By Fathali Moghaddam and James Breckenridge

Abstract

Although Americans’ views concerning illegal immigration have garnered enormous media and pundit attention, this article argues that policy preferences concerning legal immigrants with diverse racial and ethnic origins deserve the attention of homeland security professionals. Using a representative probability sample of more than 4,000 Americans, the study presented here found a majority preference for an alternative to assimilation and multiculturalism – two policies emphasized traditionally among academics. The publics’ preferred policy ­– omniculturalism – cuts across American sociodemographic differences, yet predicts critical variations in the perceived threat of terrorism, the priority of terrorism, confidence in government, and support for aggressive counter-terrorism measures.

Correspondence regarding this paper should be directed to the first author.

This article presents data suggesting that Americans’ views of policies toward immigrants are pertinent to matters of homeland security. “Homeland” is a concept shaped partly by how people psychologically differentiate “citizen” from “immigrant.” The differentiation of these categories is critical to individuals’ political and social identity. Homeland security scholars are unlikely to be aware, however, of this country’s substantial majority preference for an alternative to the traditional, yet deeply divided, incompatible policies of assimilation and accommodation. Moreover, the publics’ appraisal of the threat of terrorism, the priority they assign to homeland security institutions, their trust and confidence in homeland security organizations, and their support for counter-terrorism measures are linked to their immigration policy preference even after accounting for their race/ethnicity and socioeconomic status. Homeland security professionals would do well to consider the potential implications of these preferences.

Practitioners and researchers in the domain of security have been engaged for several decades in an important debate concerning the relative merits of a “realist” versus a “human security” approach. ¹ The realist approach focuses primarily on military security, and represents the dominant school in the domain of security studies. The human security approach is newer and involves an emphasis on health security, food security, shelter security, and other “humanitarian” concerns that are argued to be a priority for ordinary people in their everyday lives. Although the debate between the realist and human security camps has been constructive, there is a danger that both approaches are being left behind by new challenges created by accelerating globalization. Among the most important of these challenges is rapid and large-scale movement of people around the world bringing about “sudden” intergroup contact. ²

Humans have always been migrating, starting from Africa to reach all the major landmasses by about 10,000 years ago. ³ But until fairly recently, migrations were relatively slow. The human groups in interaction had more time to adapt to one another. In the modern era, using jet planes and rapid trains, large numbers of people can move long distances in a relatively short time. The availability of rapid transportation systems has been coupled with the globalization of the economy, so that a demand for cheaper labor in one part of the world can be met with a speedy supply of cheaper labor from other parts of the world. Consequently, in the last few decades there has been a rapid increase of South Asians in the United Kingdom, North Africans in France, and Turks in Germany, with the result that there are now about twenty million Muslims in the European Union.

Rising intergroup contact in recent decades has created new tensions in the European Union, and these tensions have been further intensified by a series of terrorist attacks. The most well-publicized of these attacks are the March 11, 2004, bomb explosions on trains in Madrid which resulted in close to 200 deaths and over 1,000 serious injuries, and the July 7, 2005, bomb explosions on the London public transportation system, which also resulted in multiple fatalities and serious injuries. An outcome of terrorist attacks has been a re-examination of policies for managing diversity; Europeans have been forced to ask, are we integrating minorities the best way? For example, Andrew Jakubowicz assessed reactions to the London terrorist bombings in this way: “The updraft from the bombings carried a message about the critical importance of working out what ‘multiculturalism’ could continue to mean.” ⁴ This question was brought into sharp focus when the Dutch filmmaker Theo Van Gogh was brutally murdered in Amsterdam by an Islamic fanatic on November 2, 2004. Van Gogh’s “crime” was that he had, in collaboration with the Dutch Muslim feminist Ayaan Hirsi Ali, made a short film, Submission, critical of the treatment of women in Islamic societies. Van Gogh’s murder put the spotlight on the Muslim fanatics in Europe, and forced Europeans to critically re-think their policies for managing diversity. Similarly, the threat of home-grown terrorism in the United States, highlighted by the case of about twenty young Somali-Americans apparently recruited by violent Islamic fanatics, has fueled a debate about the best policies for managing diversity in the United States, as well as the threat of terrorism, trust in government, and related security issues.

