%20-%20https://sketch.com%20--%3e%3ctitle%3e3ECBBF50-FD08-412A-943D-2CFC5E025A16%3c/title%3e%3cdesc%3eCreated%20with%20sketchtool.%3c/desc%3e%3cg%20id='Stylesheet'%20stroke='none'%20stroke-width='1'%20fill='none'%20fill-rule='evenodd'%3e%3cg%20transform='translate(-926.000000,%20-668.000000)'%20id='BUTTONS'%3e%3cg%20transform='translate(100.000000,%20597.000000)'%3e%3cg%20id='BTN-/-RIGHTS'%20transform='translate(816.000000,%2065.000000)'%3e%3cg%20id='ICON-/-Rights'%20transform='translate(10.000000,%206.000000)'%3e%3cpath%20d='M13.5752841,0%20L1.34943182,0%20C0.59375,0%200,0.586666667%200,1.33333333%20L0,22.6666667%20C0,23.4133333%200.620738636,24%201.34943182,24%20L17.6505682,24%20C18.40625,24%2019,23.3866667%2019,22.6666667%20L19,5.36%20L13.5752841,0%20Z%20M13.5752841,1.89333333%20L17.0838068,5.36%20L13.5752841,5.36%20L13.5752841,1.89333333%20Z%20M1.34943182,22.6666667%20L1.34943182,1.33333333%20L12.1988636,1.33333333%20L12.1988636,5.36%20C12.1988636,6.10666667%2012.7926136,6.69333333%2013.5482955,6.69333333%20L17.6235795,6.69333333%20L17.6505682,22.6666667%20L1.34943182,22.6666667%20Z'%20id='Shape'%20fill='%23595959'%20fill-rule='nonzero'%3e%3c/path%3e%3ccircle%20id='Oval'%20stroke='%23595959'%20stroke-width='1.5'%20fill='%23595959'%20cx='9.5'%20cy='13.5355339'%20r='5.5'%3e%3c/circle%3e%3cpolyline%20id='Path-2'%20stroke='%23FFFFFF'%20stroke-width='1.5'%20transform='translate(9.500000,%2013.035534)%20rotate(45.000000)%20translate(-9.500000,%20-13.035534)%20'%20points='8%2016.0355339%2011%2016.0355339%2011%2010.0355339'%3e%3c/polyline%3e%3c/g%3e%3c/g%3e%3c/g%3e%3c/g%3e%3c/g%3e%3c/svg%3e)
“The people who say the mosque is too close to Ground Zero, those are the same people that protest mosques in Brooklyn and Staten Island and Tennessee and Wisconsin and California. What radius will they go for?”
Ali Akram, New York City resident, quoted by Reuters.In 2010, a debate erupted about plans to construct a mosque (as part of a larger multicultural center called the Cordoba House Project) approximately two blocks from Ground Zero in New York City. Central to this debate was the notion of distance and the idea that two blocks was “too close” to place a mosque in relation to the site of the September 11 terrorist attacks at the World Trade Center. Indeed, the project was referred to as the “Ground Zero Mosque” by many political elites, such as politicians, pundits, and journalists.Footnote 1 Elites across the ideological spectrum expressed concern about the project and even the Anti-Defamation League stated, “We believe the City of New York would be better served if an alternative location could be found.”Footnote 2
Polling conducted at the time revealed that between two-thirds and three-fourths of Americans agreed that Muslims had the “right” to build the mosque, yet a majority of Americans opposed the building of the mosque.Footnote 3 Was opposition to the mosque caused by concern for the sensitivities of those who lost loved ones in the September 11 attacks or by more general intolerance for Muslim Americans? At the time, many individuals argued that it was a reflection of the former, often citing surveys that showed that although Americans opposed a mosque near Ground Zero an overwhelming majority claimed that they would support the construction of mosques in their own communities.Footnote 4 In this article, I adjudicate between these two interpretations of public opinion and examine whether opposition to the mosque was truly a function of its proposed location.
