Permitted to Build? Moral Foundations in Newspaper Framing of Mosque-Construction Controversies

Abstract

Recent proposals by American Muslims to build mosques have been met with technical and ideological opposition during the permitting process. This article examines the framing of these debates in newspaper stories between 2010 and 2013 to better understand the socially constructed position Muslims hold in American media and public spheres. Connecting framing and Moral Foundations Theory, this analysis identifies five frames: Local Regulation, Political Debate, Muslim Neighbors, Islamic Threat, and Legal Authority. These frames emphasized binding moral foundations related to in-group protection and deference to authority. A binary logistic regression found that moral evaluations were associated with mosque support, but not mosque opposition.

Keywords

framing Islam Moral Foundations Theory cluster analysis newspapers morality

Although Western media representations of Islam have been critiqued for decades (Said, 1997; Shaheen, 2001), coverage of the faith and its adherents surged in the post-9/11 era and has reached a new level of salience at the dawn of President Donald Trump’s administration, with its policies of “extreme vetting” and immigration restrictions. In recent years, media coverage of Islam has been scrutinized in a string of contentious debates widely reported in the news. These issues include international terrorism (Nacos & Torres-Reyna, 2006), wars in Muslim-majority countries (Entman, 2004), immigration (Poynting & Mason, 2007), the Islamic moral code called Shariah (Bowe & Hoewe, 2016), visual depictions the Prophet Muhammad (Shehata, 2007), and traditions related to the hijab head covering and halal food preparation (Poynting & Noble, 2003; Saha, 2012; Scott, 2010).

Faced with voluminous media attention to Islamic beliefs and practices at home and abroad, Muslims in the United States must negotiate issues of identity and belonging within their local communities. This is not an easy task, in part because there remains a disconnect between the diverse lived realities of Muslim Americans and the dominant perceptions of that community. The growing Muslim presence in the United States has occurred against a backdrop of conflict, war, and international terrorism. These long-standing domestic and global tensions make some segments of the general public profoundly nervous about having Muslims as neighbors.

Survey research shows that the Americans view Muslims most negatively of all religious groups (Pew Research Center, 2014; Wormald, 2014), though people who have personal contacts with Muslims tend to hold more positive attitudes (Cooper, 2017; Gallup Center for Muslim Studies, 2010). Many Americans express low knowledge of Islam, but high anxiety toward the faith (Panagopoulos, 2006). Nearly half of respondents in one survey admitted to at least some prejudice against Muslims (Gallup Center for Muslim Studies, 2010), while a similar percentage in another believes that Muslim Americans are not doing enough to fight extremism in their own communities (Cox & Jones, 2015). Reflecting this, some Americans endorse curtailing Muslims’ civil and religious rights in favor of security (Nisbet, Ostman, & Shanahan, 2009). Perhaps as a result of these negative attitudes, American Muslims report levels of discrimination at a higher rate than members of other religious groups (Gallup, n.d.).

Some of the outsider status of American Muslims may stem from the fact that large percentages of this community are either immigrants (61%) or the children of immigrants (17%; Pew Research Center, 2015). However, research suggests that Muslims who immigrate to the West show flexibility in their social values, absorbing many elements of their adopted culture while retaining aspects of their native culture (Norris & Inglehart, 2012). Despite perceptions to the contrary, American Muslims express negligible support for extremism, and a majority express concerns about its rise globally (Pew Research Center, 2011).

Although regarded as outsiders, Muslim Americans are largely middle class, assimilated into U.S. culture and are generally upbeat about their lives and communities (Bagby, 2009; Kohut, 2007; Pew Research Center, 2011). In the aggregate, Muslim Americans tend to support the Democratic Party politically while holding socially conservative views regarding sexuality and gender roles (Pew Research Center, 2011). They are the most racially diverse religious group in the United States, with African Americans making up the largest single bloc (Younis, 2009).

Antipathy toward Islam may have a partisan component in the United States, with Republicans tending to express more negative views of Muslims than Democrats (Gallup, n.d.; Lipka, 2015). For example, Republicans in one study expressed more opposition to the idea of voting for a Muslim president than did Democrats (Newport, 2015). Negative beliefs about Muslims may be especially concentrated and resilient among Republicans, evangelical Protestants, and television viewers who place the most trust in the Fox News network (Calfano, Djupe, Cox, & Jones, 2016).

As the world grapples with continuing Islamist terrorist attacks on the one hand and the rise of strident anti-Islam rhetoric from political actors across the globe on the other, journalists are left to figure out how to responsibly report on the day-to-day stories that involve the Muslims in their communities. Despite their outsized role in public discourse, Muslims make up a relatively small part of the total U.S. population, with estimates ranging from 0.5% to 2.2% (Association of Religion Data Archives, n.d.; Heimlich, 2011; Kohut, 2007; Mohamed, 2016). Because Muslims make up a narrow segment of the U.S. public, it is reasonable to suggest that mass media portrayals may be one of the major contributors to the formation of perceptions (and misperceptions) about Islam among people who do not interact with Muslims in their daily lives. This is an important dynamic for framing researchers, because audience members without direct experiences to draw upon become more vulnerable to mass media framing (Gitlin, 1980).

Especially in times of crisis, mass media have been shown to play a key role in influencing public opinion toward Muslims (Nisbet et al., 2009; D. A. Scheufele, Nisbet, & Ostman, 2005). If media wield a strong influence on audience attitudes, the dominant media representations of Islam and Muslims may exacerbate tensions. In a meta-analysis of 345 scholarly studies over 15 years, Ahmed and Matthes (2017) found that Muslims are usually negatively framed in news coverage, while Islam is mostly depicted as a violent religion. This is perhaps not a surprising result, considering that much of the academic research in this area exclusively examines Islam in the context of conflict-based topics such as migration, terrorism and war.

The Muslim American population may be small, but immigration and high birth rates are expected to see it double by 2030 (Heimlich, 2011). Naturally, this expanding Muslim population seeks to have physical spaces in which to worship. Reflecting this trend, the number of mosques in the United States soared 74% between 2001 and 2011 (Bagby, 2012). Because Islam occupies a controversial space in American attitudes, some of these mosque projects have met with resistance. According to Liu (2012a), at least 53 mosque projects and similar Islamic spaces have seen local regulatory opposition at the level of zoning boards and city councils in recent years.

Some of these debates have been the subject of intense news coverage. For example, the Park51 project in New York—the so-called “Ground Zero Mosque”—was the top national story for several weeks in 2010 (Holcomb, 2010; Morgan, 2010a, 2010b) and was the biggest U.S. religion story of the year (Liu, 2012a). But while the Park51 debate dominated the national discourse for a time, many smaller projects captured attention on a local level. A project proposed by the Islamic Center of Murfreesboro, Tennessee, sparked protests, lawsuits, and vandalism before opening in 2012 (Liu, 2012b). Similarly, a mosque proposed by the Islamic Center of Temecula Valley in California was the subject of fierce protests leading up to its approval by the city council in early 2011 (Liu, 2012b). In the Chicago and Atlanta areas, there were a series of smaller debates related to mosque construction and expansion projects (Liu, 2012b).

These cases typically began with land-use questions about building height, traffic, or parking, but they often morph into a forum for discussion about the status of Muslims in a community. Should Muslims be offered membership within the in-group of contemporary American society? Or should Islam be rejected as an un-American threat to security and social cohesion?

This polarized reaction to Muslims is congruent with the way Muslims are often portrayed in media coverage. A common critique of Western media representations of Islam is that they are generally monolithic and rely on “a handful of rules, stereotypes, and generalizations about the faith, its founder, and all of its people” (Said, 1997, p. xvi). The global Muslim population includes some 1.6 billion people in dozens of countries with a vast array of cultural practices (Pew, 2009). However, media coverage often places these differences of ethnicity, immigration status, sect, age, and so on, in the background in favor of an essentialized, singular Muslim whole (Morey & Yaqin, 2011; Said, 1997). When groups are framed in this way, they create “mediated notions of cultural identity” that are implicated in the construction of audience perceptions of whether the groups in question belong within the broader culture (Bantimaroudis & Kampanellou, 2007, p. 82). Although there is extensive research into how Muslims are covered in the media (see Ahmed & Matthes, 2017), there remains little research into why journalists make the reporting decisions they do. In one of the few articles on this subject, Ewart, Pearson, and Healy (2016) found that Australian journalists were highly self-reflexive about the state of reporting on Islamic issues, observing that stories were often decontextualized using arbitrary images that stereotype Muslims and presentation formats that reproduce previous problematic reporting.

This study attempts to unpack the socially constructed position that Muslims hold in the media and public spheres, as reflected in the news coverage of mosque controversies. It does so by drawing connections between two increasingly important theoretical constructs in communication research—framing and Moral Foundations Theory (MFT). This study also contributes to improving the measurement of frames, particularly at the level of moral evaluations.

Theoretical Framework

Framing describes the way people define situations, based on principles of organization that govern experience (Goffman, 1974). It plays an important role in the social construction of perceptions about issues, because these organizing principles “are socially shared and persistent over time” and “work symbolically to meaningfully structure the social world” (Reese, 2001, p. 11). The resulting frames become embedded in the social environment and are spread across discourse (Reese, 2007). Thus, it would be inaccurate to consider frames merely as features of media content. Rather, they are a common element of a quotidian existence in which people must interpret and respond to reality. However, news media framing remains an important vector of organization and interpretation for a society, in part, because the professional roles and routines of journalists guide the creation (and ultimately meaning) of content.

