; SFR = support for refugees; REL = religiosity; PC = political conservatism; RWA = right wing authoritarianism; SDO = social dominance orientation; IC = identification with community; IA = identification with America; IWAH = identification with all humanity.
*p < .05. **p < .01.
IWAH was associated with significantly less support for extreme counter-terrorism measures even when controlling for the other variables in the model. As can be seen in Table 2, RWA and SDO were associated with significantly less support for refugees. IWAH was associated with significantly more support for refugees even when controlling for all the other variables in the model.iii
Discussion
The results of Study 1 offer further evidence for the validity and utility of the concept of IWAH as an individual difference that predicts attitudes towards salient political issues. The current study replicated previous findings that Americans generally tend to identify more strongly with their community and with other Americans than with all people everywhere (McFarland et al., 2013). The current study also replicated previous findings that RWA, SDO, and political conservatism were all associated with less IWAH (McFarland et al., 2012). This finding reinforces the argument that ethnocentrism is antithetical to IWAH.
The results of the present study revealed that greater IWAH was associated with more support for helping refugees and with less support for extreme counter-terrorism measures that were likely to result in civilian casualties. IWAH explained less variance in support for counter-terrorism measures (2.5%) than in support for refugees (8%) but accounted for unique variance in stances on both of these issues even when controlling for the influence of the more frequently studied dimensions of SDO, RWA, religiosity, and political conservatism. These findings are consistent with previous findings that IWAH predicts greater concern for international human rights and interest in helping international charities (McFarland et al., 2012). The current findings extend the link between IWAH and concern for international human welfare to two currently salient political issues: the humanitarian crisis caused by refugees fleeing conflict in the Middle East and tolerance for civilian casualties in military efforts to combat terrorism. The people involved in both of these issues are from predominantly Muslim countries. Americans tend to dehumanize Muslims (e.g., Kteily et al., 2015, Study 1) and this dehumanization is associated with wanting to reduce Muslims’ access to the U.S. (e.g., Kteily et al., 2015, Study 2) and with support for more draconian tactics in combating terrorism (e.g., Kteily et al., 2015, Study 3). IWAH involves feeling a shared sense of humanity with others regardless of their religious identity and may therefore have deterred participants in the present study from dehumanizing, which in turn may have made them more concerned about minimizing the suffering of refugees and limiting the loss of human life in counter-terrorism measures.
The results of Study 1 also revealed that both SDO and RWA were associated with less support for helping refugees and with more support for extreme counter-terrorism measures likely to result in civilian casualties. These findings are consistent with previous research showing that SDO and RWA were associated with more prejudice against perceived out-groups (McFarland, 2010), less commitment to international human rights (McFarland & Mathews, 2005) less support for asylum seekers in an Australian sample (Lyall & Thorsteinsson, 2007) and that, among U.S. participants, SDO and RWA were associated with more support for military action in the Middle East as part of the war on terrorism (Crowson, 2009). The present findings extend the ability of SDO and RWA to predict political attitudes to the domains of the current refugee crisis and tolerance of civilian casualties in military operations against terrorism. It seems that the assumption of the inevitability of group-based inequality characterized by high levels of SDO may lead people to be less concerned about the plight of refugees or potential civilian casualties as these events may be conceptualized as unavoidable side effects of a competitive world in which national interests conflict. Similarly, it is possible that the importance on preserving the legitimacy of traditional structures characterized by high levels of RWA may lead people to be less concerned about the plight of refugees or potential civilian casualties because the needs of these out-group members are minimized compared to the privileged goal of protecting the national interests of one’s own country.
Political conservatism and religiosity were also both correlated with more support for extreme counter-terrorism measures and less support for refugees. The correlations with political conservatism are not surprising given that American-politically-conservative platforms typically focus on a hawkish approach towards foreign policy and national defense and on tough anti-immigration policies. The correlations regarding religiosity might seem somewhat counter-intuitive given the prominence of religious ethical teachings focusing on compassion and helping the vulnerable and needy. The correlations with religiosity in the present study are; however, interpretable through the lens of Social Identity Theory (Tajfel & Turner, 1979). Greater religiosity would mean that being religious is a more salient aspect of identity and that people therefore would value their religious identity as more important in defining themselves and would be committed to enhancing their self-esteem by viewing their religious group as superior to others. The fact that the refugees in question and the potential innocent casualties in extreme counter-terrorism efforts are from predominantly Muslim countries would result in them being classified as out-group members thereby rendering them outside the scope of people about whom one must care and provide assistance. Although religiosity and political conservatism were correlated with both support for refugees and extreme counter-terrorism measures, they did not significantly predict either in the regression analyses. One possible explanation for this is that the effect of these variables was subsumed by the influence of RWA. Both political conservatism and religiosity were associated with more RWA and RWA was a strong predictor of less support for refugees and more support for extreme counter-terrorism measures. Consequently, neither political conservatism nor religiosity seem to add any predictive utility when RWA is included in the analysis.