Two main policies have been used to manage cultural and linguistic diversity: assimilation, the washing away of intergroup differences, and multiculturalism, the highlighting, strengthening, and celebration of intergroup differences. ⁵ Both these policies are founded on psychological assumptions, some of which are questionable. ⁶ An assumption underlying assimilation policy, for example, is that intergroup differences can be washed away through contact, to eliminate any important basis for group-based divisions. But social identity research using the minimal group paradigm demonstrates that group members can use even trivial criteria as a basis for intergroup differentiation and ingroup favoritism. ⁷ By implication, no matter how similar the members of a society become through assimilation, it will be possible to manufacture dissimilarity, even on seemingly trivial criteria. Some of the key psychological assumptions underlying multiculturalism are also questionable, including the multiculturalism hypothesis, the idea that confidence in one’s own ethnic heritage will lead one to be open and accepting toward the outgroup members. Empirical evidence does not provide solid support for this hypothesis, ⁸ nor do historical examples, such as the Nazis, who arguably showed high confidence in their ingroup heritage, but were not open and accepting toward outgroups (although there is support for some interpretations of multiculturalism, particularly among minorities). ⁹

There is continued debate between supporters of multiculturalism and assimilation, ¹⁰ and some efforts to compare the two policies using empirical evidence. ¹¹ However, given that the psychological assumptions underlying both policies are in important ways flawed, we should also explore alternative policies that are already an implicit part of psychological discussions of intergroup relations. ¹² Muzafer Sherif’s concept of superordinate goals, ¹³ and Gaertner and Dovidio’s Common Group Identity Model both suggest a third alternative policy, whereby groups emphasize commonalities such as identities and goals. ¹⁴ This third alternative is reflected in the policy of omniculturalism, which proposes a two-stage process in the socialization of individuals: during stage one, the focus is on human commonalities; during stage two, intergroup differences and distinctiveness are introduced. ¹⁵ The objective of omniculturalism is to establish a solid basis of commonality between people within the framework of a primary identity, before adding an emphasis on how people also belong to groups that in some respects differ from one another.

The present study examines three research questions. The first concerns the extent to which Americans would support omniculturalism, as compared with multiculturalism and assimilation. The second concerns the support of majority and minority group members for the different policies. Some previous research has demonstrated that African Americans and other minorities show stronger support for multiculturalism, whereas white Americans show stronger support for assimilation policy. ¹⁶ A third set of research questions ­– the central focus of this article ­– concern possible differences in the attitudes of supporters of assimilation, multiculturalism, and omniculturalism, toward homeland security threats, how America should react to such threats, and the extent to which individuals trust authorities to do the right thing.

In summary, terrorist attacks in Western democracies, such as the United States, the United Kingdom, and Spain, have resulted in a re-assessment of multiculturalism and other policies for managing diversity. ¹⁷ Because assimilation has been endorsed to a greater degree by majority groups (primarily of western European descent), and because terrorist attacks are perceived as arising from minority (primarily Middle Eastern) communities, we expected support for assimilation to be associated with greater concern about future terrorist attacks, as well as stronger American reactions to terrorist attacks. Growing concerns about the possibility of “home grown” terrorism may increase the salience of these issues for American security practitioners and researchers, especially in light of current population projections, which suggest that by 2050 whites will represent a minority and one out of five Americans will be an immigrant. ¹⁸

Methods

Participants in this research were a nationally representative probability sample of 4,000 adults age eighteen and older selected randomly from an internet-enabled panel maintained by Knowledge Networks (KN) in November 2008. KN panel members are recruited through a random digit telephone dialing system based on a sample frame covering the entire United States. In contrast to “opt-in” Web surveys, which recruit participants of unknown characteristics via “blind” Internet solicitations, KN panel members are selected on the basis of known, non-zero probabilities. Individuals are not permitted to volunteer or self-select for participation in the KN panel. In addition, individuals who lack either computers or Internet access are provided equipment or access without charge. KN panel-based surveys have demonstrated acceptable concordance with a variety of “benchmark” large-scale surveys. ¹⁹

