Does Legitimacy Matter?

Abstract

Legitimacy is conceptualized as subjective individual attitudes and expectations about formal institutional authority and is often thought of as a reservoir of trust or goodwill that formal governing authorities draw on to secure acceptance and compliance with the law. Recent public opinion surveys in predominantly Muslim countries report declining support for U.S. government and policy, as well as increasing support for Muslim-based groups that attack the United States. Based on prior research within the United States showing that perceptions of legitimacy are related to both acceptance and compliance with the law, we examine whether perceptions about the legitimacy of the U.S. government may also be related to support for anti-American transnational terrorist attacks. Using data from more than 3,600 face-to-face interviews with respondents from three Muslim countries, we examine the effects of support for the American government, people, and culture on support for Muslim-based groups that attack Americans. In addition, we examine the effects of perceived domestic institutional legitimacy on support for Muslim-based groups that attack Americans. Our results indicate that individuals who have more favorable attitudes toward American citizens and culture are less likely to support attacks against Americans by Muslim-based groups. We also find that perceived legitimacy in one’s own political institutions, including government, police, and the criminal justice system, is associated with lower levels of support for groups that attack Americans. We discuss the implications of the results for research and policy.

Keywords

legitimacy support for terrorism Muslim countries anti-U.S. attitudes survey data

Recent survey results from respondents living in predominantly Muslim countries have shown increasingly negative perceptions of the U.S. government as well as increasing support for Muslim-based groups that attack the United States (Gentzkow & Shapiro, 2004; McCauley, Leuprecht, Hataley, Winn, & Biswas, 2011; McCauley & Scheckter, 2008; McCauley & Stellar, 2009; Orehek et al., 2010). Several scholars have suggested that perceptions of legitimacy in the U.S. government and negative attitudes toward American culture more generally are related to support for anti-American terrorism among Muslims living in other countries (Bernard, 2004; Garfinkle, 2004; Tessler & Robbins, 2007). In fact, researchers have long noted the importance of public opinion and support for terrorist groups and terrorism attacks (Bueno de Mesquita, 2007; Crenshaw, 1983; Hoffman, 2003; Paz, 2003). In spite of long-standing interest in legitimacy for increasing compliance with the law as well as the recent evidence of growing negative attitudes toward U.S. culture among Muslim communities, there has been little empirical research on these issues. The small body of systematic quantitative research that exists provides mixed results. For example, a recent study using a sample of respondents from Algeria and Jordan found that although perceptions of legitimacy toward U.S. foreign policy were related to support for terrorism, attitudes toward U.S. culture were not (Tessler & Robbins, 2007). Moreover, results from another study using several predominantly Muslim samples found that negative attitudes toward the United States and U.S. freedom of speech laws were associated with support for terrorism (Bueno de Mesquita, 2007).

The current study builds on prior research by examining the relationship between support for violence toward the United States and perceptions of legitimacy in U.S. government and attitudes toward U.S. people and culture among a sample of respondents from three predominantly Muslim countries. We begin with a review of the extant literature examining the effects of legitimacy on acceptance of, and compliance with, the law. Prior research (mostly on U.S. respondents) has indicated that perceptions of legitimacy in government and other formal institutions of control (e.g., police, courts) can increase support and compliance with the law (see Tyler, 1990, for review). We also review recent studies examining the effects of perceived legitimacy of U.S. institutions and culture on support for terrorism directed at the United States, in particular, terrorism attributed to groups and individuals from predominantly Muslim countries (Bueno de Mesquita, 2007; Tessler & Robbins, 2007).

Using data collected by the National Consortium for the Study of Terrorism and Responses to Terrorism (START) between 2006 and 2007, we examine whether citizen perceptions of the legitimacy of the U.S. government, people, and culture predict citizen support for Muslim-based groups that attack Americans. Based on prior research, we hypothesize that individuals in other countries who attribute lower support to the U.S. government will be more likely to support groups from the Muslim world that attack Americans. Similarly, we also predict that stronger support for American people and culture will be related to less support for Muslim-based groups that attack the United States. Tessler and Robbins (2007) argued that support for violence toward the United States among Muslims from other countries may be related to their perceptions of legitimacy in domestic institutions in their own countries. We therefore also examine the effects of the legitimacy of domestic government, police, and criminal justice institutions on support for Muslim-based groups that attack Americans. We conclude with a discussion of the implications of our study for future research, theory, and policy.

Legitimacy and Compliance With the Law

Legitimacy is often conceptualized as subjective individual attitudes and expectations about formal institutional authority (Easton, 1965; Fraser, 1974; Nieburg, 1969; Thibaut & Walker, 1975; Tyler, 1990; Weber, 1947, 1968).¹ Conceived in this way, legitimacy can be thought of as a “reservoir of favorable attitudes or good will that helps members to accept or tolerate outputs to which they are opposed or the effects of which they see as damaging to their wants” (Easton, 1965, p. 278). Thus, legitimacy generally leads to increased compliance with the law (Tyler, 1990). The vast majority of research examining the effects of legitimacy has focused on perceptions of legitimacy in political institutions (Easton, 1965; Easton & Dennis, 1969; Gibson, 1991; Gibson & Caldeira, 1995; Tyler, Casper, & Fisher, 1989). However, growing interest in the legitimacy of social institutions of formal control more generally, such as the police or courts, has led to a body of literature in criminology that links perceptions of legitimacy and procedural justice in legal authorities to acceptance and compliance with laws and legal institutions (Mastrofski, Snipes, & Supina, 1996; Paternoster, Brame, Bachman, & Sherman, 1997; Sampson & Bartusch, 1998; Sherman, 1993; Sunshine & Tyler, 2003; Tyler, 1990; Tyler & Wakslak, 2004).

Most of the previous research linking perceived legitimacy to criminal attitudes or behavior has focused on traditional street crimes, and many of these studies (Paternoster et al., 1997; Tyler, 1990) have examined individual’s perceptions of their own government and support for those corresponding government laws—thus, solely focusing on within-country relationships. Evidence within the United States indicates that legitimacy of U.S. institutions is related to greater acceptance of U.S. criminal law and lower criminal offending and recidivism (Paternoster et al., 1997; Tyler & Huo, 2002). In one of the more comprehensive studies of perceived legitimacy on compliance with the law, Tyler (1990) examined the effects of perceived legitimacy on self-reported offending using a variety of legitimacy indicators such as an obligation to comply measure, a normative support measure, and a combined measure of perceived obligation and normative support. Results from his analysis shows that support for the police significantly influenced compliance with the law, controlling for individual demographic traits, and other factors such as personal morality and deterrence. Similarly, several studies have found that positive normative orientations toward the procedures of legal authorities, often referred to as procedural justice, are also related to increased positive perceptions of legitimacy and reduced criminal offending (Jonathan-Zamir & Weisburd, 2010; Paternoster et al., 1997; Tyler, 1990; Tyler & Rasinski, 1991). For example, using a sample of known criminal offenders from the United States, Paternoster and colleagues (1997) found that when perceptions of procedural justice are more favorable, there are significant declines in individual rates of recidivism, even after controlling for prior criminal records.