Media framing is a dynamic process that includes the frame-building activities of partisans who promote particular interpretations, the frames that appear as the patterns of news content, and the consequences of framing on audience members (de Vreese, 2005; Entman, 1993; Gamson, 2001; Pan & Kosicki, 1993; Vliegenthart & van Zoonen, 2011). The components of media frames are influenced by the cognitions of journalists, newsroom routines and conventions, and the political and cultural contexts within which news stories are disseminated (D’Angelo & Kuypers, 2010; Entman, 1993; Entman, 2004; Pan & Kosicki, 1993). These framing processes operate “in the universe of shared culture and on the basis of socially defined roles” (Pan & Kosicki, 1993, p. 55).

Because issue advocates, journalists, and audience members are all entrenched in the same shared culture, framing is not a unidirectional process but one in which all the parts exert influence on the others. This shared repertoire of frames connects news producers and consumers, because journalists must shape content “within a familiar frame of reference and according to some latent structure of meaning” that resonates with audience members, who possess similar outlooks to the journalists (Van Gorp, 2007, p. 61). In fact, Entman (1993) argued that culture might be defined as a social group’s collection of shared frames.

Media Frames as Patterns in Content

The most common conceptual definition of media framing (per Matthes, 2009) is Entman’s (1993) assertion that framing is the promotion of particular interpretations of issues or events through the selection and emphasis of (a) problem definitions, (b) causal interpretations, (c) moral evaluations, and/or (d) treatment recommendations. The framing concept has been a fertile ground for researchers, becoming one of the most frequently used theoretical frameworks for studies in communication journals (Bryant & Miron, 2004). But it has been an area of research widely critiqued for methodological imprecision (Matthes, 2009; Reese, 2007).

Oftentimes, researchers examine media content for predetermined frames (Matthes, 2009). However, a frame is a complex and abstract variable that is not easily found in manifest content. Matthes and Kohring (2008) argued that coding holistic frames from a list determined ahead of time presents two main challenges to validity in framing research. The first is the coder bias that comes from wanting to find the object one is looking for; the second is the need to develop a list of frames in advance, which makes it difficult for researchers to observe new frames as they emerge.

In an attempt to ameliorate these challenges, framing researchers have begun to adopt a clustering technique introduced by Matthes and Kohring (2008) to find frames through content analysis. In this technique, frames are broken down into components based on Entman’s conceptual definition. These components are coded individually, rather than complete frames. A cluster analysis is then performed to detect patterns in the texts. When clustered patterns of frame elements group together systematically, they can be considered frames.

Although this technique does not entirely eliminate the previously identified challenges to validity, Matthes and Kohring (2008) argued that breaking a frame into its constituent parts allows researchers to operationalize the frame elements in ways that are easier to code in manifest content and more closely tied to theory. This technique also removes the researcher one step from the frame identification process. The cluster method has gained popularity in recent years and has been endorsed and/or adopted by other researchers (Bell & Entman, 2011; Bowe & Hoewe, 2016; Bowe, Oshita, Terracina-Hartman, & Chao, 2012; Donk, Metag, Kohring, & Marcinkowski, 2012; Van Gorp, 2010). However, the approach has been critiqued for its reliance on manifest content as a trade-off of validity (B. T. Scheufele & Scheufele, 2010) and for its lack of concern for the effects of frames (Cacciatore, Scheufele, & Iyengar, 2016). B. T. Scheufele and Scheufele (2010) argued that it may offer researchers only a ballpark estimate of frames, and it is therefore best suited as an exploratory tool.

The face validity of the cluster technique is strengthened by the distinct role each framing component plays in making sense of an issue. Problem definitions often predetermine the rest of the frame (Entman, 2003), because “defining the terms of a debate takes one a long way toward winning it” (Tankard, 2001, p. 95). Problem definitions are typically described as a combination of issues and actors, which, taken together, “define the central problem of a news story” (Matthes & Kohring, 2008, p. 266).

Causal attributions are important because different ways of assigning responsibility for an issue can lead to very different interpretations of events, thereby yielding vastly different frames (B. T. Scheufele & Scheufele, 2010). Causal attributions are typically assigned to the actors responsible for the success or failure of specific outcomes, the risks or benefits of an issue, or the responsibility for a controversy (Bowe et al., 2012; Donk et al., 2012; Matthes & Kohring, 2008).

Treatment recommendations offer remedies or specific calls for action for a problematic situation (Matthes & Kohring, 2008). Conceptually, treatment recommendation has been closely tied to problem definitions because each problem definition suggests a certain public policy solution (Entman, 2004).

Neglected in much of the previous research, however, is the element of moral evaluation. In part, this may be because routine journalistic standard of “objectivity” requires moral stances to be made implicitly, through the use of quotes and opinions from involved parties (Semetko & Valkenburg, 2000), making them hard to distinguish. This challenge has been exacerbated by insufficient theoretical grounding, which has made this element particularly problematic for researchers to measure (Bowe et al., 2012).

Researchers have used a variety of conflicting conceptualizations of moral evaluations, such as the benefits or risks of a public policy (David, Atun, Fille, & Monterola, 2011; Matthes & Kohring, 2008), positive or negative evaluations of technology (Donk et al., 2012), or culpability for a controversial action (Bowe et al., 2012). However, none of these descriptions are theoretically grounded, and none of them sufficiently map onto the concept of morality.

Moral Foundations Theory

This study uses MFT to operationalize framing’s moral evaluation dimension. MFT has been widely used to study political discourse (Bowe & Hoewe, 2016; Feinberg & Willer, 2013; Graham, Haidt, & Nosek, 2009; Graham, Nosek, & Haidt, 2012; Haidt & Graham, 2007). Based on insights from anthropology and evolutionary psychology, MFT posits that moral reasoning is based on a series of moral foundations (Graham et al., 2013; Haidt & Joseph, 2004). These moral foundations operate like taste receptors (Haidt & Joseph, 2004)—a finite array of commonly held structures that allow for a wide variety of possible combinations across cultures. The foundations are expressed as virtue/vice pairs that include care/harm, fairness/cheating, loyalty/betrayal, authority/subversion, and sanctity/degradation (Graham et al., 2013).

The care and fairness foundations emphasize the protection of individuals and are referred to as “individualizing” dimensions; the other three foundations connect individuals to communities and institutions and are called “binding” foundations (Graham et al., 2012). Previous research suggests that liberals tend to rely more heavily on moral intuitions that emphasize the binding foundations of care and fairness, while conservatives tend to rely on all five more equally (Haidt & Graham, 2007; Graham et al., 2009). Moving beyond simple left–right distinctions, Haidt, Graham, and Joseph (2009) found four distinct moral foundation profiles related to ideology: secular liberals, libertarians, the religious left, and social conservatives.

Although these five pairs are at the center of the bulk of MFT research, the theory’s founders suggest that there may be other moral foundations. Potential additional foundations include liberty/oppression, efficiency/waste, ownership/theft, and honesty/deception (Graham et al., 2013). This research utilizes the original five foundations, because the others have not yet been validated empirically.

Because MFT was designed to examine morality at a cultural rather than individual level (Haidt et al., 2009), it may be particularly useful for unpacking contemporary American media discourse about Islam and the culture of the mainstream media outlets analyzed. The tension over mosque construction in the United States is at the center of this study. To identify frames, MFT’s five moral foundations—care/harm, fairness/cheating, loyalty/betrayal, authority/subversion, and sanctity/degradation—were considered alongside variables measuring problem definitions, causal attributions, and treatment recommendations.

This analysis began with the following exploratory research question:

RQ1: How do the patterns of problem definitions, causal attributions, moral evaluations, and treatment recommendations combine to create frames in coverage of mosque controversies in American newspapers?

Because MFT posits a moral “flavor profile” that is connected to certain political ideologies, we should be able to use MFT to determine the ideological moral reasoning underpinning the frames—whether secular liberal, libertarian, religious left, social conservative (Haidt et al., 2009), or some heretofore undiscovered profile. Special attention was paid to the moral evaluation component of each frame to answer the following research question:

RQ2: Do the moral profiles of the frames in the news coverage of mosque-building controversies reflect particular ideologies?

As was noted in the preceding literature review, research suggests a strong partisan divide in American attitudes toward Muslims. Given that divide, it seems logical that mosque opposition may be more associated with political conservatism in the United States. Building on previous MFT research that suggests that particular moral foundation patterns are associated with certain political ideologies, (particularly Haidt et al., 2009), this study hypothesized that the moral dimensions present in articles may be predicted based on support or opposition to mosque construction. Specifically,

H1: Antimosque articles will emphasize the moral dimensions of loyalty, authority, and sanctity.

H2: Promosque articles will emphasize the moral dimensions of care and fairness.