In the present study, the response rate to invitations to participate was 71 percent. To reduce the effects of potential non-response and non-coverage bias, post-stratification sample weights, ²⁰ incorporating the probability of participant selection based on age, gender, race, and ethnicity benchmarks from the most recent available Census Bureau Current Population Survey and supplements were employed in all statistical analyses using algorithms modified for complex survey designs in the statistical software packages STATA. ²¹

MEASURES

Cultural policy preferences. Participants were grouped into one of three perspectives on cultural differences policies according to participants’ response to the following question:

“Which statement below best fits your view about immigration to the United States: When people come to America,

1. People should set aside their cultural differences and “melt into” the American mainstream;
2. People should maintain and celebrate their distinct group culture
3. People should first recognize and give priority to what they have in common with all other Americans, and then at a second stage celebrate their distinct group culture.”

We label responses 1 thru 3 Assimilation, Multiculturalism, and Omniculturalism, respectively. Participants could also choose not to declare any preference.

Political ideology. Participants identified their favored political ideology as “extremely liberal,” “liberal,” “somewhat liberal,” “moderate or middle of the road,” “slightly conservative,” “conservative,” or “extremely conservative.” In the following analyses, participants were grouped into three categories: liberal (extremely liberal or liberal), conservative (extremely conservative or conservative), or other (all other responses).

Terrorism risk perceptions. Participants rated the probability over the next five years of terrorist attacks using an anchored scale from zero (“totally unlikely to occur”) to 100 (“absolutely certain to occur”) and assessed the probability of acts of terror within the country (risk to nation – “How likely do you feel a terrorist attack is somewhere within the United States?”), as well as attacks directly involving the participant (risk to self – “How likely do you feel that you personally will directly experience an act of terrorism?”). An additional dichotomous indicator variable was included representing participants who reported that they were “very concerned” or “extremely concerned” about terrorism (“How concerned or worried are you about a terrorist attack happening in the area of the country where you live sometime during the next 12 months?”).

Emotional response to the threat of terrorism. Following the instructions, “Please help us to understand how you feel when you think about threats of terrorism using the following scale,” participants completed the Positive and Negative Affect Schedule – Expanded Form, ²² which requires participants to rate sixty emotional adjectives on a five-point scale from one (“slightly or not true of your feelings”) to five (“extremely true of your feelings”). Composite subscales assessing the degree of fear and anger were employed in the present study. These subscales have demonstrated good psychometric properties in other samples and have been significantly correlated with public perceptions about terrorism and support for various counterterrorism policies. ²³

Confidence in government, preparedness, counterterrorism measures, and security priorities. Participants were also asked whether they “agreed,” “strongly agreed,” “disagreed,” or “strongly disagreed” with a series of statements related to terrorism and terrorism policies. To simplify the presentation of results, responses were collapsed into categories indicating either agreement or disagreement. Statements assessed confidence in certain government organizations (i.e., the federal and state governments, the Immigration and Customs Enforcement agency, the Border Patrol, in response to the statement “This organization will do a good job carrying out its role in fighting terrorism”), community terrorism preparedness (“I believe my community is sufficiently prepared for a terrorist attack if it happened here”), and the importance of revenge (“It is important for United States to take revenge on the people and countries responsible for terrorist acts against this country”). In addition, participants were asked whether they agreed that in order to “protect against terrorism” the government should adopt certain measures, including “Engage in racial or ethnic profiling,” “Restrict the rights of non-citizens and foreign visitors,” or “Require all Americans to have a national identification card.” Finally, participants were asked to rank terrorism-versus disaster-related activities as the top “homeland security priority for the United States.”

