Who Is Concerned about Terrorist Attacks? A Religious Profile

: As part of the study on the psychological impact of terrorist acts on ordinary people, the objective of this study is to understand if religious identity protects individuals from feeling concerned about the possibility of terrorist attacks. The study was based on a sample from the World Values Survey, wave 6 (2010–2014), of 30,446 citizens of countries whose dominant religion is Christianity. According to the concern felt regarding the possibility of becoming the target of a terrorist attack, a religious profile was identified. Most of the sample reported high levels of worry about terrorist attacks. The most religious respondents, more faithful and more devoted to religious practices, are more worried about the occurrence of terrorist attacks. Opposite to what is mostly found in the literature, religion does not act as a protective barrier to the primary objective of terrorism, which consists in the use of violence to create fear. People worried about the probability of becoming a target in terrorist attacks are also victims of terrorism.


Introduction
Terrorism consists in the use of violence to create fear (i.e., terror, psychic fear) for political, religious, or ideological reasons. Terror is intentionally aimed at non-combatant targets, i.e., civilians or iconic symbols, and its objective is to achieve the greatest attainable publicity for a group, cause, or individual (Matusitz 2013). According to the Global Terrorism Index, from the Institute of Economics and Peace (IEP 2016), based on the Global Terrorism Database (START 2016), terrorism is a form of violence that mostly targets a small number of countries and is perpetrated by a small number of groups. The overall figures of terrorism are quite significant, with over 150,000 terrorist attacks occurring between 1970 and 2015 around the world, including 75,000 bombings, 17,000 assassinations, and 9000 kidnappings (START 2016). The IEP (2016) report notes that high levels of terrorist activity are related to high levels of political terror and political instability and low respect of human rights and religious freedoms for the United Nations (UN) or the European Union (EU). Several studies explaining the reasons for these attacks have been carried out, namely, political (McCauley and Moskalenko 2008;Ömer 2009;Midlarsky 2011), social (King and Taylor 2011;Mink 2015), economic (De Mesquita 2008;Caruso and Locatelli 2014), cultural (Pisoiu 2014;Shaffer 2015; study aimed to understand if religious identity protects individuals from feeling concerned about the possibility of terrorist attacks. A religious profile of these citizens, regarding the concern felt with the possibility of becoming the target of a terrorist attack, was identified. The authors suppose that most people worry about being targeted by a terrorist attack (1); and more religious people feel less worry than less religious or non-religious (2), possibly due to the protective nature of religion.

Sample Characterization
Women were slightly more numerous than men (56% against 44%). Respondents were aged from 16 to 97 years (M = 44.30,SD = 17.68). Most of the sample have a medium education (56%), are married or live together (61%), and are professionally active (51%) (see Table 1). The sociodemographic and religious profile reveals statistically significant differences in the distribution of the sample by the different levels of concern about the possibility of being the target of a terrorist attack (see Table 2). Table 2. Sample distribution relating the concern about the terrorist attacks.  Note: numbering in countries data refers to the importance of each country; bold values are the highest observed in the answer modality.

Religiosity-Religious Identity
In Table 4, the distribution of the items that define religious identity is presented and it can be found that the clear majority of the sample consider themselves religious (81%), considering religion very important (49%). Also, the overwhelming majority believe in God (94%), assuming that God is very important in their lives (60%).
Regarding worries about a terrorist attack, there are significant differences in religious identity. In what concerns the first item religious person (F(1) = 381.978; p < 0.001; η 2 = 0.012), subjects who do not consider themselves religious or who consider themselves atheists have lower values of concern (2.65) than those who consider themselves religious (2.96). The same happens when analyzing the item about the importance of religion (F(3) = 406.842; p < 0.001; η 2 = 0.039): Participants for whom religion is not definitely important present the lowest values of concern for terrorist attacks (2.44), followed by subjects for whom religion is not important (2.64), and then followed by subjects for whom religion is important (2.81) and very important (3.11). Subjects who believe in God have higher values of concern (2.89) than those who do not believe in God (2.39; F(1) = 442.451; p < 0.001; η 2 = 0.014). Participants who consider that God is very important (3.09) have significantly higher values than those who consider that God is not important (2.44), little important (2.58), or just important (2.69; F(3) = 466.066; p < 0.001; η 2 = 0.044).