There is also evidence that perceptions of the legitimacy of government are related to political violence. For example, Gurr (1970) linked legitimacy to political violence within countries by arguing that as legitimacy of political regimes declines, their citizens are more likely to rebel. According to this legitimacy-stability hypothesis, governments that lose legitimacy become more vulnerable to dissent and protest (Friedrichs, 1980; Gurr, 1970; Habermas, 1975; O’Connor, 1973). Similarly, LaFree (1998) argued that declining levels of perceived legitimacy in U.S. government institutions led to rising crime rates during the 1960s and 1970s. Using data from the United States, Useem and Useem (1979) reported that as confidence in the government increases, support for public protest declines. Although the research linking perceptions of legitimacy in government to political disobedience and violence has relevance for the current study, much of the literature is based on evaluating respondent’s confidence in their own government and the consequences on attitudes or behaviors. A unique contribution of this research is that we instead focus on how perceptions of the legitimacy of formal governing authorities of one country are related to support for violence toward citizens of another country.

Terrorism, Perceived Legitimacy of the U.S. Government, and Attitudes Toward American People and Culture

Scholars have also examined the connections between perceived legitimacy in the United States and support for transnational terrorism or political violence. Indeed, there is a substantial amount of research that shows that campaigns of terrorism may be conceptualized as battles over perceived legitimacy (Betts, 2002; Crenshaw, 1983, 2000; Kydd & Walter, 2006; McCauley, 2006; Sharp, 1973). Thus, Crenshaw (1983) argued that “the power of terrorism is through political legitimacy, winning acceptance in the eyes of a significant population and discrediting the government’s legitimacy” (p. 25). Similarly, Betts (2002) claimed that the contest between terrorists and counter terrorists is in effect “tripartite” involving not only insurgents and counterinsurgents but also ordinary citizens caught in the middle of these opposing viewpoints. Using the case of Islamic radicalism and transnational terrorism, Betts (2002) states that “the yet un-mobilized Muslim elites and masses of the Third World—those who were not already actively committed either to supporting Islamist radicalism or to combating it—are the target population in the middle” (p. 28). This reasoning suggests that perceptions of U.S. legitimacy influence popular support and opinion for attacks against the United States.

Using survey data from Algeria and Jordan, Tessler and Robbins (2007) showed that negative attitudes toward U.S. foreign policy are related to increased support for attacks against Americans. In the Algerian sample, the researchers examined approval for religious extremists hijacking civilian airliners in the September 11 attacks on the United States, whereas the Jordanian respondents were asked to report their level of support for a jihad against the United States and their level of trust in al Qaeda and Osama Bin Laden. Approximately 45% of the Algerian sample reported strong disapproval for the September 11 attacks, whereas 23% reported moderate to strong approval. In the Jordanian sample, approximately 30% reported moderate to strong opposition of the “jihad” against the United States, whereas 56% reported strong support.

These results indicate that those respondents who reported lower favorability of U.S. foreign policy and those that reported higher levels of agreement with the belief that the United States violates the rights of citizens in other countries were more likely to report greater support for attacks against the United States. Tessler and Robbins (2007) also found strong connections between the perceived legitimacy of domestic institutions and support for attacks against Americans. Legitimacy in domestic institutions was measured in a similar fashion across countries; however, the focus on specific institutions varied. Algerian respondents were asked to report their confidence in the government, Parliament, military, and police, whereas Jordanian respondents were asked to report on their level of confidence in the government, the legal system, and the police. For both samples, when the perceived legitimacy of domestic institutions was higher, there was less support for transnational terrorism. Most prior theorizing on the effects of legitimacy on compliance has focused on direct linkages between perceptions of legitimacy of a formal ruling institution and their corresponding laws. The findings for domestic institutions suggest that legitimacy may have broader ability to produce global effects, increasing compliance for corresponding laws as well as laws of other authorities more generally.

Although studied far less frequently, there is reason to suspect that perceptions of the U.S. government may be distinct from perceptions of the American people or culture. Recent research indicates that transnational terrorist attacks on the United States perpetrated by Muslim-based groups are fueled, in part, by increasing distaste for American culture and fear that American culture is a threat to Islam (Paz, 2003; White, 2004). Perhaps most famously, al Qaeda has repeatedly expressed strong distaste for American culture and has interpreted it as representing a threat to the Islamic way of life (White, 2004). Several studies indicate that the fear that Islam is under threat is strongly related to support for transnational terrorism (Bueno de Mesquita, 2007; Fair & Shepherd, 2006). The “clash of civilizations” thesis has been offered as an explanation for widespread distaste for Western values and lifestyles among individuals from predominantly Muslim countries (Garfinkle, 2004; Huntington, 1993, 1996; Lewis, 1990; Paz, 2003). This explanation points to inherent conflict between Islamic and Western culture that leads to anti-American ideals and actions.

Most prior research examining the effects of legitimacy has focused on perceptions of government or some other formal ruling authority (e.g., police), as the key impact of legitimacy lies in its ability to garner acceptance and compliance with formal laws more efficiently by relying on voluntary compliance rather than coercion. However, the literature linking legitimacy to transnational terrorism highlights the possibility that support for culture may also be influential for predicting support for those acts that are meant to convey statements about the target. In addition, the literature linking Western culture to perceptions of Islam being under threat suggests that perceptions of U.S. culture may also be related to support for terrorism (Huntington, 1993, 1996; Lewis, 1990).

In summary, there is little research that systematically examines the effects of perceptions of legitimacy in U.S. government and attitudes toward U.S. people and culture on support for terrorism. Tessler and Robbins (2007) found no connections between perceptions of U.S. culture as a threat to respondents’ way of life and support for terrorism. Bueno de Mesquita (2007) examined the correlates of transnational terrorism among respondents from several countries with large Muslim populations and found that the belief that the United States is a threat to Islam was associated with support for terrorism against the United States. However, support for the relationship weakened considerably in the multivariate analysis.

Current Study

The current study builds on extant research on perceptions of legitimacy and support for terrorism and political violence by examining the effects of support for the U.S. government, people, and culture on self-reported support for Muslim-based transnational terrorist attacks against Americans. We examine the independent effects of the perceived legitimacy of the U.S. government and support for American people and culture by including separate measures of each concept. We also include measures of perceived legitimacy in domestic institutions to examine whether legitimacy in formal institutions of social control has a global effect on compliance; we expect that individuals who ascribe high levels of legitimacy to ruling authorities of their own country are less likely to support violence or illegal behavior directed at targets abroad. We use survey data from respondents residing in three predominantly Muslim countries between 2006 and 2007 to test the following hypotheses:

Hypothesis 1: Individuals who report higher levels of support for the American government, as compared with those who report lower levels, will report lower levels of support for Muslim-based groups that attack Americans.