Method

This study is based on a content analysis of articles (n = 349) from five daily newspapers in the United States—the Atlanta Journal-Constitution (n = 45), the Chicago Tribune (n = 37), the New York Times (n = 83), the Nashville Tennessean (n = 145), and the Californian (Temecula, CA; n = 39). The study’s content population was downloaded from newspaper websites and online databases (LexisNexis, Gannett Newsstand, ProQuest) using the search term “mosque” for stories published between January 1, 2010, and December 31, 2013 (except for the Californian, which published its final edition on May 28, 2013). Not all of the mosque controversies identified by Liu (2012b) received extensive media coverage. The selected papers in this study were chosen to provide geographic diversity, and also because they featured coverage of mosque controversies. In some areas, the controversy centered mostly on a single controversial mosque project (e.g., Temecula and New York); in the others (Chicago, Atlanta, and Nashville), there were multiple projects that drew controversy.

The articles were downloaded and narrowed to staff-produced news stories, staff editorials, and bylined columns specifically concerning mosque construction controversies. To be included, the mosque project needed to be mentioned in the headline or lead, and the article had to be substantially about the project (rather than just a passing reference). Articles mentioning mosques in other locations or using the word “mosque” generally were discarded. In practice, there were virtually no borderline cases. Letters to the editor and wire copy were discarded. The sampling frame included the period in 2010 when a plan to construct an Islamic community center in New York was at its most controversial nationally. Local, staff-produced stories about that controversy were kept, but wire service content was discarded. The unit of analysis was the article.

Methodologically, this study relied on a combination of human-coded and computer-assisted quantitative content analysis. Scholars are increasingly recommending such hybrid approaches that combine the strengths of each method (Lewis, Zamith, & Hermida, 2013; Zamith & Lewis, 2015). Pairing traditional and computational approaches “can facilitate more expansive and powerful—yet still reliable and meaningful—forms of content analysis,” noted Zamith and Lewis (2015, p. 308). However, despite the promise of increased reliability and efficiency from the use of algorithmic text analysis tools, Lacy, Watson, Riffe, and Lovejoy (2015) cautioned that researchers must address the validity of their measures.

To determine moral evaluation, individual stories were analyzed using a moral foundations dictionary created for use with Linguistic Inquiry and Word Count (LIWC) software. This software counts virtue and vice terms related to the five foundations and calculates a score that is a percentage of foundation terms in each text (Graham et al., 2009). This is a technique that has been used in previous research to determine moral foundations in texts (Bowe & Hoewe, 2016; Clifford & Jerit, 2013; Dehghani, Sagae, Sachdeva, & Gratch, 2014; Garten, Boghrati, Hoover, Johnson, & Dehghani, 2016; Graham et al., 2009; Jairam, 2012; Motyl, 2012; Sagi & Dehghani, 2014; Teernstra, van der Putten, Noordegraaf-Eelens, & Verbeek, 2016), thus replicating a well-established method. Lacy et al. (2015) argued that algorithmic textual analysis tools (such as LIWC) are best used in studies of well-archived digital data in contexts (such as digitized newspaper stories storied in databases) that are concerned with explicitly manifest variables (like the ones in the MFT dictionary for LIWC).

Human coders analyzed the rest of the variables in this study. Problem definition was defined as the combination of central subissues and actors, following Matthes and Kohring (2008). Issues included regulatory processes, legal processes, activities by mosque proponents and opponents, threats of terrorism, immigration, vandalism/hate crimes, and politics. Because some stories included several subissues, the main subissue was considered the one identified in the headline and/or lead paragraph of the story. Actors were defined as attributed sources, because a major part of framing takes place when journalists decide whom to quote or paraphrase. Sources were coded by category (i.e., elected officials, mosque supporters, mosque opponents, neutral experts, religious leaders, etc.). The list of problem definitions and source types was generated inductively by reviewing nonstudy articles about similar controversies during the process of developing the content analysis protocol.

Causal responsibility relates to the origin of a problem (Iyengar, 1996). Based on the inductive review of nonstudy articles during protocol development, it became clear that one of the ideological roots of mosque controversies is the question of whether Islam represents a dangerous political ideology that should be opposed, or whether it is a religion entitled to the same First Amendment protections afforded other religions. Thus, in this study, causal attribution was operationalized as the presence of mosque opponents arguing that Islam is not a religion but a political force, mosque proponents asserting Muslims’ freedom of religion, “neither,” or “both.”

Entman (2003) noted that one of the functions of news frames is to endorse remedies or improvements to problematic situations. In this study, treatment recommendation was coded as calls in favor of or in opposition to the construction of a mosque. Primary attention was paid to concrete actions (legal or regulatory rulings or mosque openings). In opinion pieces, the author’s position on the question was easily discerned. News pieces were coded for whether both arguments were roughly equally treated in the article or whether one side dominated. If a remedy endorsement was not present, stories were coded “neither.”

To assess intercoder reliability, two coders coded a randomly selected subsample of 10% of the corpus. Reliability was calculated using observed agreement and Krippendorff’s α to correct for chance agreement, as recommended (Krippendorff, 2011; Lacy et al., 2015; Lombard, Snyder-Duch, & Bracken, 2002). Because multiple sources could be coded for each story, reliability for that variable was calculated for each source type. For the 12 different types of sources, observed agreement ranged between 91.4% and 100%. For sources, the average of Krippendorff’s α = 0.74. The other reliabilities were as follows: issue (80%, α = 0.76), causal attribution (85.7%, α = 0.80), and treatment recommendation (80%, α = 0.64). Although treatment recommendation has a lower corrected reliability than would normally be acceptable, it is a variable that had little variation—many of the stories were coded “both”—and that had relatively few options for coders to choose from. Both of those characteristics work against the conservative Krippendorff’s coefficient. However, because observed agreement was comparable to the other variables, the decision was made to retain treatment recommendation in the analysis. Although some scholars would counsel dropping the variable entirely (Lacy et al., 2015), similar justification for retaining such a variable has been offered in other research (Bowe et al., 2012; Brossard, Shanahan, & McComas, 2004).

Results

Patterns among the framing variables were found using Ward’s (1963) hierarchical cluster analysis method. Although other clustering algorithms have been used (David et al., 2011; Donk et al., 2012), Ward’s method is the most commonly used in this type of framing research (Bowe & Hoewe, 2016; Bowe et al., 2012; Matthes & Kohring, 2008). Variables with an occurrence of less than 10% were excluded from the cluster analysis, because such infrequent variables are not capable of contributing meaningfully to the formation of clusters. This led to the exclusion of the issues of interfaith activities, vandalism, and political campaigns, as well as law enforcement sources, and stories with no attributed sources. After consulting the scree plot to look for a strong “elbow” in the diagram, and comparing several cluster solutions, the five-cluster solution was determined to be superior in terms of interpretability. In cluster analysis, there are trade-offs in the number of clusters selected—fewer clusters are easier to interpret, while more clusters may allow for the discernment of more subtle differences between groups (Mooi & Sarstedt, 2010). Because this research is concerned with the broadly descriptive frames, cluster interpretability was paramount.

Frame Analysis

To answer RQ1, the means of the clustering variables were examined to determine which ones were most important in determining cluster membership. Taken together, the combination of dominant variables in each cluster represents patterns in the articles contained in that cluster. As a rule of thumb, means greater than 0.45 were considered as primary components of a cluster, while means between 0.25 and 0.45 were considered as secondary components. Means smaller than 0.25 are too small to be an interpretable part of a cluster. The means of the variables in each cluster are displayed in Tables 1 to 4.

Table 1.

Main News Story Topics in Mosque Construction Debate, by Frame.

Frame	Regulatory processes	Legal processes	Opponent protests	Features
Regulatory Development	0.4634	0.0976	0.0569	0.252
Political Debate	0.1343	0.1343	0.2687	0.209
Islamic Threat	0	0	0.2647	0.5882
Muslim Neighbors	0.2143	0.0833	0.0357	0.4405
Legal Authority	0	1	0	0

Note. Primary components are bolded; secondary components are shaded. Numbers indicate the means of issue presence by article. Means greater than 0.45 are primary components of a cluster; means between 0.25 and 0.45 are secondary components.

Table 2.

Sources in Mosque Construction Debate Stories, by Frame.

Frame	Local government officials	Promosque activist	Antimosque activist	Academic expert	Political figures	Muslim leaders	Other religious community leaders	Judges	General mosque supporters	General mosque opponents
Regulatory Development	0.35	0.203	0.122	0.138	0.016	0.512	0.114	0.041	0.114	0.171
Political Debate	0.403	0.224	0.597	0.149	0.284	0.672	0.284	0.06	0.209	0.164
Islamic Threat	0.147	0.176	0.5	0.147	0.265	0.265	0.265	0.059	0.029	0.176
Muslim Neighbors	0.274	0.25	0.238	0.202	0.155	0.583	0.25	0.071	0.167	0.167
Legal Authority	0.683	0.146	0.805	0.146	0.049	0.244	0	0.463	0	0.024

Note. Primary components are bolded; secondary components are shaded. Numbers indicate the means of source presence by article. Means greater than 0.45 are primary components of a cluster; means between 0.25 and 0.45 are secondary components.

Table 3.

Moral Foundations in Mosque Construction Debate Stories, by Frame.