RESULTS

More than three out of five American adults preferred omniculturalism. ²⁴ Among those who preferred another policy, more favored assimilation over multiculturalism (Table 1). Gender, age, race and ethnicity, education, income, political ideology, and urban residential status distributions within policy preference groups are listed in Table 2.

Though most members of each sociodemographic category preferred omniculturalism, distinct sociodemographic profiles differentiated proponents of assimilation or multiculturalism. Significantly greater proportions of women, adults under age forty-five, members of non-white races or ethnicities, urban residents, or political liberals, characterized multicultralists. Conversely, white non-Hispanics, older adults over age fifty-nine, individuals with annual household incomes from $10,000 to $20,000, and political conservatives were more prevalent among assimilationists. Assimilationists were also more apt to have partial or full high school educations, but were less likely to have pursued or completed college educations.

Average predicted probabilities of a terrorist attack on the nation or against the self, as well as average levels of fear and anger experienced in response to terrorism within each group are shown in Figure 1. Responses for the omnicultural group closely tracked the average national response. Assimilationists reported the most elevated appraisals of the probability of attacks against the nation or self, as well as the greatest degree of anger in response to terrorism. Omniculturalists, however, reported significantly less fear than either assimilationists or multiculturalists, but averaged significantly higher appraisals than multiculturalists of the likelihood of a terrorist attack on the nation.

Participants’ priorities and support for particular responses to the threat of terrorism reflected these divergent views of threat and emotional response (Table 3). Intense worries about terrorism were least common among multiculturalists and most prevalent among assimilationists. Significantly more assimilationists – in contrast to significantly fewer multiculturalists – viewed terrorism as the top homeland security priority, and also asserted the importance of seeking revenge against terrorist actions. Moreover, support for modifying civil liberties to prevent terrorism – racial profiling, restricting the rights of non-citizens, and requiring a national identity card – was most prevalent among assimilationists and once again, least prevalent among multiculturalists. While more assimilationists had confidence in the federal government’s capacity to counter terrorism, more multiculturalists had confidence in the Immigration and Customs Enforcement. Omniculturalists were more likely to view disaster preparedness as the top homeland security priority and to judge their communities as better prepared for crises.

With respect to security priorities and responses, omniculturalism was situated between the extremes of the alternative cultural policy preferences. Moreover, when multivariate procedures were employed to adjust statistically for sociodemographic differences among cultural preference groups, the differences among omniculturalists, multiculturalists, and assimilationists in perceived threat, emotional response, security priority, confidence in government, perceived community preparedness, and support for aggressive responses to terrorism were sustained.

DISCUSSION

The objective of this study was to explore attitudinal support among Americans for the traditional policies of assimilation and multiculturalism, as well as the new policy of omniculturalism. A second research question focused on the support of majority and minority groups for the different policies. Third, we explored the relationship between support for different policies for managing cultural diversity and security issues, specifically related to the threat of terrorist attacks, how America should react to attacks, feelings about the possibility of terrorist attacks, and trust in authorities to do the right thing in response to terrorist attacks.

With respect to support for different cultural diversity policies, omniculturalism represented a clear majority preference across all sociodemographic groups, although there were some sub-group differences: whites, men, and older adults were more prevalent among assimilationists; non-whites, women, and younger adults were more prevalent among multiculturalists. Consequently, any future exploration of the omnicultural perspective must also attend to the generational and diversity differences that underlie dissenting perspectives among a significant portion of the population. That such differences predicted the roughly 4 percent of participants who declined to state a cultural preferences, as well as the 29 percent of those who declined to participate in this survey further, ²⁵ underscores the need for careful scrutiny of the pattern of minority preferences identified in the present study.