Hypothesis 2: Individuals who report higher levels of support for the American people, as compared with those who report lower levels of support, will report lower levels of support for Muslim-based groups that attack Americans.

Hypothesis 3: Individuals who report higher levels of support for American culture, as compared with those who report lower levels, will report lower levels of support for Muslim-based groups that attack Americans.

Hypothesis 4: Individuals who report higher levels of confidence in their country’s government, police, and criminal justice system, as compared with those who report lower levels of confidence, will report lower levels of support for Muslim-based groups that attack Americans.

Method

Survey and Sample

We use data from face-to-face interviews with 3,645 respondents older than 18 years in Egypt, Indonesia, and Morocco.² Local research companies in each country used probability-based sampling methods to select respondents.³ A team of researchers from the START Consortium developed the questionnaire, which was then translated by research agencies in each country. These translations were then back translated by native speakers of Arabic, Urdu, and Indonesian working with local research teams. The researchers also matched interviewer and respondent by gender. Initial versions of the survey were pilot tested using focus groups from each country and once the researchers incorporated feedback from the pilot studies and received final approval from government agencies in each country, data were collected between December 2006 and February 2007.⁴

The three countries included in our study are among those with the highest percentage of Muslim citizens in the world but are also extremely diverse in terms of history, politics, and culture. Egypt is the fifth most populous Muslim nation in the world and acquired full sovereignty with the overthrow of the British-backed monarchy in 1952 (Central Intelligence Agency [CIA], 2011). Its population is 90% Muslim and it also has Coptic (9%) and Christian (1%) minorities (Pew Center, 2011). Morocco is a constitutional monarchy that gained independence from France in 1956 after a protracted struggle (CIA, 2011). Its population is overwhelmingly Muslim (98.7%) with tiny minorities of Christians (1.1%) and Jews (0.2%; Pew Center, 2011). Indonesia is the largest Muslim nation in the world and the fourth most populous nation on the planet (CIA, 2011). It obtained independence from the Netherlands in 1945. Its population is 86.1% Muslim, 5.7% Protestant, 3% Catholic, 1.8% Hindu, and more than 3% a mixture of other religious groups (Pew Center, 2011). The great majority of Muslims in all three countries are Sunni.

Support for Attacks Against the United States

To measure support for attacks against the United States, we create a composite measure consisting of three variables. First, respondents were asked to think of groups in the Muslim world that attack the United States, and report if they (0) disapprove of all groups, (1) approve of some and disapprove of others, or (2) approve of all groups. Second, respondents were asked, “How do you feel about al Qaeda?” and were given the following response categories: (0) I oppose attacks against the United States and do not share its attitudes, (1) I oppose attacks but share its attitudes, or (2) I support attacks against the United States and share its attitudes. Finally, we also incorporate a measure that asks respondents on a scale from 0 = lowest to 10 = highest to rate their feelings of support toward groups from the Muslim world that attack Americans.

Legitimacy/Support Variables

We create measures of the legitimacy of the U.S. government and support for American people and culture and also a measure of the legitimacy of domestic institutions. Prior research (see Fraser, 1974; Tyler, 1990, 2003, for reviews) has operationalized legitimacy through measures of the perceived obligation to obey the law and normative-based measures of support for authority. Similarly, we use a normative-based measure of support for the U.S. government. Respondents were asked to report their attitudes toward the current U.S. government using a 4-point scale ranging from 1 = very unfavorable to 4 = very favorable. Because many scholars (Huntington, 1993; Lewis, 1990; Paz, 2003) have suggested that key drivers of anti-U.S. terrorism attacks are attitudes about Western lifestyles, we also include measures of support for American people and culture, using the same 4-point scale as used to measure support for government.

Our measure of legitimacy in domestic institutions is a composite scale consisting of three questions tapping legitimacy in formal domestic institutions of authority. Respondents were asked to report their level of confidence in their national government, their justice system, and the local police using a 4-point scale ranging from 1 = none at all to a 4 = great deal. All three items are correlated with each other and have a Cronbach’s alpha of .8. All the key independent variables have less than 10% missing data.

Demographic Variables and Country Controls

Several recent studies indicate that demographic variables such as age, gender, and income may be significantly related to support for terrorism among respondents in predominantly Muslim countries (Bueno de Mesquita, 2007; Fair & Shepherd, 2006; Shafiq & Sinno, 2010; Tessler & Robbins, 2007). Age is measured using a five-category variable ranging from 18 to 20 to more than 50 years of age. We measure marital status using a binary variable comparing married with all nonmarried respondents, including divorced, widowed, and single. Employment is measured as being employed full-time or self-employed at the time of the interview. Shafiq and Sinno (2010) examined the support for suicide bombings that target U.S. citizens or Westerners in Iraq and found no effect for income or education on support. However, the authors found that compared with poorer respondents, wealthier respondents in Morocco were more likely to report support for anti-U.S. terrorism. To measure income and education, we use a variable that asked respondents to self-identify their income and educational status into three categories labeled low, medium, and high. We also include dummy variables for country location.

Attitudinal and Behavioral Variables

To control for respondents’ views of terrorism within their own countries, we include a variable measuring their perception of terrorism. Responses were on a 4-point scale that ranged from 1 = not a problem to 4 = very big problem. We also include a variable that captures perceived harm as a result of the United States. Respondents were asked to report (yes/no) if their interests have ever been harmed by U.S. actions. Exposure to international news may also influence attitudes toward U.S. culture, as well as support for groups that attack the United States. Presumably, more exposure to international news through the global media could worsen attitudes toward the United States and increase support for anti-U.S. attacks. Similarly, international news stories may be more salient to certain respondents. Thus, we create a composite measure that accounts for how many days in the average week respondents were exposed to international issues through print, television media and through discussions with family, friends, or acquaintances (0-7). All these items are correlated with each other and have a Cronbach’s alpha of .6.

Prior research indicates that perceptions of the United States as a threat to Islam are an important correlate of support for terrorist acts directed toward Americans (Bueno de Mesquita, 2007). We include a variable to capture such perceptions, as well as commitment to religious identity and religious affiliation. We include two binary variables indicating affiliation with Islamic faith and self-reported primary identification as a Muslim, as compared with identification with country, ethnic group or identification as an individual.⁵ We also include a variable that asked respondents whether they believed the goal of the United States is to weaken and divide the Islamic world, with responses measured on a 4-point scale ranging from 1 = definitely not a goal to 4 = definitely a goal.