Frame	Care	Harm	Fairness	Cheating	Loyalty	Betrayal	Authority	Subversion	Sanctity	Degradation
Local Regulation	0.1812	0.1103	0.0706	0.0437	0.7195	0.038	0.5016	0.0347	0.1304	0.0207
Political Debate	0.1973	0.3672	0.139	0.0676	0.8187	0.1876	0.5027	0.0927	0.1673	0.0463
Islamic Threat	0.1718	0.3	0.1497	0.0594	0.8182	0.2918	0.4685	0.1318	0.1874	0.0365
Muslim Neighbors	0.2095	0.347	0.1361	0.1304	0.791	0.2002	0.5687	0.0604	0.2396	0.0477
Legal Authority	0.1285	0.1173	0.2671	0.1766	0.31	0.0934	0.8198	0.1215	0.0998	0.0054

Note. Primary components of frames are in boldface; secondary components are shaded. Numbers represent means of moral foundation words per article. Means greater than 0.45 are primary components of a cluster; means between 0.25 and 0.45 are secondary components.

Table 4.

Causal Attribution and Treatment Recommendation, by Frame.

Frame	Islam/political ideology	Muslim religious rights	Both	Neither	Stop mosque	Allow mosque	Both	Neither
Local Regulation	0.0163	0.1463	0.0163	0.8211	0.0976	0.1382	0.3496	0.4146
Political Debate	0.0299	0.0299	0.9403	0	0.0149	0	0.9552	0.0149
Islamic Threat	0.4706	0.2059	0.2647	0.0588	0.6471	0	0.0294	0.3235
Muslim Neighbors	0.0476	0.4762	0.4405	0.0357	0	0.9881	0.0119	0
Legal Authority	0.2927	0.1951	0.1463	0.3659	0.122	0.2439	0.5854	0.0488

Note. Primary components of frames are in boldface; secondary components are shaded. Numbers indicate the means of causal attribution presence by article. Means greater than 0.45 are primary components of a cluster; means between 0.25 and 0.45 are secondary components.

The first frame, Local Regulation (n = 123), was mainly concerned with regulatory processes related to the activity of gaining approval for a mosque’s construction. Islamic community members were the main actors, in dialogue with government officials charged with granting approval. Binding moral foundations of loyalty and authority were the main moral evaluations present. Causal attributions were mostly absent in this cluster, while treatment recommendations were split between “neither” and “both” as secondary components. These results suggest that this was not an impassioned frame but rather more transactional in nature.

The second frame, Political Debate (n = 67), featured a wide dispersion of issues. Muslim community leaders were the single largest type of sources (0.68) followed closely by antimosque activists (0.60). The presence of secondary source types of local government officials, state and national political sources, and non-Muslim religious community leaders suggests that this was a debate that sparked interest among a wide variety of stakeholders. The main moral evaluations were, again, loyalty and authority, with a secondary moral component of harm. Harm is an individualizing foundation that is based on antagonism toward those who cause suffering (Graham et al., 2013). Causal attribution and treatment recommendation were both balanced between the two options (i.e., “both”). This balance of components suggests that this frame was dominated by the guiding journalistic principle of objectivity, in which an attempt is made for all sides of a debate to be represented.

The third frame, Islamic Threat (n = 34), was composed of feature stories and opponent protests. Main sources were antimosque activists, and secondary sources included state and national politicians, Muslim community leaders, and non-Muslim religious community leaders. The main moral evaluations were loyalty and authority, while harm and betrayal were secondary components. Betrayal is unique to this frame and is based on threats to the in-group (Graham et al., 2013). Because this frame clustered strongly around the idea that Islam is a political ideology and that the mosque in question should be stopped, it reflects a strong impression that Islam is an incompatible political system that poses an inherent danger to American society and that this threat must be stopped by regulation.

The fourth frame, Muslim Neighbors (n = 84), could be considered the opposite of the Islamic Threat frame. It was made up of feature stories that featured Muslim community leaders as main sources, with local government officials, promosque activists, and non-Muslim religious community leaders as secondary sources. Again, the main moral evaluations were loyalty and authority (with an additional component of harm). The main difference is that it strongly clustered around Muslim rights to free expression and the idea that the mosque should move forward. This frame suggests that Muslims would be harmed if the mosque projects were stopped and that it is important to extend full civil rights to Muslims in the United States, because not doing so would violate American values.

The fifth frame, Legal Authority (n = 41), completely clustered around legal processes and featured local government officials and antimosque activists as main sources. The moral evaluation was very strongly clustered around authority, with fairness and loyalty appearing as secondary components. The causal attribution was weakly clustered around “neither” and Islam as a political ideology. The treatment recommendation mainly clustered around “both,” suggesting editorial balance. The Legal Authority frame was very stable, appearing in roughly the same form in the four-, five-, and six-cluster solutions. This suggests that judicial processes occupy a particular kind of authoritative space in civic discourse.

The frames were not evenly dispersed among the stories in the corpus. The Local Regulation frame was the dominant frame, comprising 35.24% of the corpus. The feature story-based Muslim Neighbors frame was the next most prominent frame, comprising 24.07%. The other frames, in order of occurrence, were as follows: Political Debate (19.20%), Legal Authority (11.75%), and Islamic Threat (9.74%).

Moral Foundations Analysis

MFT connects morality to ideology, and RQ2 posed the question of which ideologies are reflected in the moral foundation profiles in the frames. The moral profiles generally do not fit the prototypical ideological profiles as identified by Haidt et al. (2009). None of the frames emphasize the care/harm and fairness/cheating foundations enough to fit the secular liberal or religious left profiles, and all the frames emphasize the socially binding foundations of loyalty and authority too much to fit the libertarian profile. The social conservative profile tends to feature a balance among all the five foundations, but all the frames in this study emphasize one foundation much more than the others, and none of them include the sanctity/degradation foundation to a large degree. In most cases, the dominant foundation is loyalty/betrayal, but in the case of the Legal Authority frame, it is authority. Overall, this suggests an ideological profile that emphasizes binding social foundations related to protecting the in-group and deference to the hierarchical authority that supports the social structure. This profile de-emphasizes the binding foundation of sanctity as well as the individualizing foundations of care and fairness. Thus, even though there are variations among the frames, these similarities across the five suggest that, in moral terms, these stories present a journalistic moral profile—and perhaps an ideology—that is unique in itself.

To further examine the ideological positions in the stories, the moral foundations of stories that clearly presented positions on one side of the debate or the other were analyzed. H1 hypothesized stories with antimosque treatment recommendations that would emphasize the moral foundations of loyalty/betrayal, authority/subversion, and sanctity/degradation. To answer that hypothesis, the moral foundations of stories with the treatment recommendation that a mosque project should be stopped were analyzed. Because the dependent variable was discrete, binary logistic regression was conducted. The dependent variable of treatment recommendation was recoded equal to 1 if the mosque should be stopped, with all other cases coded 0.

The results of the binary logistic regression (displayed in Table 5) show that only one moral foundation, betrayal, was a significant predictor of mosque opposition at the 0.05 level. The odds ratio for betrayal is 4.24 with a 95% confidence interval of [1.01, 17.79]. Because this was one of the foundations hypothesized to be a significant predictor, it could be considered as a weak support for the hypothesis. But with the absence of any other expected variable as a significant predictor, H1 is not supported overall.

Table 5.

Logistic Regression for Antimosque Articles.

Variables in the equation	B	SE	Wald	df	Significance	Exp(B)	95% CI for Exp(B)
Variables in the equation	B	SE	Wald	df	Significance	Exp(B)	Lower	Upper
Step 1a
Care	0.293	0.597	0.241	1	0.623	1.341	0.416	4.32
Harm	−0.885	0.661	1.792	1	0.181	0.413	0.113	1.508
Fairness	0.368	0.719	0.261	1	0.609	1.444	0.353	5.916
Cheating	1.048	0.892	1.38	1	0.24	2.851	0.497	16.371
Loyalty	−0.015	0.328	0.002	1	0.963	0.985	0.518	1.873
Betrayal	1.444	0.732	3.888	1	0.049	4.236	1.009	17.789
Authority	−0.534	0.402	1.765	1	0.184	0.586	0.267	1.289
Subversion	0.535	1.202	0.198	1	0.656	1.708	0.162	18.001
Sanctity	−1.235	0.918	1.807	1	0.179	0.291	0.048	1.76
Degradation	0.154	2.197	0.005	1	0.944	1.167	0.016	86.431
Constant	−1.868	0.432	18.675	1	0	0.154

Note. CI = confidence interval.

The analysis of H2 is the inverse of H1. The dependent variable of treatment recommendation was equal to 1 if the mosque should be allowed, with all other cases coded 0. This hypothesis posited that promosque stories would emphasize the moral dimensions of care/harm and fairness/cheating. The results of this binary logistic regression are displayed in Table 6. These results show two variables were significant predictors: care (0.001) and cheating (0.038). The odds ratio for care was 4.867 with a 95% confidence interval of [1.86, 12.64]; the odds ratio for cheating was 4.027 with a 95% confidence interval of [1.081, 15.001]. Because the two significant predictors were from the hypothesized moral foundations, H2 is supported.

Table 6.

Logistic Regression for Promosque Articles.