Preferences for cultural policies were correlated significantly with terrorism threat perceptions and emotional responses, as well as attitudes towards homeland security priorities, confidence in certain governmental organizations’ capacities to carry out their counterterrorism missions, and willingness to modify civil liberties to prevent terrorism. Although assimilationists did not differ from multiculturalists in reported fear, assimilationists expressed the highest levels of anger, an affective response associated strongly with support for aggressive counterterrorism policies in other studies. ²⁶ Indeed, support for aggressive measures was most common among assimilationists, a group which also judged the likelihood of future attacks as more probable than those who endorsed alternative cultural policies, and least prevalent among multiculturalists, a group which appraised national threats of terrorism as less likely than other groups. In several respects, the attitudes towards homeland security among omniculturalists represented a middle ground between the divergent views of assimilationists and multiculturalists.

Omniculturalism arises in part out of well-researched ideas in the social psychology of intergroup relations. Both the earlier field research of Sherif and the more recent experimental research of Gaertner and Dovidio have demonstrated that the re-categorization of the members of different groups as a single group can reduce the original intergroup biases. ²⁷ The applied benefits of superordinate goals have been demonstrated in culturally and ethnically diverse classrooms. ²⁸ The Common Ingroup Identity Model has taken the further step of carefully exploring potential antecedents, consequences, and mediating processes of re-categorization that results in a superordinate category. ²⁹ However, missing from this picture has been empirical evidence to suggest that a “third alternative” along these lines would be supported among the general population.

This study presented participants a third alternative, omniculturalism, with two steps: First, recognizing what is common to all Americans, second, celebrating distinct group cultures. Endorsement of this alternative policy represents positive feedback for research exploring the path of re-categorization, but it also highlights a need for additional research on developmental questions. In particular, at what age should the education of children emphasize what is common to everyone, and at what age should the focus be on distinct group cultures? Input from developmental science should guide schools and other socialization agents on this question. In future research, more attention also needs to be given to the difference in support shown by majority and minority group members for the three policies for managing diversity. An important limitation to the present study is that perspectives on cultural policy were measured by a single item. Future studies should include multiple measures, as well as, perhaps, comparisons among each pair of alternatives. Our statistical analyses utilized post-stratification weights to adjust for sampling biases. Nevertheless, the sociodemographic factors we found associated with a preference for assimilation or multiculturalism in this study also tended to characterize individuals in the KN panel who declined to participant. Thus, the magnitude of support for omniculturalism ­– albeit, considerable (i.e., 60 percent) – could well have been attenuated if all invited participants had been recruited successfully for the survey.

We believe that the alternative policy of omniculturalism also has potential to both gain support from diverse populations internationally and serve as an effective policy at the international level. This is because omniculturalism presents opportunities for groups to both find common ground in shared human characteristics and establish their own special (and perhaps unique) characteristics at a secondary level. A challenge in future research is to further explore these possibilities internationally.

Support for different policies for managing diversity was systematically associated with different patterns of attitudes toward security issues. Support for assimilation was associated with greater concern and anger about the possibility of a terrorist attack, as well as support for stronger reactions in the case of an attack. This included greater willingness to seek revenge, to carry out racial and ethnic profiling, and to restrict the civil liberties of foreigners in case of a terrorist attack. In contrast, supporters of multiculturalism policy downplayed the possibility of a terrorist attack and were least likely to seek revenge and agree to racial and ethnic profiling, as well as to impose restrictions on the civil liberties of foreigners as a protection against terrorism. We believe this pattern of results is explained in part by the fact that support for multiculturalism was most prevalent among minority groups, whereas support for assimilation was most prevalent among majority groups. At the same time, terrorist attacks have been seen as emanating from Islamic communities (within and outside Western societies), and the target of such attacks have often been major urban centers in the West, such as New York, London, and Madrid. Thus, majority groups support assimilation of minorities into mainstream society, and perceive terrorism (emanating from minority communities) as a greater threat and something to be angry about and avenged.

The pattern of distrust toward authorities shown by supporters of assimilation and multiculturalism was also different. Whereas supporters of assimilation expressed greater confidence in the counterterrorism capacity of the federal government, supporters of multiculturalism expressed greater trust and confidence in the present capabilities of Immigration and Customs Enforcement. These differences might be attributed to controversy regarding illegal immigration. Multiculturalists’ confidence in the status quo perhaps reflects a reluctance to support strengthening immigration controls; conversely, assimilationists’ lack of confidence might reflect greater willingness to strengthen immigration controls.