Analytical Strategy

Our primary outcome variable is a composite measure of support for Muslim-based groups that attack the United States, which consists of the three items described above. Of the 3,645 respondents who participated in the study, 2,965 respondents (81%) had valid data for the composite variable, indicating a nontrivial amount of missing data that may affect the validity of the results.⁶ Missing responses on the outcome variable are driven especially by the question tapping respondents’ views of al Qaeda: Approximately 30% of responses were missing. Approximately 25% of respondents had missing data for the question tapping support for all, some, or no groups that attack the United States, whereas only 12% had missing data for reports of overall support for groups that attack the United States. We use listwise deletion in all analyses to reduce any potential bias due to missing data (Allison, 1999). Although the al Qaeda question is conceptually and empirically related to the other two items of support for terrorism, there may be distinct perceptions of al Qaeda. To examine the robustness of our results to both conceptualization and measurement decisions, we conduct a number of analyses examining the effects of legitimacy on support for terrorism.

We use a composite measure consisting of the three questions tapping support for Muslim-based groups that attack the United States (n = 2,021). Because the three variables gauging support for terrorism are not measured on the same scale, we convert the items to z scores and create a composite measure of the average level of self-reported support for Muslim-based groups that attack Americans.⁷ All items are significantly correlated, and Cronbach’s alpha for the pooled sample was sufficient (α = .7) with similar estimates when examining the reliability of scale items within countries (α = .7).

We then examine a composite measure that excludes the al Qaeda question from the measure because it likely represents an especially extreme Muslim-based group and thus does not reflect perceptions toward other more moderate Muslim-based groups (n = 1,905). In addition, of the outcome measures, admitting support for al Qaeda may elicit the most social desirability bias from respondents. Cronbach’s alpha for the two-item scale excluding the al Qaeda measure is .6. Finally, we examine the al Qaeda measure by itself as a binary variable reflecting no support for positive attitudes toward al Qaeda or attacks by al Qaeda versus support for either attacks or positive attitudes.⁸

As with most survey data, these data are inherently nested because they contain responses from individuals who are located within countries. This nesting potentially violates ordinary least squares (OLS) assumptions of independence and homoscedasticity (Raudenbush & Bryk, 2002). Merely incorporating dummy variables to represent country-level differences between individuals ignores the nested nature of the data and may result in inefficient standard errors and inaccurate tests of significance. Several approaches exist to handle nested data, including the use of robust standard errors and multilevel or mixed-level modeling. However, none of these procedures are ideal here given the very small number of Level 2 clusters (i.e., three countries). Because our main focus is on accounting for country membership and correcting for the violation of independence, we incorporate a survey weight variable to adjust for sampling design of countries in our OLS analysis of both composite measures of support for Muslim-based groups that attack the United States.⁹ For our analysis of support for al Qaeda, we incorporate the survey weight and use logistic regression.¹⁰

Results

Descriptive Statistics

In Table 1, we present descriptive statistics for the pooled sample of respondents. According to Table 1, almost half of the sample is below the age of 29 (43%) and the average age is between 23 and 29 years of age. The sample includes more men than women (55% vs. 45%) and more married than unmarried or widowed respondents (60% vs. 40%) participated in the survey. The majority of respondents report having low (35.0%) to medium (40.2%) education; similar proportions are reported for income, with 32.9% and 44.0% reporting low to medium income. Nearly three fifths (58.7%) of the sample reported being employed or self-employed full-time, the remaining respondents reported part-time employment, retirement, or unemployment. A total of 41% (1,504) of the sample was interviewed in Egypt, followed by 31% (1,141) of respondents from Indonesia and 27% (1,000) from Morocco.

Table 1.

Descriptive Statistics, Full Sample (N = 3,645)

Variables	n	%	M	SD	Minimum	Maximum
Support for terrorism
Muslim groups that attack the United States	2,890	79.2	1.3	0.7	1	3
al Qaeda	2,557	70.1	1.8	0.8	1	3
Ratings: Support for attacks against the United States	3,193	87.6	4.4	3.2	0	10
U.S. legitimacy and support variables
Government	3,302	90.6	1.59	0.8	1	4
People	3,264	89.5	2.34	1.0	1	4
Culture	3,266	89.6	2.25	1.1	1	4
Demographic, attitudinal, and behavioral variables
Age
18-22	687	20.3			1	5
23-29	785	23.2
30-39	758	22.4
40-49	572	16.9
50+	580	17.2
Gender
Male	1,994	54.7			1	2
Female	1,651	45.3
Marital status
Married	2,135	40.6			0	1
Not married	1,456	59.5
Employed
Employed	1,493	41.5			0	1
Not employed	2,122	58.7
Education
Low	1,268	35.0
Medium	1,456	40.2			1	3
High	897	24.8
Income
Low	1,161	32.9			1	3
Medium	1,556	44.0
High	817	23.1
Muslim
No	189	5.2
Yes	3,426	94.8
Identify with religion
No	2,301	64.3
Yes	1,276	35.7
Interaction with foreigners			4.8	1.2	1	6
Media exposure			6.8	5.0	0	21
Perception of terrorism in home country			2.9	1.0	1	4
Believe United States wants to weaken Islam			3.4	0.8	1	4
Interests been harmed by United States			1.1	0.5	1	2

Not surprisingly, the vast majority of respondents (94.2%) report affiliation with the Muslim religion; however, only 35.7% report identifying themselves primarily with their religious affiliation. On average, respondents report watching or discussing international news approximately 5 days out of the week. Very few respondents report frequent interaction with foreigners from North America or Europe, with the average response between less than a month and very seldom. However, almost half the respondents report having been harmed by the United States. Similarly, the majority of respondents believe that the purpose of the United States is to weaken Islam; more than 60% of respondents indicate that weakening Islam is a definite goal of the United States. Finally, on average, respondents report terrorism to be a moderate problem within their own countries.

In Figure 1, we present the distribution for self-reported support for Muslim-based groups that attack the United States for the pooled sample (i.e., Egypt, Indonesia, and Morocco). The proportion of respondents that disapprove of all Muslim-based groups that attack the United States (46.2%) was fairly close to the percentage approving of only some groups that attack the United States (42.9%). Only 10.9% of respondents reported approving of all Muslim-based groups that attack the United States. We also see differences in levels of support across the three countries. Compared with respondents in the other two countries, Egyptian respondents were far more supportive of groups that attack U.S. targets with nearly 18% approving of all groups that attack U.S. targets, followed by Indonesians (7.9%) and Moroccans (3.6%).

Figure 1.

Self-reported support for Muslim-based groups that attack the United States, pooled and country-specific samples, 2006 to 2007

Figure 2 shows the distribution for self-reported support for al Qaeda-based views and attacks for the pooled sample. In all, 39% of respondents reported opposing both views and attacks, compared with 23.4% that favored both. As with support for groups that attack the United States, Egyptian respondents showed the strongest support for al Qaeda-based attacks against the United States (29.1%), followed by Indonesians (22.7%) and Moroccans (13.5%). It may be that proximity to the wars in Iraq and Afghanistan partly explains these patterns: Egypt is geographically closest to the wars, followed by Morocco and Indonesia. Our final measure of support for terrorism against the United States taps respondents’ support for Muslim-based groups that attack the United States on a 10-point scale. On average, 4.4% respondents reported less than moderate support, approximately 17.0% reported no support at all, and 9.0% reported complete support.