In addition, consistent with Social Identity Theory (Tajfel & Turner, 1979) and Self-categorization Theory (Turner et al., 1987), the results of Study 1 showed that both stronger identification with community and America were correlated with more support for extreme counter-terrorism measures and less support for refugees. However, neither was a significant predictor of support for refugees in the regression model. Similar to the analysis of religiosity and political conservatism, identification with America and community were strongly correlated with RWA. The negative effect of RWA on support for refugees seems to overshadow any contribution of identification with America or community. Identification with America but not identification with community was however a significant predictor of more support for extreme counter-terrorism measures in the regression analysis. This finding suggests that among those with a strong national identity, extreme counter-terrorism measures might be perceived as an effective means of protecting the safety and enduring legacy of the acting country’s cultural heritage. Consequently, a focus on shared identity with America could have resulted in an increased interest in maximizing the needs and benefits of the in-group at the expense of the out-group. A strong focus on American identity might also be associated with a tendency to devalue or dehumanize potential victims of extreme counter-terrorism measures based on a salient perceived difference in nationality.
The generalizability of the results from Study 1 is limited given the small sample size and the recruitment of a convenience sample. Further, the interpretation of the results is limited due to certain methodological issues. First, while it was our assumption that participants would conceptualize refugees and potential civilian casualties as religious out-group members, religious identity/affiliation was not explicitly measured. Second, demographic questions including political orientation and religiosity were asked before assessing the other variables of interest. Asking these demographic questions first could have primed various aspects of social identity thereby maximizing the potential effects of perceived sub-group or common-group status on the issues.
Study 2
The goal of Study 2 was to replicate, in a sample of university students, the ability of IWAH to predict greater support for refugees and less support for extreme counter-terrorism measures even when controlling for other relevant personality and demographic variables. We also aimed to improve upon some methodological limitations in Study 1, by measuring religiosity and political orientation after attitudes towards the issues. In addition to replicating the potential consequences of individual differences in IWAH for support for refugees and extreme counter-terrorism measures, we aimed to test predictors of individual differences in IWAH. Specifically, we focused on self-reported parenting practices.
The specific parenting practices of interest were control, nurturance, and over-protection. Control refers to the degree to and manner in which parents try to place guidelines, restrictions, and boundaries on children’s behavior (Buri, 1991). Parental control can be conceptualized as permissive, in which few limits or guidelines are provided, authoritarian, in which limits and guidelines are rigidly enforced through threats of punishment, or authoritative, in which limits and guidelines are flexible and shaped through age appropriate reasoned arguments (Buri, 1991). Nurturance refers to the extent to which parents express affection, concern, and warm sentiment towards their children (Buri, 1989). Over-protection refers to parental practices that are over-involved, intrusive, and stifle the development of autonomy (Parker, Tupling, & Brown, 1979).
In their description of the development of the authoritarian personality, Adorno, Frenkel-Brunswik, Levinson, and Sanford (1950) postulated an influence in child-parent dynamics. Specifically they argued that, as children, high authoritarians were likely to have held overly idealistic views of their parents and to have displayed an unquestioning obedience to parental authority out of a sense of obligation and fear of punishment. They also suggested that parental neglect and unfair parental discipline were factors leading to the development of an authoritarian personality. To the extent that an authoritarian personality would be antithetical to IWAH, it seemed reasonable to expect that peopled reared by parents using authoritarian parenting practices might have less IWAH.