The finding that support for different policies for managing cultural diversity was systematically related to attitudinal differences toward security issues reflects back in important ways on the traditional debate between the two main sides in debates about security, suggesting an interactive link between factors identified by realists and human security advocates. On the one hand, the large-scale movement of people and sudden contact between human groups can result in “host” majority groups feeling threatened, desiring the minority to assimilate, and wanting revenge for terrorist attacks. ³⁰ Furthermore, in this context the majority seems to have less confidence in federal and immigration authorities to do the right thing. These trends are no doubt to some extent associated with the majority groups perceiving the influx of “aggressive” minorities as increased competition for scarce resources. However, more than material resources are involved: minority groups support multiculturalism and seem to want to maintain their distinct identities. They are less fearful about terrorist attacks and do not support America “avenging” such attacks. Clearly, both material factors, identified by realists, and “soft” factors such as identity, identified by advocates of human security, are involved in these intergroup processes.

Since the 1990s there has been increased focus on the approximately 12-15 million illegal immigrants believed to be in the United States. For many, illegal immigrants represent a “threat” that requires an immediate solution. However, even if the “problem” of illegal immigration is solved, the far greater challenge of managing an increasingly diverse population of United States citizens looms ahead of us. In the long term, even if all 12-15 million illegal immigrants either become legal or leave the country (an unlikely event), effective policies are still urgently required for managing inter-group relations among the enormously diverse population of over 300 million Americans, which today includes 37 million legal first-generation immigrants. Such policies must receive greater attention from authorities, researchers, and others concerned with homeland security. The findings of this study highlight the value of exploring alternative policies for managing diversity, as well as critically re-thinking links between both alternative and traditional policies and homeland security.

Fathali M. Moghaddam is professor, Department of Psychology, and director of the Conflict Resolution Program, Department of Government, Georgetown University. His most recent book is The New Global Insecurity (2010); more details about his research and publications can be found at his website: www.fathalimoghaddam.com.

James N. Breckenridge, PhD, is professor of psychology and co-director of the PGSP–Stanford Consortium at the Palo Alto University. He is also associate director of the Center for Interdisciplinary Policy, Education and Research on Terrorism (CIPERT) and a senior fellow at the Center for Homeland Security and Defense (CHDS).

Correspondence regarding this paper should be directed to the first author.

This research was supported in part by funding from the Department of Homeland Security through the Center for Homeland Defense and Security at the Naval Postgraduate School, Monterey, CA.