Figure 2.

Self-reported support for al Qaeda, pooled and country-specific samples, 2006 to 2007

In Figure 3, we show the percentage of total respondents from each country reporting favorable to very favorable perceptions of the U.S. government, people, and culture. Perhaps the most striking finding in Figure 3 is that compared with perceptions of U.S. people and culture, perceptions of the U.S. government are far more negative in all three countries. For the combined sample, U.S. people are judged most favorably (48.0%), followed by culture (40.3%), and U.S. government is viewed least favorably (14.8%). Compared with either Egyptian (37.9% and 37.4%) or Indonesian (17.0% and 42.3%) respondents, Moroccan respondents are far more positive toward both U.S. culture and people (72.4% and 70.2%). Interestingly, while Egyptians are the most negative about the U.S. government (3.8%), Indonesians are the most negative about U.S. culture (17.0%). Attitudes toward American people and culture are also significantly correlated.¹¹

Figure 3.

Self-reported support for American government, people, and culture, pooled and country-specific samples, 2006 to 2007

In Table 2, we present the bivariate correlations between support for attacks against Americans by Muslim groups and the composite measure of support for attacks on American targets. Note that for the legitimacy of the U.S. government measure and the support for American people and culture measures, higher perceived legitimacy and support are significantly associated with less support for attacks against Americans. By contrast, the association between perceived legitimacy of domestic institutions and support for terrorism does not reach significance, and legitimacy of domestic institutions is negatively related to support for U.S. government, people, and culture. Thus, individuals who report higher levels of legitimacy in domestic institutions also report lower levels of legitimacy in U.S. government and less support for American people and culture. Support for American government, people, and culture is significantly related to each other as well and support for American people and culture is more strongly related to each other than support of either with support of the American government.

Table 2.

Bivariate Pearson Correlations for Support for Terrorism and Legitimacy (n = 2,650)

	1	2	3	4	5
1. Support for Terrorism Scale (three items)	1
2. Legitimacy of U.S. government	−.22***	1
3. Support for American people	−.25***	.39***	1
4. Support for American culture	−.20***	.28***	.57***	1
5. Legitimacy of domestic institutions	.01	−.07***	−.11***	−.07***	1

p < .05. ***p < .01; listwise deletion.

Table 3 shows bivariate Spearman correlations between support for al Qaeda and the primary independent variables of interest. Similar to the bivariate correlations, legitimacy in U.S. government and support for American people and culture are negatively related to support for al Qaeda, indicating that those individuals who attribute greater legitimacy to the United States and more support for American people and culture are less likely to support al Qaeda–based rhetoric or attacks. However, unlike previous correlations for support of Muslim-based groups that attack the United States, support for domestic legitimacy is negatively related to support for al Qaeda. That is, individuals reporting higher levels of legitimacy in domestic institutions are less likely to support the views and actions of al Qaeda. In addition, unlike recent research by McCauley et al. (2011) for Muslims in Ottawa, we find that as support for the United States increases, support for al Qaeda declines.

Table 3.

Bivariate Spearman Correlations for Support for al Qaeda and Legitimacy (n = 2,631)

	1	2	3	4	5
1. Support for al Qaeda	1
2. Legitimacy of U.S. government	−.10***	1
3. Support for American people	−.14***	.36***	1
4. Support for American culture	−.13***	.25***	.51***	1
5. Legitimacy of domestic institutions	−.10***	−.07***	−.10***	−.05***	1

p < .05. ***p < .01; listwise deletion.

Table 4 shows the result from our multivariate regression analysis of legitimacy on support for Muslim-based groups that attack Americans, including all demographic, attitudinal, and behavioral control variables. Column 1 provides results from the analysis of support for Muslim-based groups using the composite measure, including all three items of support for terrorism.

Table 4.

Regression Results: Support for Terrorism, Pooled Sample, 2006 to 2007

	Model 1 (three-item measure)	Model 2 (two-item measure)	Model 3 (al Qaeda)
	b (SE)	b (SE)	b (SE)
Legitimacy and support variables
Legitimacy of U.S. government	0.02 (.03)	0.02 (.03)	−0.01 (.08)
Support for U.S. people	−0.11**** (.02)	−0.12**** (.02)	−0.11* (.06)
Support for U.S. culture	−0.05*** (.02)	−0.04** (.02)	−0.12** (.06)
Perceived domestic legitimacy	−0.01** (.01)	−0.01 (.01)	−0.07**** (.02)
Demographic variables
Age	−0.04*** (.01)	−0.04** (.02)	−0.08*** (.04)
Gender	−0.01 (.03)	−0.03 (.04)	−0.01 (.11)
Marital status	−0.09** (.04)	−0.07 (.04)	−0.16 (.12)
Muslim affiliation	0.40**** (.08)	0.37**** (.09)	1.04*** (.26)
Education	−0.03 (.02)	−0.00 (.03)	−0.03 (.08)
Employment	0.03 (.03)	0.03 (.04)	0.11 (.11)
Income	−0.01 (.02)	−0.03 (.02)	0.04 (.07)
Attitudinal and behavioral variables
Interaction with foreigners	−0.02 (.02)	−0.00 (.02)	−0.05 (.05)
International news exposure	0.01 (.00)	0.00 (.00)	0.01 (.01)
Perception of domestic terrorism	0.00 (.02)	0.01 (.02)	−0.13 (.05)
Perception the United States has harmed interests	0.15**** (.04)	0.14**** (.04)	0.37**** (.11)
Perception the United States is a threat to Islam	0.06**** (.02)	0.06* (.03)	0.20*** (.07)
Religious identity	0.30**** (.03)	0.33**** (.04)	0.44**** (.11)
Country location variables
Egypt	0.18**** (.05)	0.33**** (.06)	−0.41** (.17)
Morocco	−0.25**** (.07)	−0.25**** (.07)	−0.43** (.21)

Note: SE = standard error.

p < .10. **p < .05. ***p < .01. ****p < .001.

We find support for two out of our three hypotheses about support for terrorist attacks against the United States. As predicted, we find that respondents who express greater support for the American people show less support for groups that target the United States. Moreover, these findings hold despite the strong effects for identification with religion as well as perceptions of the United States as a threat to Islam. We also find significant effects for perceptions of American culture on support for attacks against Americans. Individuals reporting higher levels of support for American culture are less likely to support Muslim-based groups that attack the United States. There is also support for our hypothesis regarding the legitimacy of domestic institutions and support for terrorism. Although not as strongly related to terrorism as is support for U.S. people or culture, perceptions of legitimacy in domestic institutions are significantly related to lower support for terrorism directed at the United States. But contrary to our predictions, we find no evidence that as the perceived legitimacy of the U.S. government increases, support for groups that attack the United States declines.