McFarland et al. (2012) speculated about the possible developmental origins of individual differences in IWAH being related to parenting practices. Specifically, they posited that children who were neglected by their parents and children who were spoiled or overindulged by their parents would grow up to be adults who lacked a sense of IWAH. These predictions were based on Adler’s (1927/1954) concept of social interest and his theorizing that lack of parental displays of affection produced a sense that the world was a dangerous place and that this perception resulted in a consequential lack of concern for others out of fear or hostility. Adler also theorized that children who were spoiled by their parents would be too self-absorbed to have a civic interest in the welfare of others. McFarland et al. (2012) also speculated that both excessive punishment by parents and a lack of parental nurturance and affection would lead to low levels of IWAH in adults. This speculation was based on the assumption that ethnocentrism characterized by high levels of SDO and RWA is antithetical to IWAH and on Duckitt’s (2001) dual process model that assumes parental insistence on obedience to authority out of fear of punishment increases RWA, whereas a deficit of parental displays of caring increases SDO in the form of indifference to social inequalities. In contrast to these speculations about the relation of parenting behaviors to the development of IWAH, McFarland et al. (2013) reviewed the results of an unpublished study conducted in their lab, in which adults’ IWAH scores were unrelated to their retrospective assessment of their parents on dimensions such as being affectionate and supporting and being punitive.
In spite of the previous null findings (McFarland et al., 2013), given the logically compelling theoretical link between parenting practices and IWAH, further research seemed warranted. Consequently, in Study 2, we further examined the possible link between parenting practices and individual differences in IWAH using established self-report measures of the theoretically grounded parenting dimensions of nurturance, control, and overprotection. We hypothesized that the authoritarian type of parental control involving rigid discipline based on fear of punishment would be associated with more RWA and therefore with less IWAH. Similarly, we hypothesized that a lack of parental nurturance would be associated with more SDO and therefore lower levels of IWAH. Further, we hypothesized that recipients of permissive and over-protective parenting would have lower levels of IWAH because such parenting practices would promote being self-absorbed and lacking a sense of duty or obligation to others.
Method
Participants
Participants were 126 students enrolled in introductory psychology classes at a small state-supported university located in the southeastern United States. Data were collected during the fall 2015 and spring 2016 semesters. The participants ranged in age from 18-21 years (M = 18.92, SD = 0.88). Of the participants, 34 were men, 91 were women, and one person failed to specify her/his gender. Of the participants, 71 identified their race/ethnicity as White, 42 as Black, 5 as Hispanic, 2 as Asian, and 6 as other or unspecified. Of the participants, 12 identified their religious affiliation as Catholic, 92 as Protestant, 1 as Buddhist, 1 as Hindu, 1 as Muslim, 5 as atheist, 7 as agnostic, and 7 as other or unspecified. For their participation, students received credit towards the partial fulfilment of a research participation requirement.
Materials
Participants read the same passage about refugees and answered the same questions about support for refugees described in Study 1. Similarly, support for extreme counter-terrorism measures, IWAH, and RWA were all measured the same way as described in Study 1. SDO was measured the same way as described in Study 1, with the exception that all 14 items from the Social Dominance Orientation Scale were included (Pratto, Sidanius, Stallworth, & Malle, 1994).
Religion
In addition to the measures of religious attendance and religious importance described in Study 1, participants in Study 2 were also asked to answer yes or no in response to the question “Do you consider yourself to be a religious person?” and were asked to specify their religious affiliation. The average religious attendance was 2.79 (SD = 2.50) and the average rated religious importance was 5.06 (SD = 1.96). Religious attendance and importance were strongly correlated (r = .50) so they were transformed into z scores before being averaged to form a composite measure of religiosity.
Political orientation
Participants rated their overall political orientation, their position on foreign policy issues, their position on economic issues, and their position on social issues. These ratings were made on a 5-point scale anchored at 1 (very liberal) and 5 (very conservative). Responses to the four questions were averaged to obtain a composite measure of political orientation.
Parenting styles
Participants completed the Parental Authority Questionnaire (Buri, 1991), the Parental Nurturance Scale (Buri, 1989), and the Parental Bonding Instrument (Parker, Tupling, & Brown, 1979). Participants were asked to make responses to all three scales in reference to their primary caregiver (defined for them as the person who was the most important influence on them as they were growing up).
The Parental Authority Questionnaire is a 30-item measure that assesses respondents’ perceptions of the ways in which their parents attempted to exert control over their behavior as they were growing up. Ten items assess the authoritarian style of exerting control characterized by a dogmatic and inflexible insistence on obedience based on fear of punishment. An example item was, “Whenever my primary caregiver told me to do something as I was growing up, she/he expected me to do it immediately without asking any questions.” Ten items assess the permissive approach to control characterized by a lack of setting boundaries or enforcing rules. An example item was, “As I was growing up, my primary caregiver seldom gave me expectations and guidelines for my behavior.” Ten items assess the authoritative approach to control characterized by setting appropriate rules and trying to get children to understand and internalize the rationale behind those rules. An example item was, “As I was growing up, once family policy had been established, my primary caregiver discussed the reasoning behind the policy with the children and family.” Participants made their ratings on a scale from 1 (strongly disagree) to 5 (strongly agree).