---

1. A. Collins, ed., Contemporary Security Studies (Oxford University Press, 2007); M.I. Midlarsky, Handbook of War Studies (Boston: Unwin Hyman, 1989); R. Paris, “Human Security: Paradigm Shift or Hot Air?” International Security 26 (2001):87-102; M. Weissberg, “Conceptualizing Human Security,” Swords and Ploughshares: A Journal of International Affairs XIII (2003): 3-11.↵
2. F.M. Moghaddam, Multiculturalism and Intergroup Relations: Psychological Implications for Democracy in Global Context (Washington, DC: American Psychological Association Press, 2008).↵
3. S. Wells, The Journey of Man: A Genetic Odyssey (Princeton, NJ: Princeton University Press, 2002).↵
4. A. Jakubowicz, “Anglo-multiculturalism: Contradictions in the Politics of Cultural Diversity at risk,” International Journal of Media and Cultural Politics 2 (2006): 255.↵
5. W.E. Lambert and D.M. Taylor, Coping with Cultural and Racial Identity in Urban America (New York: Praeger, 1990).↵
6. Moghaddam, Multiculturalism and Intergroup Relations.↵
7. H. Tajfel, C. Flament, M.G. Billig, and R.F. Bundy, “Social Categorization and Intergroup Relations,” European Journal of Social Psychology 1 (1971): 149-177.↵
8. Lambert and Taylor, Coping with Cultural and Racial Identity.↵
9. M. Verkuyten, “Ethnic Group Identification and Group Evaluation among Minority and Majority Groups: Testing the Multiculturalism Hypothesis,” Journal of Personality and Social Psychology 88 (2005): 121-138.↵
10. B.J. Fowers and B.J. Davidov, “The Virtue of Multiculturalism: Personal Transformation, Character, and Openness to the Other,” American Psychologist 61 (2006): 581- 594.↵
11. J.A. Richeson and R.J. Nussbaum, “The Impact of Multiculturalism versus Color-blindness on Racial Bias,” Journal of Experimental Social Psychology 40 (2004):417-423; C. Wolsko, B. Park, C.M. Judd, and B. Wittenbrink, “Framing Interethnic Ideology: Effects of Multicultural and Color-blind Perspectives on Judgments of Groups and Individuals,” Journal of Personality and Social Psychology 78 (2000): 635-654.↵
12. B. Park and C. M. Judd “Rethinking the Link between Categorization and Prejudice within the Social Cognition Perspective, Personality and Social Psychology Review 9 (2005): 108-130.↵
13. M. Sherif, Groups in Harmony and Tension: An Integration of Studies on Intergroup Relations (New York: Octagon Books, 1973).↵
14. S.L. Gaertner and J.F. Dovidio, Reducing Intergroup Bias: The Common Ingroup Identity Model (Philadelphia, PA: Psychology Press, 2000).↵
15. F.M. Moghaddam, “Omniculturalism: Policy solutions to Fundamentalism in the Era of Fractured Globalization,” Culture & Psychology 15 (2009): 337-347.↵
16. J.D. Vorauer, A. Gagnon, and S.J. Sasaki, “Salient Intergroup Ideology and Intergroup Interaction,” Psychological Science 20, No. 1, (2009): 444-446; Wolsko et al., “Framing Interethnic Ideology.”↵
17. Jakubowicz, “Anglo-multiculturalism.”↵
18. J.S. Passel and D.V. Cohn, Pew Social and Demographic Trends: U.S. Populations Projections: 2005-2050, http://pewsocialtrends.org/pubs/703/population-projection-united-states.↵
19. L.C. Baker, M.K. Bundorf, S. Singer, and T.H. Wagner, Validity of the Survey of Health and the Internet and Knowledge Network’s Panel and Sampling (Palo Alto, CA: Stanford University, 2003), http://www.knowledgenetworks.com/ganp/reviewer-info.html; J.M. Dennis and R. Li, “More Honest Answers to Web Surveys? A study of Data Collection Mode Effects,” Journal of Online Research (October 2007):1-15; T. Heeren, E.M. Edwards, J.M. Dennis, S. Rodkin, and R.W. Hinson, “A Comparison of Results from an Alcohol Survey of a Prerecruited Internet Panel and the National Epidemiologic Survey on Alcohol and Related Conditions,” Alcoholism: Clinical and Experimental Research 32 (2008): 222-229.↵
20. In contrast to “opt in” Internet-based surveys, in which only the reported demographics of participants who choose to volunteer for the survey are available, complete population demographics for the KN Panel are known prior to survey recruitment. Consequently, a unique advantage of sampling from a pre-recruited web-enabled panel is that the sociodemographic characteristics of panel members who declined the invitation to participate can be unambiguously described. In this study, people who declined to participate were more likely to be female, under age thirty, black or Hispanic, and have a high school or less education. Statistical analyses that fail to account for response rate differences among such subgroups of participants can bias estimates of effects and yield imprecise and misleading standard errors and confidence intervals. Sampling weights are typically employed to