As expected, compared with younger respondents, older respondents were less likely to report support for groups that attack Americans. Compared with unmarried respondents, married respondents reported less support for groups that attack Americans. Muslim respondents were significantly more likely than non-Muslim respondents to voice support for Muslim-based groups that attack Americans. Similarly, respondents who identified strongly with their religion were more likely to support attacks against Americans. We found no effect on support for attacks on Americans for gender, education, employment, income, international news exposure, interaction with foreigners, or perceptions of terrorism within one’s own country.

Not surprisingly, individuals who reported that the U.S. government had directly harmed their interests were more likely to support attacks on Americans. The belief that the United States’s purpose is to weaken and divide Islam and experiences of having one’s interest directly harmed by the U.S. government were both related to increased support for Muslim-based groups that attack Americans.

Finally, in all models, the coefficients for the country dummy variables were significant. Consistent with the differences shown in Figure 2, compared with Indonesians, Egyptians were more likely to support Muslim-based groups that attack the United States. By contrast, Moroccans were less likely than Indonesians to support attacks against the United States.¹²

Column 2 in Table 4 shows results from the analysis of support for Muslim-based groups that attack the United States, excluding the al Qaeda measure. Despite the measurement change, we find similar effects for positive views toward American people and culture on support for Muslim-based groups that attack the United States. Once again, respondents who report more favorable attitudes about U.S. people and culture are less likely to support violence toward the United States committed by Muslim-based groups. However, unlike the Model 1 results, we do not find in Model 2 a significant relationship between perceived legitimacy of domestic institutions and support for terrorism directed at the United States.

We find similar effects in Model 2 for many of the individual attitudinal and behavioral measures that were related to the three-item terrorism composite measure used for Model 1. For example, older individuals are less likely to support Muslim-based groups that attack the United States. Religious affiliation and identity are again significant predictors of support for Muslim-based groups that use terrorist violence toward the United States. The findings for perceptions of harm caused by the United States and the belief that the United States’s purpose is to weaken Islam are also related to increased support for terrorism. Finally, compared with Model 1 results, the impact of country locations in Model 2 remains similar. Whereas Egyptians are more likely than Indonesians to support Muslim-based groups that attack the United States, Moroccan respondents were significantly less likely to support terrorism as compared with Indonesians.¹³

Column 3 in Table 4 displays results from the logistic regression analysis of the item measuring support for al Qaeda.¹⁴ Again, despite changes in the measurement of support for terrorist attacks, we find significant effects for favorable views toward U.S. people and culture (although the measure of support for U.S. people is only significant at p < .10). In addition, we find that those individuals reporting higher levels of legitimacy in their own institutions of formal authority are more likely to report lower levels of support for attacks on U.S. targets by al Qaeda.¹⁵ This is particularly interesting because the effects for domestic legitimacy are not significant in Model 2 that excludes the al Qaeda measure.

To summarize, despite differences across the three models of terrorism summarized in Table 4, there are also major similarities. In general, we hypothesized that higher levels of perceived legitimacy of U.S. government and domestic institutions, and higher levels of support for U.S. people and culture, will be related to lower levels of support for terrorist attacks on Americans. For the nine equations examining the levels of support for U.S. people and culture and the perceived legitimacy of domestic institutions, all nine were in the expected negative direction and all but two were statistically significant (and one of these two was marginally significant at p < .10). By contrast, two of the three results for the three models of the legitimacy of the U.S. government were not in the expected negative direction and none were significant. In short, we find strong support for our hypotheses that as favorable views toward U.S. people and culture increase, support for terrorist attacks on Americans declines, and moderately strong support for our hypothesis that as perceived legitimacy of domestic institutions increases, support for terrorist attacks on Americans declines. We also found that those who view the United States’s purpose as destructive to Islam are more supportive of attacks on U.S. targets. We found no support for the hypothesis that greater perceived legitimacy of the U.S. government is associated with lower support for anti-American terrorism.

Discussion and Conclusion

In contrast to several earlier studies (Bueno de Mesquita, 2007; Tessler & Robbins, 2007), we found strong support for the conclusion that respondents who view U.S. people and culture more favorably are less likely to support Muslim-based groups that attack Americans. It may be that these differences are explained by the diverse ways these measures were used in prior research. For example, Tessler and Robbins (2007) examined whether respondents in Algeria and Jordan perceived U.S. culture as harmful to their own culture, whereas Bueno de Mesquita (2007) asked respondents in 13 predominantly Muslim countries whether they had favorable perceptions of U.S. culture. Instead, we use a normative-based measure that asked respondents to report their attitudes toward U.S. people and culture from very favorable to very unfavorable.

Moreover, in contrast to Tessler and Robbins (2007) and Bueno de Mesquita (2007), we did not find that the perceived legitimacy of the U.S. government significantly affected support for Muslim-based groups that attack Americans. This result was consistent for both the pooled and country-specific analyses and again may be a consequence of differing measures of government legitimacy used in the previous and current studies. Tessler and Robbins found evidence that negative perceptions of U.S. foreign policy were significantly related to approval of terrorist attacks against the United States among Algerian and Jordanian respondents. For example, in Algeria, respondents were asked to report whether they believed U.S. foreign policies toward other countries were bad or good, and in Jordan, respondents were asked to report their level of approval of the statement, “The United States often violates other people’s human rights around the world.” By contrast, we asked respondents to indicate how favorably they viewed the current U.S. government. It may be that the more specific measures used by Tessler and Robbins and Bueno de Mesquita are tapping opposition to recent U.S. foreign policy rather than more general attitudes toward the U.S. government.

Although we include a measure that directly taps beliefs about the United States as a threat to Islam, variables representing religious affiliation and identification still remain significant across the models: Respondents who identified more strongly with Islam were more supportive of attacks on U.S. targets. It is important to note that other scholars have found evidence showing that violence in the name of Islam is condemned and receives little support among large majorities of Muslims (Esposito, 2002) and many argue that it is important not to “conflate the religion of Islam with the actions of those who hijack Islamic discourse to justify acts of terrorism” (Tessler & Robbins, 2007, p. 307). Moreover, our strong findings for the connection between religious identification and support for terrorism may be based in large part on the fact that many of these respondents were being interviewed in the midst of two large-scale U.S.-led military actions directed at predominantly Muslim countries. Nevertheless, the results show how closely intertwined religious identity is with perceptions of terrorism in these countries.

We currently focus on one aspect of support for terrorist groups that attack Americans. However, support for terrorism can manifest itself at many different levels, from support of some ideas associated with Muslim-based groups, to financial support of such groups, to direct participation in attacks. Existing research conducted in multiple predominantly Muslim countries, although mixed, indicates that correlates (educational attainment, income) differ for support for attacks against civilians and support for attacks against Western military targets (Shafiq & Sinno, 2010). This leaves open the possibility that there may be different correlates for different types of support for transnational terrorism. For example, it may be that the legitimacy of U.S. foreign policy is more significantly associated with support for violence against military or government targets, whereas support for the American people may be more strongly associated with support for violence against civilian targets.