EXPLAINING PUBLIC OPPOSITION TO THE MOSQUE
Most news accounts of the debate over the Cordoba House Project emphasized the notion that the primary reason for opposing the building of a mosque so close to Ground Zero was that it would be insensitive to those who had been affected by the attacks on September 11. This argument was cited by a variety of opinion leaders ranging from Republicans Sarah Palin and Rudolph Giuliani to Democrats Harry Reid and Howard Dean. The importance of the location of the Cordoba House relative to the site of the World Trade Center attacks was also evident through the attempts of some politicians to resolve the issue. The most prominent example of such a proposal was made by then-governor David Paterson, who offered state assistance in finding an alternative location for the mosque farther away from Ground Zero.
Elites cited a sensitivity to the location of the project as distinct from opposition motivated by intolerance, however, strong theoretical reasons indicate that the proximity of the mosque was significant because it activated individuals to think of the project in terms of inter-group conflict. Indeed, tolerance for out-groups is often conditioned by the proximity of groups to each other. Perhaps the best-known example of this phenomenon in the political science literature is V.O. Key's (Reference Key1949) theory of racial threat, which posits that majority groups will react to the increasing influence of minority groups when those groups are located more proximately to the members of the majority group. Indeed, Key found that whites were more likely to vote in elections when the percentage of African Americans living nearby increased. Key's racial threat theory is an example of a broader tendency of majority group members to react in intolerant ways when their status is threatened by minority out-groups (Blumer Reference Blumer1958; Marcus et al. Reference Marcus, Sullivan, Theiss-Morse and Wood1995; Stouffer Reference Stouffer1992). Muslim Americans are a particularly likely target for a lack of tolerance because they tend to be viewed less favorably by Americans than most other minority groups (Panagopoulos Reference Panagopoulos2006; Schafer and Shaw Reference Schafer and Shaw2009; Sides and Gross Reference Sides and Gross2007).
Central to the racial threat thesis is that the proximity of the minority group made that group salient for the political actions taken by majority group members. Similarly, more recent research has found that where one lives relative to the Mexican border has notable affects on attitudes about immigration policies (e.g., Branton et al. Reference Branton, Dillingham, Dunaway and Miller2007). As an out-group becomes more proximate to an in-group, the tensions between the out-group and in-group become more pronounced, and therefore will more likely affect whether group threat affects how one thinks and acts in the social and political realm. In the case of the Cordoba House Project, spatial proximity was not defined in relation to where majority group members lived, but relative to a site that serves for most Americans as the most important memorial to the September 11 attacks (Low Reference Low2004). On the one hand, the proximity of the Cordoba House Project to the World Trade Center site may have caused many citizens to feel threatened by the out-group behind the project, thereby causing those individuals to be less tolerant and supportive of the project. If that proximity was crucial in reducing support for the Cordoba House Project, then a similar project located farther from the World Trade Center site should be less likely to activate feelings of group threat and, therefore, win more support from the majority group.
On the other hand, the project's proximity to the World Trade Center may be of less importance for affecting support for the project because most Americans live far from the memorialized location and therefore will have little direct interaction and experience with what happens in the area. In fact, the importance of the arguments made about the project's proximity to the World Trade Center site may have been most significant in providing citizens with cues on the debate (Zaller and Feldman Reference Zaller and Feldman1992). Politicians were not perfectly divided along party lines on this issue. Several noteworthy Democrats (including Harry Reid and Democratic members of the New York congressional delegation) opposed the location of the mosque while some Republicans (for example, Orrin Hatch and Ron Paul) supported it. Nevertheless, most of the vocal critics of the project were Republicans, and many Democrats, including President Obama, gave their support for the right of the organizers to build the mosque if not the actual plan to do so. Thus, conservatives and Republicans may have taken cues from Republicans/conservatives leading them to be more likely to oppose the project while Democrats taking cues from Democrats/liberals would have been more likely to support it.
DATA
To examine the effect of distance on attitudes toward the Cordoba House Project, I analyze data from the 2010 Cooperative Congressional Election Study (CCES).Footnote 5 To gauge whether support for the mosque was related to where the mosque was located in relation to Ground Zero, I included this question on the post-election survey: “Would you support or oppose the building of a mosque and Islamic cultural center ____ blocks from the site of the World Trade Center in New York City where the September 11th terrorist attacks occurred?” Rather than asking all respondents about a center located two blocks from the World Trade Center site, each of 2,068 respondents was randomly assigned an integer between 2 and 16.
