The Ideological Effects of Framing Threat on Immigration and Civil Liberties

Key policy questions that touch on civil liberties and immigration are: How far should the government go in tapping phones, monitoring email and credit card transactions? Should “racial profiling” be an acceptable security and policy practice? Should tighter restrictions be placed on visas to individuals from certain (e.g., Arab) countries? Muslims? How extensive should security be at airports or other areas in the public sphere?

According to a 2005 survey of registered voters, those claiming illegal immigration to be a problem were equally likely to base their opinion on issues of terrorism and homeland security (31%) as jobs and the economy (Polling Report FOX News/Opinion Dynamics Poll. April 25–26, 2005).

In the US, the Patriot Law of 2001 and the Enhanced Border Security and Visa Entry Reform Act in 2002 notably paved the way for electronic innovations, visa screening, racial and ethnic profiling, acceleration procedures, and unprecedented security checks as well as the formation of a new Office of Homeland Security to coordinate activities with a reorganized INS. The formation of this new Office brought 22 federal agencies under one umbrella, and has been considered the first significant addition to the US government since 1947, when Harry Truman merged the various branches of the US Armed Forces into the Department of Defence to better coordinate the nation’s defence against military threats (US Department of Homeland Security: www.​dhs.​gov).

The American literature tends to posit an important relationship between ideological leanings and immigration preferences. Based on national poll data, Espenshade et al. (1996) found that immigration attitudes were strongly divided by ideological underpinnings with conservatives more likely to endorse restrictionist policies. These findings were corroborated in research conducted by Citrin et al. (1997).

While our experiment included a news frame regarding immigration as it pertains to economic threat, it is not specifically addressed within the paper for theoretical reasons. We are here concerned primarily with investigating the differences between threats which harm higher order needs (i.e., national identity) versus those which threaten more basic needs (i.e., security). For the sake of transparency however we have included the data for the economic manipulations within all of the tables.

Though the frames are slightly different in nature, they were intended to reflect the political discourse at the time of the survey. Consequently, while the frames may arguably seem to differ by threat intensity they do so in a way that has real world relevance. Future research involving media content analysis is warranted to systematically gauge the external validity of the experimental manipulation however.

The experiment was run over the course of a 6-month (from December 2006 to May 2007). Recognizing the limits of the experimental design, including possible bias of frame (national security) at the study’s specific time, sample (undergraduate students), and place (New York), the findings are meant to be suggestive of general trends, with need for further national experimental data to corroborate the study.

While the literature on inter-group conflict and race/ethnicity in American politics has been suggestive of demographic differences (Welch and Sigelman 2000), there is still a dearth of work related to diverse migrant or minority groups (e.g., Latinos or Asians), beyond a focus on black/white relations (Fosset and Kiecolt 1989; Glaser 1994). Yet, the import of race/ethnicity of immigrant populations is also believed to interact with attitudes towards security, as demonstrated in WWII with Japanese-Americans, and more recently by the link between 9/ll and Arab populations. While recognizing that among demographic data, inter-group variables may significantly affect attitudinal variations toward immigrants (see Quillian 1995; Tolbert and Hero 1996), the primary focus of the investigation here however is on ideological effects. Thus, given our experimental method, we assume that any group differences between subjects are randomly dispersed.

These questions comprised the following: do you think that the number of immigrants from foreign countries who are permitted to come to the United States to live should be (Increased a lot, Increased a little, Left the same as it is now, Decreased a little, Decreased a lot)? Would you vote for a candidate who would be strict on immigration (Yes, No)? Immigrants who do not have legal documents should be sent back to their countries (Strongly agree, Agree somewhat, Disagree somewhat, Disagree strongly)? The questions were combined to create a scale whose overall score was .528 (on a 0–1 scale: 0 represented support for greater leniency and 1 represented greater restrictions).