Religiosity-Religious Practice
Concerning the religious practice, only 30% of the sample are an active member of the church and 40% attend religious services at least once a month; most of the sample (68%) prays at least once a day (see Table 5). Regarding worries about a terrorist attack, significant differences in religious practice were found. Concerning the first item active/inactive membership of a church or religious organization, the participants who are active members (2.97) have higher values than those who are not members (2.89) or those who are inactive members (2.84; F(2) = 31.402; p < 0.001; η 2 = 0.002). Regarding the question How often do you attend religious services, subjects who attend very little or never attend religious services present less concern about the terrorist attacks (2.72) than the participants who go once a year (2.76), on special days (2.81), once a month (2.94), and at least once a week (3.05; F(4) = 146.351; p < 0.001; η 2 = 0.016). With regard to the item How often to you pray, subjects who never pray have lower values (2.51) than those who pray once a year (2.78), on special days (2.74), several days a week (2.95), and every day (3.09; F(4) = 197.584; p < 0.001; η 2 = 0.021).

Discussion
This study aimed to understand if religious identity protects individuals from feeling concerned about the possibility of terrorist attacks. The study is justified by (1)  Most respondents reported high levels of worry about terrorist attacks, confirming the first hypothesis, which predicted that most people worry about being targeted by a terrorist attack. This can be explained by the overall numbers of terrorism, mentioned above, that are quite impressive (over 150,000 terrorist attacks between 1970 and 2015 around the world) and whose mass circulation is widespread through the media. Although there are few opportunities for the individual to become a victim of terrorism, most people feel threatened when being reminded about it through newspaper articles, pictures of attacks, etc. (Fischer et al. 2010). As internet-based contact is not regulated by the same standards and ethics as those of traditional print and television journalism, it may be possible to be inferred that this kind of information is mostly misinformation (Comer 2019), which may contribute to the perception of threat. Disruptive events, such as terrorist attacks, have many impacts both at the psychosocial and societal level; post-traumatic stress disorder, depression, fear, and anxiety are frequently associated with terrorist attacks (Oksanen et al. 2018). Also, danger perception of terror attacks was reported as being associated with both psychological distress and with severe stress symptoms (Mahat-Shamir et al. 2018).
Based on the data in this study, and regarding sociodemographic characteristics, countries where terrorist attacks seem to be most feared are unstable countries from a political point of view and with a history of social violence. This is in accordance with Shechory-Bitton and Cohen-Louck (2018), who reported that this fear of terrorism is perceived differently than other types of fear, namely, because of the geographic location. Participants who are male, older, not married/not living together, with higher education, and professionally non-active revealed less worries than participants who are women, younger, married/living together, less educated, and professionally active. Lerner et al. (2003) stated that females experienced more pessimistic risk than males, which may explain the results in relation to gender. Also, Cohen-Louck and Levy (2018) found that men reported higher perceptions than women regarding risk perception of terrorism. Contrary to the results in this study, Williamson et al. (2019) found a positive association between fear of terrorist attacks and age and a negative one with married participants (less fear). However, the same authors reported that as educational achievement increases, the fear of terrorism decreases, in agreement with the results in this study. Subjects who are married/living together present higher values than those who are not married/not living together, possibly due to the fact that having a family makes them feel more concerned about their future in general and with terrorist threats, in particular. Respondents with no formal education report significantly higher levels of concern than those with more education, relating to the fact that less educated people may be less aware of the significance of the probability of a terrorist attack. Professionally active respondents tend to report significantly more worry than non-active ones, possibly because non-active people have less to lose than active individuals.
Most of the sample considers itself religious, with being religion very important. Also, this majority believe in God and consider that God is very important in their lives. These results seem to contradict those of other authors (Voas 2008;Voas and Chaves 2018), who state that there is a religious decline in all countries, mainly in religious practice (Molteni and Biolcati 2018), supporting the 'believing without belonging' theory (Davie 1990;Flanagan and Davie 1995), which suggests a widening gap between stable or increasing private and intimate religiosity and decreasing public religiosity.
Subjects who do not consider themselves religious or who consider themselves atheists have less concern with terrorist attacks than those who consider themselves religious. Also, participants for whom religion is not important present the lowest values of concern for terrorist attacks than subjects for whom religion is important. These results do not confirm the supposed second hypothesis, which predicts that more religious people feel less worry about terrorist attacks. These results also seem to contradict those found by Fischer et al. (2006), who claim that intrinsic religiosity helps to cope with the increased salience of terrorism; and even that high probability of terrorism only negatively affects the mood of non-religious participants but not of intrinsically religious persons. However, Adamczyka and LaFree (2015) showed that more religious respondents are more likely to express concerns about terrorism, although this relationship were mediated by their level of conservatism. Also, Haner et al. (2019) found that gender, religiosity, and psychological distress were most consistently associated with fear of terrorism and worry about being a victim of a terrorist attack.
Subjects who believe in God and consider that God is important present more concern than those who do not believe in God and consider that God is not important. These results may be in accordance with Jong and Halberstadt (2016), who found that the narrative reporting that non-religious people fear death more than religious people is not necessarily true. According to Pennycook et al. (2016), there is evidence that atheists and agnostics are more reflective than religious believers, which may explain the results obtained in this study.
Almost a third of the sample are an active member of the church, about two in five attend religious services at least once a month, and the majority prays at least once a day. Active members present higher values of worry than those who are not members or inactive members. Subjects who attend very little or never attend religious services and that never pray present less concern about the terrorist attacks than the participants that do it. These results clearly contradict those of Kupor et al. (2015), as well as those of Jonas and Fischer (2006), who consider that religion has a protective effect on people's lives. Although attending religious services has been shown to be one of the strongest religious predictors of well-being (George et al. 2002), it only happens with those who internalize their religious beliefs and try to live out their religion on a daily basis; otherwise, anxiety may emerge (Steffen et al. 2017), contributing to the fear of terrorism.