The Parental Nurturance Scale is a 24-item measure of respondents’ perceptions about the extent to which their parents expressed concern and displayed affection towards them as they were growing up. An example item was “My primary caregiver expressed her/his warmth and affection for me.” Participants made their ratings on a scale from 1 (strongly disagree) to 5 (strongly agree).
The Parental Bonding Instrument is a 25-item measure of respondents’ perceptions of attitudes and behaviors parents displayed towards them during the first 16 years of their life. Twelve of the items assess care or the extent to which parents displayed signs of warmth and affection. An example item was “Spoke to me with a warm and friendly voice.” Ratings were made on a scale from 1 (very like) to 4 (very unlike). Thirteen of the items assess overprotection or the extent to which parents excessively restricted children’s behavior and inhibited their autonomy. An example item was “Felt I could not look after myself unless she/he was around.” Ratings were made on a scale from 1 (very like) to 4 (very unlike). Responses were recoded so that higher scores indicated more care and more overprotection respectively.
Procedure
Participants signed up for a study that was presented to them as an investigation of “perceptions of parents, personality, and attitudes.” Participants arrived in groups of no more than 10 to a psychology laboratory. All participants completed the measures of support for extreme counter-terrorism, support for refugees, and IWAH before completing measures of SDO, RWA, demographics, religion, political conservatism, and parenting styles. Whether participants completed the IWAH measure before or after the measures of support for extreme counter-terrorism and support for refugees was counterbalanced. The order in which participants completed the measures of support for extreme counter-terrorism and support for refugees was also counterbalanced. All participants were debriefed about the true purpose and hypothesis of the study before being thanked and dismissed.
Results
Differences in Levels of Identification Across IWAH Subscales
Scores on the three IWAH subscales (community, America, all humanity) were subjected to a within-subjects analysis of variance, which revealed an overall difference in levels of identification, F(2, 250) = 66.9, p < .001, η2p = .436. Pairwise comparisons were conducted using paired sample t-tests. Participants identified more strongly with their community (M = 3.75, SD = 0.74) than with all humanity (M = 3.07, SD = 0.66), t(125) = -9.8, p < .001, d = .87. Participants identified more strongly with America (M = 3.49, SD = 0.67) than with all humanity, t(125) = -7.5, p < .001, d = .66. Participants identified more strongly with their community than with America, t(125) = 5.1, p < .001, d = .45.iv
Simple Order Correlations
Simple order correlations among all variables are presented in Table 3. Support for extreme counter-terrorism measures was associated with more religiosity, more political conservatism, more SDO, more RWA, more identification with community, and more identification with America. Support for extreme counter-terrorism measures was however not significantly related to IWAH. Support for refugees was associated with less political conservatism, less SDO, less RWA, and more IWAH. None of the measures of parenting styles were associated with IWAH. Further, none of the measures of parenting style were associated with support for extreme counter-terrorism measures or with support for refugees.v
Table 3
Simple Order Correlations Among All Variables in Study 2
| Variable | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | 15 | 16 |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 1. REL | -- | .18* | .06 | .49** | .34** | .32** | .13 | .20* | -.03 | .19* | -.18* | .15 | .21* | -.06 | .07 | -.01 |
| 2. PC | -- | -- | .36** | .61** | .17 | .28** | -.04 | .22* | -.35** | .12 | -.18* | .11 | .06 | -.01 | -.12 | -.12 |