The choice of the lower range of blocks was based on the actual proposed location for the center, which was commonly described in the news as two blocks from the World Trade Center site. The choice for the upper end of the range was selected with two considerations in mind. First, during the controversy, the New York Times publicized the fact that two mosques were already located near the World Trade Center: one four blocks away and the other 14 blocks from the site.Footnote 6 Thus, the range chosen for this experiment would include within it these two long-standing (and seemingly uncontroversial) mosques. The limit of 16 blocks was also chosen to preserve a reasonable sample size at each value/condition, thereby allowing for the detection of any potential threshold or nonlinear effects. Figure 1 shows a map of downtown Manhattan for perspective on how far away from the World Trade Center 16 blocks would be. The most direct route to 16 blocks would take an individual about .8 mile away from the World Trade Center, or about a 15-minute walk.Footnote 7
Map of Downtown Manhattan
The dependent variable is binary with 1 indicating that respondents supported the project and 0 indicating that they opposed it. The 12% of respondents who answered that they were not sure whether they supported the mosque were discarded from the analysis after I conducted a multinomial analysis of the data and found that a “not sure” response was unrelated to distance and other key variables included in the models presented. I also excluded 12 self-identified Muslim respondents from the analysis, although their inclusion did not affect the findings.Footnote 8 After these adjustments, 36% of respondents indicated that they supported the mosque whereas 64% were opposed.
In addition to testing for the effect of distance on support for the mosque, I also examine the other potential effects described previously. Ideology and party identification are both captured using seven-point scales. The party identification variable ranges from “strong Democrat” (coded 1) to “strong Republican” (coded 0). Ideology ranges from “very liberal” (coded 1) to “very conservative” (coded 0). I expect that liberals and Democrats will be more supportive of the mosque than Republicans and conservatives.
I also expect that individuals who are less acceptant and tolerant of out-groups will be less willing to support the building of the mosque, regardless of distance. Unfortunately, the survey did not include questions asking respondents to evaluate Muslims as a group. However, Kalkan et al. demonstrate that “positive affect for Muslims is largely determined by favorable views of other minority groups,” suggesting that individuals who are less tolerant of some out-groups are also likely to be less tolerant of Muslims (Kalkan, Layman, and Uslaner Reference Kalkan, Layman and Uslaner2009, 13). Thus, as a proxy for lack of out-group tolerance, I use each respondent's attitudes on immigration. Specifically, I draw from a question that asked respondents for their opinions on “a policy that deports the 11 million illegal immigrants living in the U.S.” Respondents could rate themselves on a scale ranging from 0 to 100, where 0 indicated that they strongly supported the policy and 100 indicated that they strongly opposed it. After controlling for ideology and partisanship, this question should be a reasonable proxy for tapping individuals' tolerance for outsiders.Footnote 9
Finally, I also control for additional factors that may have influenced a respondent's support for the mosque. Education takes on six values ranging from respondents without a high school degree (coded 1) to those with a postgraduate degree (coded 6). I expect that individuals with higher levels of education will be more supportive of the mosque because tolerance tends to increase with higher levels of education (Bobo and Licari Reference Bobo and Licari1989). I also included a control for an individual's level of religiosity. This variable was created as an index combining variables measuring how important religion is to the respondent's life, how often they attend church services, and how often they pray. The three component variables were included in a factor analysis and each loaded at .74 or above on the first factor. A standardized index of religiosity was generated from this factor, with higher values representing more religious individuals. I expect that more religious individuals will be less likely to support the mosque since religious differences will be more salient to them. Because individuals who know more about politics may have processed information about the debate differently than those with less information, I included a measure of each respondent's knowledge. This measure was based on how many of six factual questions about politics the respondent answered correctly.Footnote 10 Finally, gender is included as a dummy with female respondents coded 1 and males coded 0. Previous research finds that women are less supportive of out-groups exercising rights; therefore, I expect that women will be less likely to support the mosque relative to men (Marcus et al. Reference Marcus, Sullivan, Theiss-Morse and Wood1995).Footnote 11
RESULTS
Because the primary variable of interest, distance, was randomly assigned to respondents, it should be entirely uncorrelated with other factors that might influence one's support for the mosque.Footnote 12 Thus, determining whether distance influences support can be achieved through a simple bivariate analysis. Figure 2 plots the probability of supporting the building of a mosque depending on how many blocks away respondents were told the mosque would be from the World Trade Center. The plot is produced using a kernel-weighted local polynomial regression to account for potential non-linearities in the relationship between distance and support. The gray-shaded region indicates the 95% confidence interval around the estimates.