Our findings suggest that there is an important distinction between views of the U.S. government and views of the American people and culture. Favorable attitudes toward civilian populations that are targets of violent acts may be associated with reduced support for those groups that perpetrate attacks on civilians. This might suggest that policies and public narratives for reducing terrorism should emphasize civilian losses and cross-cultural connections between ordinary citizens. The finding that the perception of the United States as a threat to Islam is related to anti-American attacks may reflect the fact that in some cases U.S. policy has increased support for terrorism. Although not tested in this study, it is certainly possible that perceptions of the legitimacy of the U.S. government are in turn directly related to prior U.S. policies. It is important to note the limitations of the cross-sectional design used in the current study. Future research should use longitudinal data to examine the extent to which these types of actions and incidents lead to increases in support for terrorist groups as well as unfavorable attitudes toward U.S. government, people, and culture.

We also find support for the conclusion that perceptions of legitimacy may have more general effects on compliance with the law. There is growing evidence that the relationship between procedural justice and legitimacy is applicable in a wide variety of contexts (Gibson & Caldeira, 1995; Jonathan-Zamir & Weisburd, 2010). Similarly, our results show that respondents who viewed their own institutions more favorably were also less supportive of anti-American attitudes and attacks on the United States—especially those by al Qaeda. Policy implications from studies of procedural justice and legitimacy of authorities on compliance with the law have focused on the benefits of increasing compliance with the law of those specific authorities, but the findings from this study suggest that such policy approaches may also yield returns in terms of broader compliance. To the extent that fostering support for local formal authorities not only leads to reductions in crime locally but also diminishes support for transnational violence, it would have very significant policy implications. However, it is also important to point out that this finding disappears in the pooled analysis when the question about al Qaeda is excluded from the composite measure of support for terrorism.

We also found that support for al Qaeda produced somewhat different results from more generic forms of support for groups that attack U.S. targets. In particular, when we gauged support for terrorism specific to al Qaeda, the connection between the legitimacy of domestic institutions and opposition to terrorist attacks was stronger and the connection between support for U.S. people and culture and opposition to terrorist attacks was weaker. Part of these differences may reflect sample selection: Compared with questions about terrorism more generally, significantly fewer respondents answered items specific to al Qaeda. The difference may also reflect the fact that support for al Qaeda among moderate Muslims has declined substantially after continuing evidence of its violence and brutality directed at civilians (Zakaria, 2010). In either case, it is worth emphasizing that we find a strong connection between the perceived legitimacy of domestic formal control institutions such as the police and opposition to al Qaeda. Indeed, these results are also consistent with Tessler and Robbins (2007) who found significant effects for perceived domestic legitimacy when examining two outcome measures of terrorism that explicitly refer to the extremist Muslim-based groups and a jihad against the United States. In both cases, individuals who expressed higher perceived levels of legitimacy in domestic institutions were less likely to report support for extremist Muslim-based groups.

We began this article by reviewing recent survey evidence that shows increasingly negative perceptions of the United States among Muslims in other countries. Our descriptive results tell a similar story, with large majorities of respondents in all three countries reporting unfavorable to very unfavorable views of the U.S. government. Yet the evidence regarding Muslim perceptions toward U.S. culture and the American public is far more complex. Our results indicate that despite negative views of the U.S. government, favorable perceptions of American people and culture are related to less self-reported support for Muslim-based groups that attack Americans. Furthermore, the belief that the United States is a direct threat to Islam is related to increased support for groups engaged in transnational terrorism against the United States. Although policy makers have suggested that increased communication may improve relationships between the United States and the Muslim world, recent evidence (Gentzkow & Shapiro, 2004) also suggests that exposing individuals from predominantly Muslim countries to a broader range of media sources may reduce animosity and negative perceptions toward the United States. Further research is necessary to determine more precisely the source of the strong negative attitudes toward the U.S. government and the source of the belief that the United States is a threat to Islam in predominantly Muslim countries.

Footnotes

Acknowledgements

We want to recognize Derrick Franke who contributed greatly to earlier drafts of this article and Clark McCauley and James Forest for their helpful comments. We also want to thank Steven Kull for providing data on focus groups.

Authors’ Note

Any opinions, conclusions, or recommendations in this article are those of the authors and do not necessarily reflect the views of Department of Homeland Security (DHS).

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Support for this research was provided by the Science and Technology Division of the Department of Homeland Security (DHS) through the National Consortium for the Study of Terrorism and Responses to Terrorism (START).

Notes

Bios

Gary LaFree is a director of the National Consortium for the Study of Terrorism and Responses to Terrorism (START) at the University of Maryland, as well as professor in the Department of Criminology and Criminal Justice. Much of his current research is related to the development and analysis of the Global Terrorism Database, a major project being supported by START.

Nancy A. Morris completed her PhD in Criminology and Criminal Justice at University of Maryland, College Park. She is an assistant professor in the L. Douglas Wilder School of Government and Public Affairs at Virginia Commonwealth University. Her research interests and activity thus far can be classified into three broad policy relevant areas: cross-national criminology, juvenile delinquency and developmental criminology; and longitudinal research methodology.

References

Allison

P. D.

(1999). Logistic regression using the SAS system: Theory and application. Cary, NC: SAS Institute.

Bernard

(2004, Spring). Five pillars of democracy: How the West can promote an Islamic reformation. Rand Review. Retrieved from http://www.rand.org/publications/randreview/issues/spring2004/pillars.html

Betts

R. K.

(2002). Fixing intelligence. Foreign Affairs, 81, 43-59.

Bueno de Mesquita

(2007). Correlates of public support for terrorism in the Muslim world (United States Institute of Peace Working Paper). Washington, DC: United States Institute for Peace.

Central Intelligence Agency. (2011). The World Factbook. Retrieved from https://www.cia.gov/library/publications/the-world-factbook/index.html

Crenshaw

(1983). Terrorism, legitimacy and power. Scranton, PA: Wesleyan University Press.

Crenshaw

(2000). The psychology of terrorism: An agenda for the 21st century. Political Psychology, 21, 405-420.

Easton

(1965). A systems analysis of political life. Chicago, IL: University of Chicago Press.

Easton

Dennis

(1969). Children in the political system: Origins of political legitimacy. New York, NY: McGraw-Hill.

10.

Esposito

(2002). Unholy war: Terror in the name of Islam. New York, NY: Oxford University Press.

11.

Fair

C. C.

Shepherd

(2006). Who supports terrorism? Evidence from fourteen Muslim countries. Studies in Conflict and Terrorism, 29, 51-74.

12.

Fraser

(1974). Validating a measure of national political legitimacy. American Journal of Political Science 18, 117-134.

13.

Friedrichs

D. O.

(1980). Legitimacy crisis in the United States: A conceptual analysis. Social Problems, 27, 105-135.

14.

Garfinkle

(2004). Anti-Americanism, U.S. foreign policy, and the war on terrorism. In Garfinkle

(Ed.), A practical guide to winning the war on terrorism (pp. 197-218). Palo Alto, CA: Hoover Books Online.