| 3. SDO | -- | -- | -- | .44** | -.05 | -.01 | -.32** | .27** | -.39** | .09 | .08 | .09 | -.01 | -.03 | -.07 | .03 |
| 4. RWA | -- | -- | -- | -- | .31** | .40** | .02 | .44** | -.42** | .25** | -.19* | .19* | .15 | -.03 | -.18* | -.08 |
| 5. IC | -- | -- | -- | -- | -- | .68** | .37** | .20* | .07 | .16 | -.04 | .25** | .21* | .05 | -.01 | -.19* |
| 6. IA | -- | -- | -- | -- | -- | -- | .53** | .39** | -.11 | .15 | -.13 | .21** | .10 | -.01 | -.09 | -.28* |
| 7. IWAH | -- | -- | -- | -- | -- | -- | -- | .10 | .30** | -.05 | .04 | -.01 | -.02 | .04 | -.06 | .09 |
| 8.SFECT | -- | -- | -- | -- | -- | -- | -- | -- | -.38** | .10 | -.07 | .12 | .04 | -.08 | -.05 | .06 |
| 9. SFR | -- | -- | -- | -- | -- | -- | -- | -- | -- | -.14 | .11 | -.11 | -.04 | -.03 | .08 | -.08 |
| 10. APC | -- | -- | -- | -- | -- | -- | -- | -- | -- | -- | -.34* | -.15 | -.25** | -.10 | .11 | .01 |
| 11. PPC | -- | -- | -- | -- | -- | -- | -- | -- | -- | -- | -- | .05 | .08 | -.06 | -.05 | .14 |
| 12. ATPC | -- | -- | -- | -- | -- | -- | -- | -- | -- | -- | -- | -- | .63** | -.07 | -.06 | -.19* |
| 13. PN | -- | -- | -- | -- | -- | -- | -- | -- | -- | -- | -- | -- | -- | .08 | -.05 | -.09 |
| 14. PAC | -- | -- | -- | -- | -- | -- | -- | -- | -- | -- | -- | -- | -- | -- | -.47** | -.11 |
| 15. PO | -- | -- | -- | -- | -- | -- | -- | -- | -- | -- | -- | -- | -- | -- | -- | .06 |
| 16. Age | -- | -- | -- | -- | -- | -- | -- | -- | -- | -- | -- | -- | -- | -- | -- | -- |
| α | -- | .86 | .88 | .91 | .89 | .84 | .82 | .73 | .83 | .81 | .70 | .82 | .92 | .97 | .78 | -- |
| M | 0.00 | 3.21 | 2.51 | 92.85 | 3.75 | 3.49 | 3.06 | 2.77 | 3.02 | 3.60 | 2.28 | 3.59 | 4.22 | 2.78 | 2.36 | 18.92 |
| SD | 0.88 | 0.85 | 1.06 | 30.31 | 0.74 | 0.67 | 0.66 | 0.87 | 0.84 | 0.65 | 0.57 | 0.68 | 0.71 | 1.03 | 0.54 | 0.88 |
Note. REL = religiosity; PC = political conservatism; SDO = social dominance orientation; RWA = right wing authoritarianism; IC = identification with community; IA = identification with America; IWAH = identification with all humanity; SFECT = support for extreme counter-terrorism measures; SFR = support for refugees; APC = authoritarian parental control; PPC = permissive parental control; ATPC = authoritative parental control; PN = parental nurturance; PAC = parental care; PO = parental overprotection.
*p < .05. **p < .01.
Partial Correlations
The partial correlation of IWAH and support for extreme counter-terrorism measures was not statistically significant when controlling for identification with America and identification with community, r(122) = -.13, p = .139. The partial correlation of IWAH and support for refugees was still statistically significant even when controlling for identification with America and identification with community, r(122) = .43, p < .001. None of the partial correlations of IWAH and measures of parenting style were statistically significant when controlling for identification with America and identification with community (rs < .16, ps > .05).
Regression Analyses
The support for extreme counter-terrorism measures and support for refugees scores were subjected to separate hierarchical multiple regression analyses. The grand mean centered continuous variables age, religiosity, political conservatism, RWA, and SDO were entered in Step 1. Identification with community and identification with America were entered in Step 2, and IWAH was entered in Step 3. As can be seen in Table 4, both RWA and identification with America were associated with more support for extreme counter-terrorism measures. IWAH was not a significant predictor of support for extreme counter-terrorism measures. As can be seen in Table 4, both RWA and SDO were associated with significantly less support for refugees. Identification with community was associated with significantly more support for refugees. IWAH was associated with more support for refugees even when controlling for all the other variables in the model.