Kernel-Weighted Local Polynomial Regression Showing Relationship between Location of Mosque and Support for Mosque
Overall, the results presented in figure 2 indicate that support for the mosque only increased slightly when respondents were told that it would be located increasingly farther away from the World Trade Center. However, given the width of the 95% confidence intervals, we cannot be certain that there is any relationship between distance and support for the mosque. Of course, being unable to confidently state that a relationship exists does not necessarily mean that the relationship does not exist. Therefore, to determine how likely it is that no relationship exists between the location of the mosque and support for the mosque, I estimated two simple logit models. In the first model, I allowed logit to produce an estimate of the coefficient for the number of blocks away from Ground Zero the question stated that the mosque would be located (β blocks ). In the second model, I constrained β blocks to equal 0. Then, I calculated an approximation of the Bayesian posterior probability of a model given the data to determine the probability of the model where the coefficient for blocks is constrained to equal 0 (relative to the model where the coefficient for blocks is allowed to vary).Footnote 13 Because β blocks was the only term in the models, the calculation essentially provides the probability that the coefficient for distance is 0 given the data. The value produced by this calculation was .91; thus, we can be more than 90% confident that support for the mosque was unrelated to how far away from Ground Zero the mosque would be located.Footnote 14
One possible explanation for the lack of relationship between distance and support is that respondents who were already aware of the controversy over the mosque answered with their opinions of the mosque and its actual proposed location relative to the World Trade Center instead of the location specified in the survey question. To provide some leverage on determining whether the lack of effect can be attributed to this explanation, I reproduced the plot from figure 2 separately for respondents who had higher and lower levels of political knowledge.Footnote 15 If the previous explanation is causing the non-effects in figure 2, then we would expect to find a stronger effect for distance among individuals who are less engaged in politics (and therefore less knowledgeable) because those individuals are less likely to have heard about the mosque or already formed an opinion about it. Figure 3 presents these separate plots. The plots indicate that respondents' opinions about the mosque were unaffected by where it would be located regardless of their level of knowledge. Thus, it is unlikely that the non-finding for distance can be attributed to some respondents ignoring the information about distance in the question.
Kernel-Weighted Local Polynomial Regression Showing Relationship between Location of Mosque and Support for Mosque Depending on R's Political Knowledge
Yet another explanation for the lack of a relationship between distance and support is the possibility that most Americans have little sense about how far a “block” really is. To get a sense of whether this is dampening the effect of distance, I limited the analysis to respondents who lived within 50 miles of Ground Zero (N = 91) because these respondents are more likely to understand how far a block is. When I limited the analysis to just these respondents, the results were unchanged.Footnote 16
While we observe no relationship for distance in the full sample, perhaps the effect of proximity instead may be conditional on an individual's general level of tolerance for out-groups. Specifically, individuals with high levels of tolerance may support the project regardless of distance whereas those with low tolerance may oppose the project without regard to its location. However, for those in the middle of the tolerance scale, distance might matter more for conditioning support or opposition to the project. To investigate this possibility, figure 4 plots the relationship between distance and support for the Cordoba House Project depending on the individual's level of support for deporting immigrants.Footnote 17 Overall, figure 4 offers little support for the hypothesized conditional effects. As expected, individuals who place themselves at either end of the deportation scale were consistent (and quite far apart) in their support for or opposition to the mosque. Although a slight upward trend appears in support for the Cordoba House Project among those indicating moderate support for deporting immigrants, the increase is substantively small and not statistically significant.Footnote 18
Kernel-Weighted Local Polynomial Regression Showing Relationship between Location of Mosque and Support for Mosque Depending on R's Level of Support for Deporting Immigrants
The analysis of the survey experiment indicates that how far the mosque would be located from Ground Zero appeared to have little, if any, effect on how Americans viewed the project.Footnote 19 This raises the possibility that attitudes on the issue were not at all tied to the location of the mosque. To determine what factors did affect attitudes toward the mosque, table 1 presents the estimates from a logit model predicting support for the mosque. Five coefficients in the model are statistically significant at p < .05, each in the expected direction. To better understand the strength of the effects for each of these variables, figure 5 presents the effects for each variable on the predicted probability of supporting the mosque while holding the other variables in the model at their means.