15.

Gentzkow

M. A.

Shapiro

J. M.

(2004). Media, education and anti-Americanism in the Muslim world. Journal of Economic Perspectives, 18, 117-133.

16.

Gibson

J. L.

(1991). Institutional legitimacy, procedural justice, and compliance with Supreme Court decisions: A question of causality. Law & Society Review, 25, 631-636.

17.

Gibson

J. L.

Caldeira

(1995). The legitimacy of transnational legal institutions: Compliance, support, and the European Court of Justice. American Journal of Political Science, 39, 459-489.

18.

Gurr

(1970). Why men rebel. Princeton, NJ: Princeton University Press.

19.

Habermas

(1975). Legitimation crisis. Boston, MA: Beacon Press.

20.

Hoffman

(2003). Al Qaeda, trends in terrorism and future potentialities: An assessment. Santa Monica, CA: RAND.

21.

Huntington

(1993). The clash of civilizations? Foreign Affairs, 72, 22-49.

22.

Huntington

(1996). The clash of civilizations and the remaking of world order. New York, NY: Simon & Schuster.

23.

Jonathan-Zamir

Weisburd

(2010). How do majority communities view the potential costs of policing terrorism? Findings from a community survey in Israel. Policing: A Journal of Policy and Practice, 4, 169-181.

24.

Kydd

A. H.

Walter

B. F.

(2006). The strategies of terrorism. International Security, 31, 49-80.

25.

LaFree

(1998). Losing legitimacy: Street crime and the decline of social institutions in America. Boulder, CO: Westview Press.

26.

Lewis

(1990). The roots of Muslim rage: Why so many Muslims deeply resent the West and why their bitterness will not be easily mollified. Atlantic Monthly, 266, 47-60.

27.

Mastrofski

S. D.

Snipes

J. B.

Supina

A. E.

(1996). Compliance on demand: The public’s response to specific police requests. Journal of Research in Crime & Delinquency, 33, 269-305.

28.

McCauley

(2006). Jujitsu politics: terrorism and responses to terrorism. In collateral damage: The psychological consequences of America’s war on terrorism. Westport, CT: Praeger.

29.

McCauley

Leuprecht

Hataley

Winn

Biswas

(2011). Tracking the war of ideas: A poll of Ottawa Muslims. Terrorism and Political Violence, 23, 804-819.

30.

McCauley

Scheckter

(2008). What’s special about U.S. Muslims? The war on terrorism as seen by Muslims in the United States, Morocco, Egypt, Pakistan, and Indonesia. Studies in Conflict and Terrorism, 31, 973-980.

31.

McCauley

Stellar

(2009). U.S. Muslims after 9/11: Poll trends 2001-2007. Perspectives on Terrorism, 3, 35-47.

32.

Nieburg

H. L.

(1969). Political violence. New York, NY: St. Martin’s Press.

33.

O’Connor

(1973). The fiscal crisis of the state. New York, NY: St. Martin’s Press.

34.

Orehek

Fishman

Dechesne

Doosje

Kruglanski

Cole

. . .Jackson

(2010). Need for closure and the social response to terrorism. Basic and Applied Social Psychology, 32, 279-290.

35.

Paternoster

Brame

Bachman

Sherman

(1997). Do fair procedures matter? The effect of procedural justice on spouse assault. Law & Society Review, 31, 163-204.

36.

Paz

(2003). Islamists and anti-Americanism. Middle East Review of International Affairs, 7, 53-61.

37.

Pew Center. (2011). Forum on religion and public life. Available from http://pewforum.org/

38.

Raudenbush

S. W.

Bryk

A. S.

(2002). Hierarchical linear models: Applications and data analysis methods (2nd ed.). Newbury Park, CA: SAGE.

39.

Sampson

R. J.

Bartusch

D. J.

(1998). Legal cynicism and (subcultural?) tolerance of deviance: The neighborhood context of racial differences. Law & Society Review, 32, 777-804.

40.

Shafiq

M. N.

Sinno

A. H.

(2010). Education, income, and support for suicide bombings: Evidence from six Muslim countries. Journal of Conflict Resolution, 54, 146-178.

41.

Sharp

(1973). The politics of non-violent action. Boston, MA: Porter Sargent.

42.

Sherman

(1993). Defiance, deterrence and irrelevance: A theory of the criminal sanction. Journal of Research in Crime & Delinquency, 30, 445-473.

43.

Sunshine

Tyler

T. R.

(2003). The role of procedural justice and legitimacy in shaping public support for policing. Law & Society Review, 37, 513-548.

44.

Tessler

Robbins

M. D.

(2007). What leads some ordinary Arab men and women to approve of terrorists acts against the United States? Journal of Conflict Resolution, 51, 305-328.

45.

Thibaut

Walker

(1975). Procedural justice: A psychological analysis. Hillsdale, NJ: Erlbaum.

46.

Tyler

T. R.

(1990). Why people obey the law. New Haven, CT: Yale University Press.

47.

Tyler

T. R.

(2003). Process based regulation. Procedural justice, legitimacy, and the effective rule of law. In Tonry

(Ed.), Crime and justice (Vol. 30, pp. 283-358). Chicago, IL: University of Chicago Press.

48.

Tyler

T. R.

Casper

J. D.

Fisher

(1989). Maintaining allegiance toward political authorities: The role of prior attitudes and the use of fair procedures. American Journal of Political Science, 33, 629-652.

49.

Tyler

T. R.

Huo

(2002). Trust in the law: Encouraging public cooperation with the police and courts. New York, NY: Russell Sage.

50.

Tyler

T. R.

Rasinski

(1991). Procedural justice, institutional legitimacy, and the acceptance of unpopular U.S. Supreme Court decisions: A reply to Gibson. Law & Society Review, 25, 621-630.

51.

Tyler

T. R.

Wakslak

C. J.

(2004). Profiling and police legitimacy: Procedural justice, attributions of motive and acceptance of police authority. Criminology, 18, 267-274.

52.

Useem

(1979). Government legitimacy and political stability. Social Forces, 57, 840-852.

53.

Weber

(1947). The Theory of Social and Economic Organizations. New York: Oxford University Press.

54.

Weber

(1968). Economy and society: An outline of interpretive sociology ( Fischoff

Gerth

Henderson

A. M.

Kolegar

Wright Mills

Parsons

. . . Wittich

, Trans.). In Roth

Wittich

(Eds.), New York, NY: Bedminster Press.

55.

White

J. K.

(2004). Terrorism and the remaking of American politics. In Crotty

(Ed.), The Politics of terror: US responses to 9/11 (pp. 37-63). Lillington, NC: Northeastern University Press.

56.

Zakaria

(2010). The Jihad against the Jihadis: How moderate Muslim leaders waged the war on extremists—And won. In Badey

T. J.

(Ed.), Violence and terrorism (pp. 144-147). New York, NY: McGraw-Hill.