Table 4
Predictors of Support for Extreme Counter-Terrorism Measures and Support for Refugees in Study 2
| Predictor Variable | Step 1
|
Step 2
|
Step 3
|
|||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| B | SE | 95% CI | β | B | SE | 95% CI | β | B | SE | 95% CI | β | |
| SFECT
|
||||||||||||
| REL | -.018 | .094 | -.204 – .168 | -.018 | -.052 | .091 | -.232 – .128 | -.053 | -.052 | .091 | -.232 – .128 | -.053 |
| Age | .083 | .081 | -.076 – .243 | .085 | .161 | .079 | .005 – .317 | .164* | .164 | .079 | .007 – .321 | .167* |
| PC | -.085 | .107 | -.297 – .127 | -.083 | -.123 | .102 | -.325 – .078 | -.121 | -.126 | .102 | -.328 – .076 | -.123 |
| RWA | .013 | .004 | .006 – .020 | .467** | .010 | .003 | .003 – .017 | .349** | .010 | .003 | .003 – .017 | .347** |
| SDO | .072 | .076 | -.079 – .223 | .088 | .125 | .073 | -.020 – .270 | .152 | .117 | .077 | -.035 – .269 | .142 |
| IC | -.145 | .125 | -.393 – .102 | -.123 | -.145 | .125 | -.394 – .103 | -.123 | ||||
| IA | .556 | .144 | .270 – .841 | .430** | .584 | .163 | .262 – .906 | .453** | ||||
| IWAH | -.050 | .129 | -.306 – .205 | -.038 | ||||||||
| r2 | .212 | .310 | .311 | |||||||||
| SFR
|
||||||||||||
| REL | .169 | .087 | -.004 – .342 | .177 | .142 | .088 | -.032 – .317 | .149 | .144 | .083 | -.021 – .309 | .151 |
| Age | -.101 | .075 | -.249 – .047 | -.107 | -.103 | .076 | -.254 – .048 | -.109 | -.127 | .073 | -.271 – .017 | -.134 |
| PC | -.089 | .099 | -.286 – .108 | -.090 | -.076 | .099 | -.271 – .119 | -.077 | -.053 | .094 | -.239 – .132 | -.054 |
| RWA | -.010 | .003 | -.017 – -.004 | -.374** | -.010 | .003 | -.017 – -.004 | -.376** | -.010 | .003 | -.016 – -.004 | -.358** |
| SDO | -.159 | .071 | -.299 – -.019 | -.201** | -.154 | .071 | -.295 – -.012 | -.194* | -.079 | .070 | -.218 – .061 | -.100 |
| IC | .272 | .121 | .032 – .513 | .239* | .273 | .115 | .045 – .500 | .240* | ||||
| IA | -.217 | .140 | -.495 – .060 | -.175 | -.475 | .149 | -.771 – -.180 | -.382** | ||||
| IWAH | .450 | .118 | .216 – .684 | .355** | ||||||||
| r2 | .270 | .301 | .378 | |||||||||
Note. SFECT = support for extreme counter-terrorism measures; SFR = support for refugees; REL = religiosity; PC = political conservatism; RWA = right wing authoritarianism; SDO = social dominance orientation; IC = identification with community; IA = identification with America; IWAH = identification with all humanity.
**p < .01. *p < .05.
Discussion
The results of Study 2 partially replicated the finding from Study 1 in that IWAH was associated with more support for refugees even when controlling for other personality and demographic variables. These findings offer further support for the idea that support for policies that seem callous towards the suffering of refugees from predominantly Muslim countries in the Middle East may stem from a tendency to focus on the ways in which these people are different from the self. In contrast, feeling a sense of shared identity and common humanity with refugees increases compassion towards them and interest in assisting them.
Unlike in Study 1, where IWAH explained a statistically significant but small amount of variability in lack of support for extreme counter-terrorism measures, in Study 2, IWAH was not a significant predictor of this measure. This might be a product of differences in the sample. The sample in Study 2 was entirely students in introductory psychology classes. They were younger and higher on RWA than were the participants in Study 1. There was also less variability in support for extreme counter-terrorism measures in Study 2. Therefore, perhaps the lack of relationship between IWAH and support for extreme counter-terrorism was due to the fact that there were fewer participants in the Study 2 sample on the low end of support for extreme counter-terrorism measures.
The results of Study 2 also replicated the results of Study 1 in terms of the effects of RWA and identification with America. RWA was a significant predictor of both more support for extreme counter-terrorism measures and less support for refugees. Identification with America was a significant predictor of more support for extreme counter-terrorism.
The results of Study 2 differed from those of Study 1 in terms of the effects of SDO and identification with community. In Study 2, SDO was a significant predictor of less support for refugees but was not a significant predictor of support for extreme counter-terrorism measures. In Study 2, identification with community was associated with more support for refugees. These differences may have been partly a function of the greater ethnic diversity represented in the Study 2 sample. For example, SDO was significantly correlated with support for extreme counter-terrorism among White participants but the relation between the two variables was not statistically significant among non-White participants. The motive to defend existing status hierarchies may lead to greater tolerance of potential harm to those who are viewed as threats to disrupt those hierarchies but this effect may be limited to majority group members who benefit from the existing social structures promoting group-based-inequalities.