This measure constituted the following questions: How strongly do you favor or oppose requiring that everyone in the United States carry a national identification card that would have detailed information about each person (strongly favor, somewhat favor, somewhat oppose, strongly oppose)? Should foreigners who dislike our government and criticize it be allowed to study here? How strongly do you favor or oppose allowing the U.S. government to regularly monitor the personal telephone calls and e-mails of ordinary Americans? A group that wants to buy advertising space in a newspaper to advocate war against another country- Should be turned down by the newspaper, Should have as much right to buy advertising space as a group that favors world peace? How strongly do you favor or oppose allowing the U.S. government to regularly monitor the movement and whereabouts of ordinary Americans? What concerns you more right now: that the government will fail to enact strong, new anti-immigration laws with adequate enforcement, OR that the government will enact and implement new anti-immigration laws, which excessively restrict the average person’s civil liberties? Do you think the US should put some groups, especially those of certain nationalities or religions (Muslims for instance), under special surveillance- Put under surveillance, It’s a mistake to target a special group?

The question read as follows: “How important do you believe this issue of immigration as a threat to (our culture, national security, or the economy) to be?” Subjects in the control group did not receive this question as it pertained directly to the news frame.

The question took the following form: “Do you believe the argument of the House member, that immigration is a threat to (our culture, national security, or the economy), is a weak or a strong one?” The control group did not receive this question as it pertained directly to the news frame.

The difference in levels of support for liberal immigration policies between the various frames is lightly demonstrated (F[3, 399] = 1.98, p = .116). Since we are mainly interested in the effects of the national security frame, as compared to the other frames, however conducting an overall test of significance may not be appropriate here. Thus, instead of conducting an omnibus test, comparisons between two or more treatment conditions that are components of a larger experimental design should be made.

While ideology appears to play a pivotal role in the intersection of framing and civil liberties we found very little evidence for convergence of opinion. Thus we focus our attention on support for immigration policies regarding our third hypothesis. It is interesting to note however that conservatives were consistently more likely to support greater restrictions across the various frames than were liberals (see Table 3). In terms of the framing effects both liberals and conservatives were more likely to support greater restrictions under national security frames than under national identity frames. The findings are significant at the .05 and .1 level respectively using a t-test. These results substantiate the thesis that security framing increases support for civil liberty restrictions.

The first Eurobarometer suvey on “Racism, Xenophobia and Intolerance” of 11,795 adults reported that, among European publics, the degree of importance attached to the problem of immigration surprisingly varies little with age, sex, or educational level, but becomes distinguishable on the political/ideological level (Commission of the European Communities 1989). According to the findings, the more one identifies with the right of the political spectrum, the more importance one attaches to the “problem” of immigration (and vice versa). Longitudinal data on Members of the European Parliament corroborate these findings. The attitudes of Members of European Parliament in the 2004 new-security climate echoed earlier 1994 attitudes (see Lahav 2004, p. 139), suggesting a greater propensity of conservative partisans (15% difference) to report the immigration issue to be “very important” compared to liberals (Lahav 2009, p. 221).

As measured by the extent to which the respondent agreed/disagreed with the following question “I find myself ‘torn’ between two sides of the issue…Sometimes I see the positive aspects of this issue, but sometimes I more easily see the negative aspects of this issue”.

This may explain why liberals supported liberal immigration policies to a greater extent under the national identity frame as they were more likely to reject this argument (see Table 4).

When attitudes towards immigrants from certain regions were probed, respondents not only reacted much more negatively to Middle Eastern and Turkish groups than to Europeans, Asians or even Africans, more generally, but the different framing effects were only notable among the former groups. For example when asked about immigration preferences for individuals coming from Europe (0 = increased a lot, 1 = decreased a lot) the average respondent score was .469 under the security frame and .47 under the identity frame. In sharp contrast, the average for those hailing from the Middle Eastern and Turkey was .63 and .54 for the security frame, while the identity frame yielded .57 and .51 respectively.

Parties may stress particular issues because they work in their favor, and, in some sense they “own” them (Petrocik 1996). Since many issues will never be emphasized by particular parties, certain areas of resulting space will be out of bounds to some parties and effectively owned by others. According to Budge and Farlie, for example, the British Conservatives could never appear more committed to social reforms than the Labour Party. The best Conservative strategy thus was to divert attention from the need for social reforms to the need for law and order, on which Labour could not credibly propose a firmer line than the traditional party of order (1983, p. 260).