reduce bias of this kind. Details regarding the Knowledge Networks panel design and post-stratification sample weighting are available on-line at http://www.knowledgenetworks.com/ganp/docs. Briefly, an iterative process is used to create weights that are inversely proportional to the probability of selecting each subject, i.e., the proportion of people in the population belonging to each “cell” or cross-classification by age, gender, race/ethnicity, education, income, and geographic region groups. Participants in over-represented cells are weighted less; participants in under-represented cells are weighted more. Iteration is continued until the distribution of weighted data converges on the most recently available U.S. Census distributions for each cell. Sampling weights are employed in subsequent statistical analyses to adjust for response rate and coverage biases and to strengthen the representativeness of results.↵
21. Bureau of Labor Statistics and U.S. Census Bureau, Current Population Survey, (Washington, DC: U.S. Census Bureau, 2008), http://www.census.gov/cps/; StataCorp, Stata Statistical Software: Release 10 (College Station, TX: StataCorporation, 2007).↵
22. D. Watson and L.A. Clark, The PANAS-X. Manual for the Positive and Negative Affect Schedule – Expanded Form (Iowa City, IA: University of Iowa, 1994).↵
23. J.N. Breckenridge and P.G. Zimbardo, “The Political Psychology of Terrorism Five Years after September 11,” paper presented at the International Society of Political Psychology (2007); Breckenridge and Zimbardo, “The Psychology of Political Violence: Implications for Constructive Public Policy?” paper presented at the American Psychological Association (2007).↵
24. Some participants (3.8 percent) declined to state a cultural policy preference. A logistic regression of response status (no stated preference versus any stated preference) on the sociodemographic variables listed in Table 2 was statistically significant (F(17, 4000) = 2.57, p < .001), indicating that non-response could not be construed as randomly missing data. Participants who had received some college education were twice as likely as those with less than a high school education to endorse one of the three cultural policy perspectives (AOR (adjusted odds ratio) = .484, p < .01). Participants with annual incomes between $60,000 and $174,000 were from three to five times more likely to respond than participants in the lowest income category (AOR= .319, p <.001 and AOR = .179, p < .001, for incomes $60,000-99,000 and $100,000-174,000, respectively). The remaining analyses in this paper are confined to participants who chose one of the three perspectives on cultural differences, but include all demographic indicators as covariates throughout. A multinomial logistic regression of declared cultural policy alternatives (assimilation, multiculturalism, omniculturalism) on all predictors (gender, age, race and ethnicity, education, income, political ideology, and urban/rural status) was also statistically significant (F(40,3840) = 4.62, p < .0001).↵
25. See previous note.↵
26. See note 23, Breckenridge and Zimbardo.↵
27. Sherif, Groups in Harmony and Tension; S.L. Gaertner and J.F. Dovidio, “Understanding and Addressing Contemporary Racism: From Aversive Racism to the Common Ingroup Identity Model,” Journal of Social Issues 61 (2005): 615-639.↵
28. E. Aronson, C. Stephan, J. Sikes, N. Blaney, and M. Snapp, The Jigsaw Classroom (Beverly Hills, CA.: Sage, 1978).↵
29. Gaertner and Dovidio, “Understanding and Addressing Contemporary Racism.”↵
30. F.M. Moghaddam, “Catastrophic Evolution, Culture, and Diversity Management,” Culture & Psychology 12 (2006):7415-434.↵

---

This article was originally published at the URLs https://www.hsaj.org/?article=6.3.7 and https://www.hsaj.org/?fullarticle=6.3.7.

Copyright © 2010 by the author(s). Homeland Security Affairs is an academic journal available free of charge to individuals and institutions. Because the purpose of this publication is the widest possible dissemination of knowledge, copies of this journal and the articles contained herein may be printed or downloaded and redistributed for personal, research or educational purposes free of charge and without permission. Any commercial use of Homeland Security Affairs or the articles published herein is expressly prohibited without the written consent of the copyright holder. The copyright of all articles published in Homeland Security Affairs rests with the author(s) of the article. Homeland Security Affairs is the online journal of the Naval Postgraduate School Center for Homeland Defense and Security (CHDS). https://www.hsaj.org