The results of Study 2 failed to find any support for the hypothesized relation between parenting styles and IWAH. This lack of relation is consistent with previously reported results showing no association between participants’ levels of IWAH and their retrospective assessment of their parents (McFarland et al., 2013). These null findings are somewhat surprising given the seemingly sound theoretical rationale offered by McFarland et al. (2012) for the role of parenting practices in the development of individual differences in IWAH. Authoritarian methods of exerting parental control were associated with more RWA in the current study, suggesting that the theoretical rationale for predicting an association of parenting practices to individual differences in IWAH was sound.
The results of Study 2 are limited due to the characteristics of the sample. There was little variability in the religiosity of the sample, with very few participants identifying as atheist or agnostic. A large number of questions was used to assess parenting behaviors and this may have taxed participants’ motivation and promoted inattentive responding. Further, the current study relied on emerging adults’ retrospective reports about their parents’ practices. Future research might consider using reports directly from parents. Parenting behaviors may vary as a function of socio-economic status and education level and these variables were not assessed in the present study. Additional research using more diverse samples and different methodologies is needed to further explore the possible role parents and other agents of socialization may play in the development of individual differences in IWAH and how parenting practices and IWAH are related to political issues involving the suffering of out-group members.
General Discussion
The results of the present research revealed that, in two different samples, IWAH was associated with greater support for refugees. Consequently, it seems that the tendency to feel a shared commonality with people regardless of ethnicity, nationality, religion, or other potentially divisive categorizations is associated with more support for offering aid and asylum to refugees. This association could reflect the fact that people high on IWAH conceptualize refugees as members of the superordinate category human beings resulting in a perceived necessity of extending them the same sense of compassion and bearing the same moral obligation towards helping them as would be extended to any other member of that category. In contrast, people low on IWAH may conceptualize refugees as members of subordinate categories based on ethnicity, nationality, or religion. The results of this categorization might be to perceive refugees as out-group members thereby potentially reducing perceived compassion or moral obligations to offer aid or at least minimizing the importance of the suffering of these out-group members as less important than the obligation to protect the perceived economic and security interests of the in-group.
Additional consistent findings across both studies were that RWA was associated with less support for refugees but more support for extreme counter-terrorism measures and that identification with America was associated with more support for extreme counter-terrorism measures. The findings regarding RWA are consistent with previous research showing that authoritarianism predicts prejudice against outgroups (McFarland, 2010), less commitment to human rights (McFarland & Mathews, 2005), less support for asylum seekers (Lyall & Thorsteinsson, 2007), and greater support for military action in the Middle East when it is perceived as reducing threats of terrorism (Crowson, 2009; McFarland, 2005). The findings regarding identification with America are consistent with Social Identity Theory (Tajfel & Turner, 1979) and Self-categorization Theory (Turner et al., 1987), suggesting that the tendency to focus on nationalistic identity results in the privileging of the rights and welfare of those perceived as sharing the common identity and a callousness towards those perceived as falling outside the parameters of the shared identity.
There were some inconsistencies in the findings across the two studies. Most notably, IWAH was associated with less support for extreme counter-terrorism in Study 1 but not in Study 2. Similarly, while SDO was associated with less support for refugees in both studies, it was associated with more support for extreme counter-terrorism only in Study 1 and not in Study 2. Identification with community was associated with more support for refugees but only in Study 2 and not in Study 1. The differences in the findings across the two studies could be a function of the demographic differences in the samples. The sample in Study 2 was composed of exclusively university students; the respondents were younger and more ethnically diverse than those in Study 1. In addition, the discrepancies might also have to do with the situational salience of social identity. In Study 1, questions about religiosity and political orientation were asked at the beginning of the survey. Consequently, aspects of social identity based on nationality or religion would have been salient prior to the assessment of attitudes towards refugees and extreme counter-terrorism measures. In contrast, demographic questions were assessed at the end of the survey in Study 2. Consequently, aspects of social identity may have been less salient. Additional research is needed to resolve these discrepancies.
The consistent finding across both studies that IWAH was associated with greater support for refugees is especially relevant given the growing humanitarian crisis stemming from the increasing number of refugees, the call for greater aid from countries not contiguous to the main areas from which refugees are fleeing (Guterres, 2014; UN, 2017), the political opposition to supporting refugees (Marcos, 2015; Seipel, 2015) and the widespread negative attitudes towards immigrants and asylum seekers in both America (Hartig, 2018) and Europe (Wike et al., 2016). Although correlational, the present findings point to the possible utility of fostering a sense of shared human identity as a means of increasing compassion and a desire to provide aid in response to the refugee crisis.