Results from Logit Model Estimating Support for Mosque
N = 1,575. Adjusted Count R2 = 0.608. Significance levels:
* p < .05,
** p < .01. WTC = World Trade Center
Relationship between Independent Variables and Predicted Probability of Supporting Mosque
Based on the predicted effects presented in figure 5 the most important factors shaping views on this issue were partisanship, ideology, and support for the deportation of immigrants, although political knowledge and religiosity had somewhat weaker effects on support for the mosque. Given the politicization of this issue it is not surprising to find that Republicans and conservatives were much more likely to oppose building the mosque while Democrats and liberals were much more supportive. The partisan divide was documented in several other surveys collected during the debate as well.
What is notable is that even after controlling for ideology and partisanship, an individual's support for the deportation of illegal immigrants was strongly associated with opposition to the mosque. When a respondent was strongly supportive of deporting illegal immigrants, his or her predicted probability of supporting the mosque was less than .2. However, if the same respondent strongly opposed deporting illegal immigrants, the predicted probability of supporting the mosque increased to .75. Thus, individuals who are less tolerant of outsiders were also less willing to support building the mosque, regardless of how many blocks away from Ground Zero the mosque would be located.
CONCLUSION
During the controversy surrounding the Cordoba House Project, most elites reasoned that a mosque two blocks from the former World Trade Center site would be “too close.” The implication was that it was not the building of a mosque per se that was problematic, but merely the location chosen for that mosque. The analysis here casts considerable doubt on the notion that a mosque would have been tolerated had it been located farther from Ground Zero, at least as long as the public was being asked to evaluate the mosque in conjunction with a reminder of the September 11 terrorist attacks. Rather, opposition to the mosque is explained largely by partisanship, ideology, and general intolerance toward out-groups; as such, the findings here speak more generally to the growing body of literature demonstrating that many Americans hold negative views toward Muslims and are generally intolerant of that group, particularly in the wake of September 11 (e.g., Kalkan, Layman, and Uslaner Reference Kalkan, Layman and Uslaner2009; Panagopoulos Reference Panagopoulos2006; Schafer and Shaw Reference Schafer and Shaw2009; Sides and Gross Reference Sides and Gross2007).
Did the debate about the location of the project affect the extent to which respondents were willing to express intolerant attitudes on this issue? A survey conducted in August 2011 by Opinion Research Corporation found that 57% of respondents either opposed or strongly opposed the building of a mosque near the site of the World Trade Center.Footnote 20 However, in the same survey, 76% said that they would either “strongly favor” or “favor” the building of a mosque in their own community. These contrasting findings are perhaps best understood as a product of the interaction between social desirability bias and elite discourse. Social desirability bias arises when survey respondents give responses to sensitive questions that they think are more socially acceptable, even if those responses do not accurately reflect their true attitudes. When asked whether they would oppose the building of a mosque, most Americans understand that given the centrality of religious freedom in the United States, the socially acceptable response is not to express opposition. Thus, when asked about a mosque in their own community, a large majority of Americans claim to be supportive. However, the widely publicized discourse over the Cordoba House Project gave citizens a socially acceptable reason to oppose the project by highlighting concern about the sensitivities of those who lost friends and relatives in the September 11 attacks. The fact that this rationale was voiced by important individuals in both parties further legitimized opposition to the mosque. In short, the findings from this article suggest that proximity to the World Trade Center did not cause most Americans to oppose the mosque's construction, but gave those Americans a socially acceptable excuse to express their general intolerance for the project.