Variables in the equation	B	SE	Wald	df	Significance	Exp(B)	95% CI for Exp(B)
Variables in the equation	B	SE	Wald	df	Significance	Exp(B)	Lower	Upper
Step 1a
Care	1.582	0.487	10.57	1	0.001	4.867	1.875	12.635
Harm	0.213	0.411	0.268	1	0.605	1.237	0.553	2.77
Fairness	0.356	0.528	0.454	1	0.5	1.428	0.507	4.022
Cheating	1.393	0.671	4.311	1	0.038	4.027	1.081	15.001
Loyalty	0.213	0.213	1.002	1	0.317	1.237	0.815	1.877
Betrayal	0.899	0.574	2.449	1	0.118	2.457	0.797	7.576
Authority	0.49	0.252	3.785	1	0.052	1.632	0.996	2.673
Subversion	−0.489	0.962	0.258	1	0.611	0.613	0.093	4.043
Sanctity	0.823	0.436	3.558	1	0.059	2.277	0.968	5.355
Degradation	0.58	1.365	0.181	1	0.671	1.787	0.123	25.956
Constant	−2.014	0.334	36.394	1	0	0.133

Note. CI = confidence interval.

Using a chi-square test of independence with a threshold value of 0.25% foundation words referenced in the articles (indicating at least a secondary foundation), a binary present/nonpresent categorization separated articles accordingly (Table 7). After testing, there was a significant difference between frames when looking at the presence of each of the foundations. This implies that the proportions of present versus nonpresent foundations differed between frames.

Table 7.

Chi-Square Tests of Moral Foundations.

	Chi-square	p value	Cramer’s V
Care/harm	36.4863	2.30E-07	0.3233**
Fairness/cheating	17.6752	1.43E-03	0.2250**
Loyalty/betrayal	50.3753	3.01E-10	0.3799**
Authority/subversion	12.7906	1.23E-02	0.1914*
Sanctity/degradation	13.0225	1.12E-02	0.1932*

p < .05. **p < .005.

Discussion

This study examined the coverage of mosque-building controversies to understand what physical and metaphorical space Muslims are able to occupy in contemporary America. A cluster analysis of frame elements suggested five frames with differing patterns of problem definitions, causal attributions, moral evaluations, and treatment recommendations. This array of multiple and oppositional frames is not surprising—in fact, it is what one should expect if approaching framing from a constructionist perspective (Vliegenthart & van Zoonen, 2011). Nor should these different frames be considered in isolation from one another. They occurred simultaneously and work together to influence audience understandings. For example, a newspaper could publish a Local Regulation story one day and a Muslim Neighbors story the next, with audience interpretations building cumulatively from previous articles and other sources of knowledge (such as family members and other individual contacts).

Moreover, these frames did not occur in equal proportions. The most frequently occurring frame—the Local Regulation frame—contained 123 articles, but its dominant concern with technical matters rendered it fairly unexciting overall. On the contrary, the least occurring frame was the Islamic Threat frame. Made up of 34 stories, this frame accounted for less than 10% of the study population. Although the Islamic Threat frame was the smallest, it was also the most negative, contrasted with the Muslim Neighbors frame, which was more positive and included more than double the number of stories. However, one should be careful in equating the sheer number of articles in a frame with frame strength. In agenda setting research, a growing body of literature supports the idea that negatively valenced information has stronger effects than positive information (Coleman & Wu, 2010; Ragas & Kiousis, 2010; Wu & Coleman, 2009). This cuts to the very heart of the question of whether it is sufficient to rely solely on quantitative analysis, measuring frames by “the sheer weight of accumulated sentences” (Reese, 2001, p. 17). Without accounting for the relative weight of positive versus negative information, the answer may be no.

The finding that moral foundations were only predictive of promosque positions suggests that mosque opponents did not base their arguments primarily on moral evaluations, but rather may have emphasized other framing components. In particular, causal attribution and treatment recommendation may have played a key framing role in those stories. Many of the antimosque stories argued that Islam is a political ideology that poses some sort of existential threat to the American way of life. This suggests an emphasis on the causal attribution variable. Moreover, this interpretation is consistent with the one moral foundation that was significant in these stories—betrayal. This finding reinforces one of the broader assumptions underpinning this research, which is that all the framing components are important when conducting research using the cluster technique. Researchers should continue to improve the theoretical grounding for these components. Also, perhaps the five original MFT dimensions are insufficient to capture the entire moral range of the debate. Validating the other foundations—particularly liberty/oppression, efficiency/waste, ownership/theft, and honesty/deception (Graham et al., 2013)—would allow researchers to discern more nuanced moral reasoning.

This study cannot directly answer questions about the agenda setting or priming effects of negatively valenced coverage, and the constructionist paradigm in which it is based raises doubts about whether we can isolate the effects of a particular group of articles outside the cultural milieu in which it is embedded. But it may stand that infrequent but particularly negative frames can set the tone of a debate and influence audience understandings of more neutral information. For example a Legal Authority story could thus be read from either an Islamic Threat or a Muslim Neighbors perspective, with audience members drawing widely divergent interpretations depending on which point of view is more salient. These selections are consequential because, as Oliver, Dillard, Bae, and Tamul (2012) found, the way in which news stories are formatted can increase compassion and empathy toward stigmatized groups.

The Binding Role of Journalism

The addition of MFT allows some important insights for interpreting frames and understanding the role journalism plays in fostering social cohesion. The foundations identified by MFT can be broadly divided into two groups—“individualizing” foundations that are related with personal liberation and “binding” foundations that are concerned with the protection of group cohesion (Graham et al., 2009; Haidt, 2008). Overall, the frames emphasized socially binding foundations of loyalty/betrayal and authority/subversion.

Generally speaking, the loyalty foundation relates to notions of allegiance to one’s in-group (at the expense of an opposing out-group; Graham et al., 2013). This in-group loyalty has evolutionary roots in the competition between groups over scarce resources (Graham et al., 2013; Haidt et al., 2009). Muslims are frequently depicted in an out-group position in American media coverage (Hoewe, Bowe, & Makhadmeh, 2014; Ibrahim, 2010). This was also true in these frames, in which mosque opponents often presented Muslims as affiliated with a terroristic political movement aimed at destroying the homeland. Yet there was a loyalty-based counter-narrative offered as well, in which mosque opponents were framed as agents of intolerance, violating many of the United States’ most deeply held principles of pluralism and free expression. In both cases, protecting the values that bind Americans together as a people was emphasized. The second binding foundation stressed, authority, is related to the legitimacy of institutions and the maintenance of social order through the establishment of hierarchical relationships characterized by obedience and deference to superiors (Graham et al., 2013). Therefore, it is no surprise that the Legal Authority frame placed the most emphasis on this concept of all the frames.

Although individualizing foundations were not main the components of any of the frames, the binary logistic regression results showed that care and cheating were significant predictors of mosque support. Care includes concepts such as safety, peace, security, and protecting; cheating encompasses bias, injustice, bigotry, and discrimination. Stories with a treatment recommendation in support of mosque construction endorsed the desire of Muslim communities to have a safe place to worship and portrayed those in opposition as forces of prejudice or segregation.

Harm was a secondary component of three of the frames (Political Debate, Islamic Threat, and Muslim Neighbors) and encompasses concepts such as war, violence, killing, and attacks. The reverberations of both the 9/11 attacks and the wars in Muslim-majority nations Iraq and Afghanistan were frequently referenced in this corpus—even though these broader issues have little to do with local zoning questions about parking or building heights. This continued association of Islam with the topics of terrorism and war may be so salient for some audience members as to drown out other associations. Further differentiating the Legal Authority frame from the others, this was the only frame in which the individualizing foundation of fairness is a secondary component, suggesting that legal solutions must not only uphold in-group societal values, but they also must be fair to the parties involved in order to be perceived as just.

Implications for Journalists

It is important remember that these frames are the product of a dynamic and socially constructed discourse in which journalists play an official role. Some of the core elements of the practice of journalism include the ability to define a community and identify its heroes and villains (Kovach & Rosenstiel, 2007), which may be reflected in the emphasis of socially binding moral foundations in these stories. This preference for community-building moral foundations may not be surprising when considering the traditional function of journalism in American society. This function dictates that American journalists work to inform the public, and in turn, those informed citizens have a responsibility to the greater community (Tuchman, 1980). Considering how embedded media institutions are in the functioning of that social order, perhaps news content at its core serves a binding function.

The civic function of promoting social order and an informed electorate is at the center of American journalistic practice, and this function is reflected in the moral evaluations in these frames. This study, however, cannot untangle whether these were individual moral decisions made by journalists or moral positions imposed upon news content by the media routines and professional practices. In either case, the current state of reporting on Islam has led to calls from some scholars to improve the training for journalists on these issues. Such training would help journalists balance legitimate concerns about terrorism with respect for religious freedom.

Journalists have an important role to play in helping Muslims and non-Muslim communities build trust in one another by adhering to the Society of Professional Journalists’ (2014, para. 17) Code of Ethics’ exhortation to “Boldly tell the story of the diversity and magnitude of the human experience” and “Seek sources whose voices we seldom hear.” Some journalists have called for a cultural change in newsrooms that would allow for a more nuanced presentation of Muslims in a variety of situations, rather than reflexively connecting them to conflict (Ewart et al., 2016). Reflecting this call for a cultural shift, recent years have seen the development of a variety of scholarly resources to improve training for journalists reporting on Islam and Muslims (e.g., Ewart & Pearson, 2016; Hoewe, Bowe, & Zeldes, 2012; Michigan State University School of Journalism, 2014; Pintak & Franklin, 2013).