Conclusion
The results of this study seem to question the idea that religion has a protective effect on people's lives, in the sense that the more religious believers and assiduous persons in religious practices are not protected from feeling more concern about terrorist attacks. It was found that religiosity is not protective of concern created by violence, which is the primary target of terrorism (Kupor et al. 2015). This study suggests, in accordance with the one from Brañas- Garza et al. (2013), that religiosity impacts on individual decisions and behavior, and its intensity affects social interactions and attitudes. Thus, it is possible to state that the more religious participants are other victims of terrorist attacks, although not the immediate ones (Matusitz 2013). This study contributes to the body of knowledge about religion and terrorism as it sheds light on the relationship between religion and fear, showing that religion does not protect religious people from feeling fear and or concern about terrorism, as is mostly claimed in the literature.

Limitations
The main limitation of this study concerns the fact that all the sample belong to a single religion, Christianity. However, it was a methodological option to avoid that the type of religion influences the variables studied, namely, religious practice; this would certainly vary according to the type of religion and not because of the concern about terrorist attacks. Another limitation concerns the selection of countries being mostly Christian: Countries with a Christian majority were chosen; however, this majority varies greatly from countries, with a percentage of 51% of Christians to countries with 90% of Christians. Finally, this study was a cross-sectional one, preventing the establishment of causal relationships between the variables under study, and thus allowing only the establishment of differences and associations.

Research Implications
Future research should include all religions, especially the most significant ones. In addition, countries whose percentage of religious majority is equivalent should be sought. At last, longitudinal research designs can best fit the subject under analysis.