McFarland et al. (2013) speculated that IWAH might be increased by exposing people to case studies of moral heroes who had demonstrated altruism out of a shared sense of humanity. For example, they cited the case of Japanese diplomat Chiune Sugihara who at risk to himself and in violation of policy issued visas to thousands of Jewish refugees thereby saving them from the Nazis. In addition, McFarland (2011) speculated about the possible role of media as a socializing agent that might be used to foster a greater sense of shared humanity. He highlighted the children’s television show The Big Blue Marble, which aired on PBS from 1974-1983 and featured stories about children from all around the world, as a possible means of teaching people to have a greater appreciation of diverse cultures and consequently perhaps a greater sense of IWAH. McFarland (2011) also referred to photographer Edwards Steichen’s attempt to promote a sense of shared humanity in his work The Family of Man (1955), in which he captured images of people from diverse cultures engaged in universal human activities such as birth, marriage, family gatherings, and play.
There are two different theoretical predictions about how priming IWAH might effect reactions to refugees. The common in-group identity model (Gaertner & Dovidio, 2000) predicts that priming people to think of refugees as common-group members rather than sub-group members should increase support for offering refugees aid and asylum. In contrast, the in-group projection model (Wenzel, Mummendey, & Waldzus, 2007) suggests that in-group members oftentimes think of their in-group as being more prototypically representative of the larger superordinate category. Consequentially, asking in-group members to re-conceptualize out-group members as belonging to a superordinate category can result in more negative feelings towards the out-group members. This increased negativity is based on efforts to legitimize initial perceptions of the in-group’s greater claim to representativeness in this category and therefore greater deservingness of any benefits ensuing from membership in the superordinate category. From this perspective, priming people to think of refugees as in-group members (fellow human beings) rather than out-group members (based on ethnicity, nationality, or religion) might actually decrease positive sentiments towards refugees and consequentially decrease support for offering them aid and asylum. If members of a potential host country viewed themselves as somehow more prototypically human (if they viewed refugees as infrahuman based on their ethnicity, nationality, or religion) then attempts to force them to focus on their shared humanity with refugees might backfire and strengthen perceptions that refugees were less entitled to aid or access to the country due to their less legitimate standing as full members of the superordinate category.
The limited available research seems to favor the common in-group identity prediction. For example, Reese, Proch, and Finn (2015) showed that priming German university students with images of cultural diversity (a poster with hands of different skin tones holding up a globe) increased IWAH and willingness to donate to global charities. Similarly, priming people with evidence for genetic similarities among ethnic groups in conflict (Arabs and Jews) reduced inter-ethnic hostility and increased support for peaceful means of conflict resolution (Kimel, Huesmann, Kunst, & Halperin, 2016). Extrapolating from these results, it seems that manipulations that promote a focus on common group identity among people result in more favorable attitudes towards and reactions to others who belong to different subgroups. A fruitful avenue for future research would be to see if priming IWAH could influence reactions to refugees. It seems plausible that priming members of potential host nations with stimuli highlighting their common group identity (shared humanness) might increase willingness to provide aid or asylum to refugees.
Empathy is an additional variable warranting further research as a mechanism potentially underlying the relation between IWAH and support for refugees and potentially offering a means of enhancing acceptance of refugees. Previous research has established that empathy is associated with more support for global human rights (McFarland & Mathews, 2005) and more IWAH (McFarland et al., 2012). Although empathy was not measured in the current studies, it is plausible that feeling a shared sense of humanness with refugees corresponded to a greater ability to empathize with their emotional state and therefore to feeling a greater sense of compassion for their suffering and desire to alleviate that suffering. Consequently, interventions aimed at increasing empathy by promoting real or imagined contact with refugees or by promoting an appreciation for diverse cultures might be effective in creating more accepting attitudes towards refugees. However, increasing empathy does not always translate into greater helping when groups differ in status and have different goals (Zaki & Cikara, 2015).
Future research is needed to determine if it is possible to change attitudes towards refugees by enhancing IWAH. If it is possible, it is important to determine whether a change in adopting a superordinate criterion for group identity, enhanced empathy, or some other mechanism is responsible for the effect, and whether the effect of such manipulations would apply to those with ethnocentric personalities or be limited to people with moderate to high dispositional levels of IWAH.