Limitations and Future Research

This research makes a contribution to framing theory using MFT to operationalize moral evaluations, but like all research, it contains limitations. The results will need to be squared with other studies of moral foundations in public discourse. Although the foundations of loyalty and authority were strongly emphasized in controversies about mosque construction, Bowman, Lewis, and Tamborini (2014) found ideologically related differences in moral foundations in a content analysis of headlines related to the killing of Osama bin Laden in which conservative-region newspapers emphasized authority and loyalty and liberal-region papers emphasized fairness and cheating (or, as they referred to it, reciprocity). Unlike that study, this research did not consider regional differences. Furthermore, because of their brevity, it could hold that headlines are more starkly framed because there is no room to portray nuances or multiple positions. Also, those studies only considered moral foundations and did not examine the other components of framing. Perhaps moral foundations group differently when incorporated into a full framing analysis than they do when considered in isolation.

A geographically based analysis is beyond the scope of this study, but researchers should analyze where there are regional differences in the coverage of American Muslims—particularly whether the presence of a large Muslim population in a community has an influence on the coverage. Using a community structure approach, Chang (2016) found increased support for undocumented immigrants in news editorials produced in communities that were more pluralistic and less politically or religiously conservative. This type of regional difference may also extend to other groups perceived as “outsiders.” Researchers should examine whether more cosmopolitan communities with large Muslim populations may have more sympathetic coverage of Muslims.

Furthermore, the results of this story cannot be generalized to coverage of Muslims broadly. Controversies about mosque construction in the post-9/11 era may be a topic in which moral evaluations play a unique role. Future researchers should expand this type of research into different arenas to see whether Muslims are covered differently in other topics.

Finally, the relatively low intercoder reliability for the treatment recommendation variable and the lack of an established theoretical grounding for it suggest that future researchers should turn their attention there next. Different framing components can be dominant across discourse on a single topic, which means that all the components in the cluster technique must be theoretically justified to be meaningful and valid. Such improvements to the theoretical and methodological rigor of framing would go a long way toward resolving fears that framing runs the risk of becoming an unfocused catch-all term or a “passe-partout” (Reese, 2007; Van Gorp, 2007).

Conclusion

Considered alongside other studies of media coverage of Islam, this research bolsters arguments that the position of Muslims within American pluralism is still conflicted. Ibrahim (2010) found that American Muslims were portrayed as a most peaceful part of an American in-group, while foreign Muslims were portrayed as violent jihadis in out-group terms. However, Hoewe et al. (2014) found that American Muslims were consistently portrayed as members of an out-group by associating them with references to foreign nations in a decade of coverage of Shariah. Moreover, Bowe and Hoewe (2016) found that supporters of an amendment to ban Shariah from use by courts were more likely to frame their positions in moral terms than were opponents of amendment.

This study analyzed the ways Islam is framed in several mainstream print media outlets in an attempt to understand the socially constructed position the faith holds in the American mediascape. As media frames are an important part of both reflecting and creating culture, they cannot be disassociated from their cultural milieu. Religion remains important in the United States, with around 70% of Americans professing to be either very or moderately religious (Newport, 2014). At the same time, Americans are expected to give precedence to conforming to national values over religious commitments (Haddad & Ricks, 2009).

In the interpretation of many Americans, Islam represents “a unified and unchanging codex of religious, juridical and moral rules, supposedly regulating the life of both the individual Muslim and the Islamic community in all its aspects” (Jung, 2010, p. 16). This essentialist view sets up a cultural conflict in which Muslim loyalty to the United States is always under suspicion. This distrust was a major feature of the debates surrounding these mosque proposals. It is so embedded that, even after the U.S. Department of Justice and multiple judges issued rulings that Islam is a religion deserving First Amendment protections, mosque opponents continued to be quoted in stories questioning Islam’s status as a legitimate faith. Portrayed as such, it remained an ever-open question, immune to any authoritative resolution.

These varying results might change over time or be related to the specific contours of a debate. However, the in-group/out-group distinction is key, where Muslims are “always required to perform and thereby prove their loyalty” by disassociating themselves from international conflicts to become a part of the in-group (Morey & Yaqin, 2011, p. 88). This research shows that this de facto out-group status remains one important part of the dynamic in the conversation surrounding American Islam—even in routine bureaucratic matters like zoning.

Encouraging a news discourse that does not reflexively connect all Islam-related stories to festering global conflicts may be one small but important step in challenging that out-group status and reducing societal tensions. As Pintak (2016, para. 22) noted,

This isn’t about political correctness, it’s about differentiating the many threats and winning allies in the war against extremism at home and abroad. . . . Because at the end of the day, words shape perceptions, which shape policy, which often determines whether people live or die.

Footnotes

Declaration of Conflicting Interests

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

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research was supported with funding from the College of Communication Arts & Sciences and the Graduate School of Michigan State University.

Author Biography

Brian J. Bowe (PhD, Michigan State University) is an assistant professor of journalism at Western Washington University. He researches the connections between media and culture in multiple contexts. His work focuses on framing and agenda setting, with particular focus on media coverage of Islam. He has taught journalism and communications at Grand Valley State University, the Sorbonne University, and Michigan State University. His research has been published in journals such as Public Understanding of Science, Journalism, and Media, Culture & Society. He has written several books about popular music, including biographies of The Ramones, The Clash, and Judas Priest. He coedited the anthology CREEM: America’s Only Rock ‘N’ Roll Magazine (2007, Collins).

References

Ahmed

Matthes

(2017). Media representation of Muslims and Islam from 2000 to 2015: A meta-analysis. International Communication Gazette, 79, 219-244. doi:10.1177/1748048516656305

Association of Religion Data Archives. (n.d.). Religious adherents, 2010. Retrieved from http://www.thearda.com/internationalData/countries/Country_234_2.asp

Bagby

(2009). The American mosque in transition: Assimilation, acculturation and isolation. Journal of Ethnic and Migration Studies, 35, 473-490. doi:10.1080/13691830802704640

Bagby

(2012). The American mosque 2011 (No. 1). Washington, DC: Council on American–Islamic Relations. Retrieved from http://www.cair.com/images/pdf/The-American-Mosque-2011-part-1.pdf

Bantimaroudis

Kampanellou

(2007). The cultural framing hypothesis: Attributes of cultural alliances and conflicts. The Harvard International Journal of Press/Politics, 12, 80-90. doi:10.1177/1081180X07299793

Bell

C. V.

Entman

R. M.

(2011). The media’s role in America’s exceptional politics of inequality: Framing the vash tax cuts of 2001 and 2003. The International Journal of Press/Politics, 16, 548-572. doi:10.1177/1940161211417334

Bowe

B. J.

Hoewe

(2016). Night and day: An illustration of framing and moral foundations in the Oklahoma Shariah amendment campaign. Journalism & Mass Communication Quarterly, 93, 967-985. doi:10.1177/1077699016628806

Bowe

B. J.

Oshita

Terracina-Hartman

Chao

W.-C.

(2012). Framing of climate change in newspaper coverage of the East Anglia e-mail scandal. Public Understanding of Science, 23, 157-169. doi:10.1177/0963662512449949

Bowman

Lewis

R. J.

Tamborini

(2014). The morality of May 2, 2011: A content analysis of U.S. headlines regarding the death of Osama bin Laden. Mass Communication and Society, 17, 639-664. doi:10.1080/15205436.2013.822518

10.

Brossard

Shanahan

McComas

(2004). Are issue-cycles culturally constructed? A comparison of French and American coverage of global climate change. Mass Communication and Society, 7, 359-377.

11.

Bryant

Miron

(2004). Theory and research in mass communication. Journal of Communication, 54, 662-704. doi:10.1111/j.1460-2466.2004.tb02650.x

12.

Cacciatore

M. A.

Scheufele

D. A.

Iyengar

(2016). The end of framing as we know it . . . and the future of media effects. Mass Communication and Society, 19, 7-23. doi:10.1080/15205436.2015.1068811

13.

Calfano

B. R.

Djupe

P. A.

Cox

Jones

(2016). Muslim mistrust: The resilience of negative public attitudes after complimentary information. Journal of Media and Religion, 15, 29-42. doi:10.1080/15348423.2015.1131041

14.

Chang

(2016). Framing of the immigration reform in 2006: A community structure approach. Journalism & Mass Communication Quarterly, 92, 836-856. doi:10.1177/1077699015594462

15.

Clifford

Jerit

(2013). How words do the work of politics: Moral foundations theory and the debate over stem cell research. The Journal of Politics, 75, 659-671. doi:10.1017/S0022381613000492

16.

Coleman

H. D.

(2010). Proposing emotion as a dimension of affective agenda setting: Separating affect into two components and comparing their second-level effects. Journalism & Mass Communication Quarterly, 87, 315-327. doi:10.1177/107769901008700206

17.

Cooper

(2017). Americans who interact with Muslims hold more positive views of Muslims, refugees. Public Religion Research Institute. Retrieved from http://www.prri.org/spotlight/social-contact-muslims-refugee-ban/

18.