Participants and Procedure
Data used in the study were collected from wave 6 of the WVS (2016). These data were obtained between 2010 and 2014 and include basic information on demographics, value and attitude measures, and organizations' affiliation of more than 85,000 respondents in 57 countries. WVS has a wellstructured sample of respondents and is not limited to special groups of individuals. It contains a question regarding the concern about terrorist attacks and a set of questions, not only about religious affiliation, but also on the intensity of individual beliefs.
The participating countries in this study were all countries included in WVS wave 6 whose dominant religion is Christianity (Pew Research Center's Forum on Religion & Public Life 2012): Argentina, Australia, Armenia, Brazil, Belarus, Chile, Colombia, Cyprus, Ecuador, Estonia, Georgia, Germany, Ghana, Mexico, Netherlands, New Zealand, Peru, Philippines, Poland, Romania, Russia, Rwanda, Slovenia, South Africa, Zimbabwe, Spain, Sweden, Trinidad and Tobago, Ukraine, United States, and Uruguay. Of these countries, only subjects who considered themselves Christian were selected, comprising a total of 31,749 citizens. This selection sought to avoid differences between religions (identities and practices) determining the level of concern about terrorist attacks. Since the statistical analysis was limited to observations without outliers and without missing data, the final sample resulted in 30,446 respondents. The socio-demographic variables used were gender, age (in years and in categories), marital status (whose response modalities were converted into only two: Not married/not living together and married/living together), highest education level (these response modalities were converted into four: No formal education, low education, medium education and high education), and employment status (whose response modalities were converted into two: Inactive and active).

Worry of Terrorist Attacks
The question about the worry of terrorist attacks used is part of a list of situations. It was measured with one 4-point item and participants were asked "To what degree are you worried about the following situations?" (1 very much-4 not at all). The situation was "a terrorist attack". Responses were recorded, with higher scores indicating higher levels of worries.

Religiosity
To identify the religious profile, a set of religious indicators was used. Response scales were adjusted so that the higher values corresponded to higher religiosity levels. Religiosity was indicated by two topics: Religious identity and religious practice.
1. Religious identity. Religious identity was measured with the following four items: (a) "Independently of whether you attend religious services or not, would you say you are" (1 a religious person, 2 not a religious person, 3 an atheist). This item has been recommended as the best single measure of personal piety (Stark 2002). We combined the second and third response ratings as the non-religious group (0), which was compared to the first group, the religious one (1). (b) "How important is religion in your life?" (1 not at all important, 2 not very important, 3 rather important, 4 very important). (c) "Do you believe in God" (1 yes, 2 no). This item has been commonly used to assess religious beliefs (Norenzayan and Hansen 2006). We recoded the responses as 1 (yes) and 0 (no). (d) "How important is God in your life?". Participants answered this question on a 10-point scale (1 not at all important-10 very important). The responses were recoded as 1 not at all important to 4 very important. 2. Religious practice. The measure of religious practice contained three items: (a) "Apart from weddings and funerals, about how often do you attend religious services these days?" Participants answered this question on a 7-point scale (1 more than once a week-7 never, practically never). (b) "Apart from weddings and funerals, about how often do you pray?" Participants responded to this question on an 8-point scale (1 several times a day-8 never, practically never). The 2 items have shown good validity to estimate religious actions (Sethi and Seligman 1993). The responses were recoded to the 2 items in a 5-point scale (1 never-5 often), with higher scores indicating more religious practice. (c) "Active/Inactive membership: Church or religious organization" (1 not a member, 2 inactive member and 3 active member).

Data Analyses
A univariate analysis was applied to characterize the sample. Chi-square analyses were used to examine changes in the demographic characteristics and the religious profile in relation to the concern felt about terrorist attacks. To identify the profile of groups based on the worry of terrorist attacks, contingency tables and one-way ANOVA tests were computed. Significance was set at p ≤ 0.05. Data analysis was conducted using SPSS software, Version 25.
Funding: This research received no external funding.

Conflicts of Interest:
The authors declare no conflict of interest.