Cox

Jones

R. P.

(2015). Nearly half of Americans worried that they or their family will be a victim of terrorism. Public Religion Research Institute. Retrieved from http://www.prri.org/research/survey-nearly-half-of-americans-worried-that-they-or-their-family-will-be-a-victim-of-terrorism/

19.

D’Angelo

Kuypers

J. A.

(2010). Introduction: Doing news framing analysis. In D’Angelo

Kuypers

J. A.

(Eds.), Doing news framing analysis: Empirical and theoretical perspectives (pp. 1-13). New York, NY: Routledge.

20.

David

C. C.

Atun

J. M.

Fille

Monterola

(2011). Finding frames: Comparing two methods of frame analysis. Communication Methods and Measures, 5, 329-351. doi:10.1080/19312458.2011.624873

21.

Dehghani

Sagae

Sachdeva

Gratch

(2014). Analyzing political rhetoric in conservative and liberal weblogs related to the construction of the “Ground Zero Mosque.” Journal of Information Technology & Politics, 11, 1-14. doi:10.1080/19331681.2013.826613

22.

de Vreese

C. H

. (2005). News framing: Theory and typology. Information Design Journal & Document Design, 13(1), 51-62.

23.

Donk

Metag

Kohring

Marcinkowski

(2012). Framing emerging technologies risk perceptions of nanotechnology in the German press. Science Communication, 34, 5-29. doi:10.1177/1075547011417892

24.

Entman

R. M.

(1993). Framing: Toward clarification of a fractured paradigm. Journal of Communication, 43(4), 51-58. doi:10.1111/j.1460-2466.1993.tb01304.x

25.

Entman

R. M.

(2003). Cascading activation: Contesting the White House’s frame after 9/11. Political Communication, 20, 415-432. doi:10.1080/10584600390244176

26.

Entman

R. M.

(2004). Projections of power: Framing news, public opinion, and U.S. foreign policy. Chicago, IL: The University of Chicago Press.

27.

Ewart

Pearson

(2016). Reporting Islam: International best practice for journalists. Brisbane, Australia: Griffith University. Retrieved from https://reportingislamdotorg.files.wordpress.com/2016/09/reportingislamreportagehandbookv012016.pdf

28.

Ewart

Pearson

Healy

(2016). Journalists’ and educators’ perspectives on news media reporting of Islam and Muslim communities in Australia and New Zealand. Journal of Media and Religion, 15, 136-145. doi:10.1080/15348423.2016.1209391

29.

Feinberg

Willer

(2013). The moral roots of environmental attitudes. Psychological Science, 24, 56-62. doi:10.1177/0956797612449177

30.

Gallup

Inc

. (n.d.). Islamophobia: Understanding anti-Muslim sentiment in the West. Retrieved from http://www.gallup.com/poll/157082/islamophobia-understanding-anti-muslim-sentiment-west.aspx

31.

Gallup Center for Muslim Studies. (2010, January). Religious perceptions in America: With an in-depth analysis of U.S. attitudes toward Muslims and Islam. Washington, DC: Gallup.

32.

Gamson

W. A.

(2001). Foreword. In Reese

Gandy

O. H.

Grant

A. E.

(Eds.), Framing public life (pp. ix-xvi). Mahwah, NJ: Lawrence Erlbaum.

33.

Garten

Boghrati

Hoover

Johnson

K. M.

Dehghani

(2016). Morality between the lines: Detecting moral sentiment in text. Proceedings of IJCAI 2016 workshop on Computational Modeling of Attitudes, New York, NY. Retrieved from http://morteza-dehghani.net/wp-content/uploads/morality-lines-detecting.pdf

34.

Gitlin

(1980). The whole world is watching: Mass media in the making & unmaking of the New Left (Rev. 2003 ed.). Berkeley: University of California Press.

35.

Goffman

(1974). Frame analysis: An essay on the organization of experience. Boston, MA: Northeastern University Press.

36.

Graham

Haidt

Koleva

Motyl

Iyer

Wojcik

S. P.

Ditto

P. H.

(2013). Moral foundations theory: The pragmatic validity of moral pluralism. In Devine

Plant

(Eds.), Advances in experimental social psychology (Vol. 47, pp. 55-130). Burlington, VT: Academic Press.

37.

Graham

Haidt

Nosek

B. A.

(2009). Liberals and conservatives rely on different sets of moral foundations. Journal of Personality and Social Psychology, 96, 1029-1046. doi:10.1037/a0015141

38.

Graham

Nosek

B. A.

Haidt

(2012). The moral stereotypes of liberals and conservatives: Exaggeration of differences across the political spectrum. PLoS ONE, 7(12), Article e50092. doi:10.1371/journal.pone.0050092

39.

Haddad

Y. Y.

Ricks

(2009). Claiming space in America’s pluralism: Muslims enter the political maelstrom. In Sinno

A. H.

(Ed.), Muslims in Western politics (pp. 13-34). Bloomington: Indiana University Press.

40.

Haidt

(2008). Morality. Perspectives on Psychological Science, 3, 65-72. doi:10.1111/j.1745-6916.2008.00063.x

41.

Haidt

Graham

(2007). When morality opposes justice: Conservatives have moral intuitions that liberals may not recognize. Social Justice Research, 20, 98-116. doi:10.1007/s11211-007-0034-z

42.

Haidt

Graham

Joseph

(2009). Above and below left-right: Ideological narratives and moral foundations. Psychological Inquiry, 20, 110-119. doi:10.1080/10478400903028573

43.

Haidt

Joseph

(2004). Intuitive ethics: How innately prepared intuitions generate culturally variable virtues. Daedalus, 133(4), 55-66.

44.

Heimlich

(2011). America’s Muslim population 2030. Retrieved from http://www.pewresearch.org/fact-tank/2011/03/02/americas-muslim-population-2030/

45.

Hoewe

Bowe

B. J.

Makhadmeh

(2014). Broadcasting Sharia: American TV news’ illustration of social identity and the emergence of a threat. Journal of Media and Religion, 13, 67-81. doi:10.1080/15348423.2014.909195

46.

Hoewe

Bowe

B. J.

Zeldes

G. A.

(2012). Using a wiki to produce journalistic best practices. Communication Teacher, 26, 22-32. doi:10.1080/17404622.2011.625365

47.

Holcomb

(2010). Elections, Katrina and economy split the news agenda. Retrieved from http://www.journalism.org/2010/08/30/pej-news-coverage-index-august-2329-2010/

48.

Ibrahim

(2010). The framing of Islam on network news following the September 11th attacks. International Communication Gazette, 72, 111-125.

49.

Iyengar

(1996). Framing responsibility for political issues. The ANNALS of the American Academy of Political and Social Science, 546, 59-70.

50.

Jairam

(2012). Applying moral theory in practice: Frequently violated moral domains in banned advertisements (Master’s thesis). The University of Texas at Austin.

51.

Jung

(2010). The origin of difference: Edward Said, Michel Foucault and the modern image of Islam. In Ismael

T. Y.

Rippin

(Eds.), Islam in the eyes of the West: Images and realities in an age of terror. London, England: Routledge.

52.

Kohut

(2007). Muslim Americans: Middle class and mostly mainstream. Washington, DC: Pew Research Center.

53.

Kovach

Rosenstiel

(2007). The elements of journalism: What newspeople should know and the public should expect (Compl. updated and Rev. ed.). New York, NY: Three Rivers Press.

54.

Krippendorff

(2011). Agreement and information in the reliability of coding. Communication Methods and Measures, 5, 93-112. doi:10.1080/19312458.2011.568376

55.

Lacy

Watson

B. R.

Riffe

Lovejoy

(2015). Issues and best practices in content analysis. Journalism & Mass Communication Quarterly, 92, 791-811. doi:10.1177/1077699015607338

56.

Lewis

S. C.

Zamith

Hermida

(2013). Content analysis in an era of big data: A hybrid approach to computational and manual methods. Journal of Broadcasting & Electronic Media, 57, 34-52. doi:10.1080/08838151.2012.761702

57.

Lipka

(2015). The political divide on views toward Muslims and Islam. Retrieved from http://www.pewresearch.org/fact-tank/2015/01/29/the-political-divide-on-views-toward-muslims-and-islam/

58.

Liu

(2012a). Controversies over mosques and Islamic centers across the U.S. Retrieved from http://www.pewforum.org/2012/09/27/controversies-over-mosques-and-islamic-centers-across-the-u-s-2/

59.

Liu

(2012b). Religion in the news: Islam and politics dominate religion coverage in 2011. Retrieved from http://www.pewforum.org/2012/02/23/religion-in-the-news-islam-and-politics-dominate-religion-coverage-in-2011/

60.

Lombard

Snyder-Duch

Bracken

C. C.

(2002). Content analysis in mass communication: Assessment and reporting of intercoder reliability. Human Communication Research, 28, 587-604.

61.

Matthes

(2009). What’s in a frame? A content analysis of media framing studies in the world’s leading communication journals, 1990-2005. Journalism & Mass Communication Quarterly, 86, 349-367.

62.

Matthes

Kohring

(2008). The content analysis of media frames: Toward improving reliability and validity. Journal of Communication, 58, 258-279. doi:10.1111/j.1460-2466.2008.00384.x

63.

Michigan State University School of Journalism. (2014). 100 questions and answers about Muslim Americans with a guide to Islamic holidays. East Lansing, MI: David Crumm Media.

64.

Mohamed

(2016). A new estimate of the U.S. Muslim population. Retrieved from http://www.pewresearch.org/fact-tank/2016/01/06/a-new-estimate-of-the-u-s-muslim-population/

65.

Mooi

Sarstedt

(2010). Cluster analysis. In Mooi

Sarstedt

(Eds.), A concise guide to market research (pp. 237-284). Berlin, Germany: Springer.

66.

Morey

Yaqin

(2011). Framing Muslims: Stereotyping and representation after 9/11. Cambridge, MA: Harvard University Press.

67.

Morgan

(2010a). Koran and cultural center put Islam in the news. Retrieved from http://www.journalism.org/2010/09/13/pej-news-index-report-september-612-2010/

68.

Morgan

(2010b). Mosque controversy, Iraq war dominate the news. Retrieved from http://www.journalism.org/2010/08/23/mosque-controversy-iraq-war-dominate-news/

69.

Motyl

(2012). Party evolutions in moral intuitions: A text-analysis of US political party platforms from 1856-2008 (SSRN Scholarly Paper No. ID 2158893). Rochester, NY: Social Science Research Network. Retrieved from https://papers.ssrn.com/abstract=2158893

70.

Nacos

B. L.

Torres-Reyna

(2006). Fueling our fears: Stereotyping, media coverage, and public opinion of Muslim Americans. New York, NY: Rowman & Littlefield.

71.

Newport

(2014). Mississippi most religious state, Vermont least religious. Retrieved from http://www.gallup.com/poll/167267/mississippi-religious-vermont-least-religious-state.aspx

72.

Newport

(2015, September 22). Six in 10 Americans would say “yes” to Muslim President. Gallup. Retrieved from http://www.gallup.com/opinion/polling-matters/185813/six-americans-say-yes-muslim-president.aspx

73.

Nisbet

E. C.

Ostman

Shanahan

(2009). Public opinion toward Muslim Americans: Civil liberties and the role of religiosity, ideology, and media use. In Sinno

A. H.

(Ed.), Muslims in Western politics (pp. 161-199). Bloomington: Indiana University Press.

74.

Norris

Inglehart

R. F.

(2012). Muslim integration into Western cultures: Between origins and destinations. Political Studies, 60, 228-251. doi:10.1111/j.1467-9248.2012.00951.x

75.

Oliver

M. B.

Dillard

J. P.

Bae

Tamul

D. J.

(2012). The effect of narrative news format on empathy for stigmatized groups. Journalism & Mass Communication Quarterly, 89, 205-224.

76.

Pan

Kosicki

G. M.

(1993). Framing analysis: An approach to news discourse. Political Communication, 10, 55-75. doi:10.1080/10584609.1993.9962963

77.

Panagopoulos

(2006). Trends: Arab and Muslim Americans and Islam in the aftermath of 9/11. The Public Opinion Quarterly, 70, 608-624.

78.

Pew Forum on Religion & Public Life. (2009). Mapping the Global Muslim population: A report on the size and distribution of the world’s Muslim population. Washington, DC: Pew Research Center.

79.

Pew Research Center. (2011). Muslim Americans: No signs of growth in alienation or support for extremism. Retrieved from http://www.people-press.org/files/legacy-pdf/Muslim%20American%20Report%2010-02-12%20fix.pdf

80.

Pew Research Center. (2014). How Americans feel about religious groups. Retrieved from http://www.pewforum.org/files/2014/07/Views-of-Religious-Groups-07-27-full-PDF-for-web.pdf

81.

Pew Research Center. (2015). America’s changing religious landscape. Washington, DC: Author. Retrieved from http://www.pewforum.org/2015/05/12/americas-changing-religious-landscape/

82.

Pintak

(2016, July 19). Not all Islamists are out to kill us. Foreign Policy. Retrieved from https://foreignpolicy.com/2016/07/19/not-all-islamists-are-out-to-kill-us-united-states-trump-2016-isis/

83.

Pintak

Franklin

(Eds.). (2013). Islam for journalists: A primer on covering Muslim communities in America. Retrieved from https://www.rjionline.org/downloads/islam-for-journalists

84.

Poynting

Mason

(2007). The resistible rise of Islamophobia. Journal of Sociology, 43, 61-86. doi:10.1177/1440783307073935

85.

Poynting

Noble

(2003). “Dog-whistle” journalism and Muslim Australians since 2001. Media International Australia, Incorporating Culture & Policy, 109, 41-49.

86.

Ragas

M. W.

Kiousis

(2010). Intermedia agenda-setting and political activism: MoveOn.org and the 2008 presidential election. Mass Communication and Society, 13, 560-583. doi:10.1080/15205436.2010.515372

87.

Reese

S. D.

(2001). Prologue—Framing public life: A bridging model for media research. In Reese

S. D.

Gandy

O. H.

Grant

A. E.

(Eds.), Framing public life (pp. 7-31). Mahwah, NJ: Lawrence Erlbaum.

88.

Reese

S. D.

(2007). The framing project: A bridging model for media research revisited. Journal of Communication, 57, 148-154. doi:10.1111/j.1460-2466.2006.00334.x

89.

Sagi

Dehghani

(2014). Measuring moral rhetoric in text. Social Science Computer Review, 32, 132-144. doi:10.1177/0894439313506837

90.

Saha

(2012). “Beards, scarves, halal meat, terrorists, forced marriage”: Television industries and the production of “race.” Media, Culture & Society, 34, 424-438. doi:10.1177/0163443711436356

91.

Said

(1997). Covering Islam: How the media and the experts determine how we see the rest of the world (Rev. ed.). New York, NY: Vintage Books.

92.

Scheufele

B. T.

Scheufele

D. A.

(2010). Of spreading activation, applicability, and schemas. In D’Angelo

Kuypers

J. A.

(Eds.), Doing news framing analysis: Empirical and theoretical perspectives (pp. 135-158). New York, NY: Routledge.

93.

Scheufele

D. A.

Nisbet

M. C.

Ostman

R. E.

(2005). September 11 news coverage, public opinion, and support for civil liberties. Mass Communication and Society, 8, 197-218.

94.

Scott

(2010). The politics of the veil. Princeton, NJ: Princeton University Press.

95.

Semetko

H. A.

Valkenburg

P. M.

(2000). Framing European politics: A content analysis of press and television news. Journal of Communication, 50, 93-109.

96.

Shaheen

J. G.

(2001). Reel bad Arabs: How Hollywood vilifies a people. New York, NY: Olive Branch Press.

97.

Shehata

(2007). Facing the Muhammad cartoons: Official dominance and event-driven news in Swedish and American elite press. The Harvard International Journal of Press/Politics, 12, 131-153. doi:10.1177/1081180X07307869

98.

Society of Professional Journalists. (2014). SPJ code of ethics. Retrieved from http://www.spj.org/ethicscode.asp

99.

Tankard

J. W.

(2001). The empirical approach to the study of media framing. In Reese

Gandy

O. H.

Grant

A. E.

(Eds.), Framing public life (pp. 95-106). Mahwah, NJ: Lawrence Erlbaum.

100.

Teernstra

van der Putten

Noordegraaf-Eelens

Verbeek

(2016). The morality machine: Tracking moral values in Tweets. In H. Boström, A. Knobbe, C. Soares, & P. Papapetrou (Eds.), Advances in intelligent data analysis XV (pp. 26-37). Stockholm, Sweden: Springer.

101.

Tuchman

(1980). Making news: A study in the construction of reality. New York, NY: Free Press.

102.

Van Gorp

. (2007). The constructionist approach to framing: Bringing culture back in. Journal of Communication, 57, 60-78. doi:10.1111/j.0021-9916.2007.00329.x

103.

Van Gorp

. (2010). Strategies to take subjectivity out of framing analysis. In D’Angelo

Kuypers

J. A.

(Eds.), Doing news framing analysis: Empirical and theoretical perspectives (pp. 84-109). New York, NY: Routledge.

104.

Vliegenthart

van Zoonen

(2011). Power to the frame: Bringing sociology back to frame analysis. European Journal of Communication, 26, 101-115. doi:10.1177/0267323111404838

105.

Ward

J. H.

(1963). Hierarchical grouping to optimize an objective function. Journal of the American Statistical Association, 58, 236-244.

106.

Wormald

(2014). How Americans feel about religious groups. Retrieved from http://www.pewforum.org/2014/07/16/how-americans-feel-about-religious-groups/

107.

H. D.

Coleman

(2009). Advancing agenda-setting theory: The comparative strength and new contingent conditions of the two levels of agenda-setting effects. Journalism & Mass Communication Quarterly, 86, 775-789.

108.

Younis

(2009, March). Muslim Americans exemplify diversity, potential. Gallup. Retrieved from http://www.gallup.com/poll/116260/Muslim-Americans-Exemplify-Diversity-Potential.aspx

109.

Zamith

Lewis

S. C.

(2015). Content analysis and the algorithmic coder: What computational social science means for traditional modes of media analysis. The ANNALS of the American Academy of Political and Social Science, 659, 307-318. doi:10.1177/0002716215570576