Psychological Study of Perceived Religious Discrimination and Its Consequences for a Muslim Population

This study focuses on the consequences of perceived religious discrimination on a Muslim population. Specifically, the purpose of this research was to assess the impact of perceived religious discrimination in the news media on a number of individual variables (self-esteem and perceived stress). It was also aimed at studying identification as a coping mechanism. A total of 88 Muslim participants took part in this study. The mediation analyses carried out confirm that group identification plays a mediating role in the indirect relationship between perceived religious discrimination in the news media and perceived stress. This discussion highlights the benefits of social support against the deleterious effects of discrimination.


Introduction
Most classical studies on intergroup bias aimed at assessing and understanding the attitudes and behaviors of the members of privileged groups toward discriminated groups.The question of the perception of discrimination by the targets is nonetheless fundamental.Indeed, a large body of research has shown that being the target of discrimination can cause psychological and physical suffering (Bourguignon et al. 2012;Clark et al. 1999;Every and Perry 2014;Lagacé and Tougas 2006;Major and O'Brien 2005;Noh and Kaspar 2003).These negative effects concern all perceived discrimination, be they ethnic (Krieger and Sidney 1996), related to disability (Snyder et al. 2010), gender (Schmitt et al. 2003), age (Redman and Snape 2006) or religious orientation (Kunst et al. 2016).However, while perceived discrimination negatively affects the physical and psychological well-being of stigmatized individuals, a review of the literature reveals that these individuals use adjustment strategies to counteract these deleterious effects and enable them to adapt socially (Bourguignon and Herman 2007;Branscombe et al. 1999;Crocker and Major 1989).
Thus, the study presented below is part of this perspective and focuses on the evaluation of perceived religious discrimination of Muslims in the news media.If we consider the current typology in this field, our study can be viewed as exploring group or systemic discrimination (in the terminology of Every and Perry 2014) rather than interpersonal discrimination.Indeed, group discrimination stems from the way in which society and its institutions function.More precisely, it is this functioning that creates, maintains and perpetuates intergroup inequalities.In addition, group discrimination is evidenced by the repetition of acts and unfavorable decisions toward a target group.By contrast, interpersonal discrimination takes place when a person treats another person unfairly due to the social category to which he or she belongs.This type of unequal treatment is the consequence of either stereotypical beliefs about the target person or negative evaluations of the target group (Schmidt 2010;Whitley and Kite 2013).The distinction between these two forms of discrimination is fundamental to the extent that they do not produce the same effects.For example, Bourguignon et al. (2006Bourguignon et al. ( , 2012) ) have distinguished between personal discrimination and group discrimination and they have shown that the effects of these discriminations on self-esteem are different.In other words, they have shown that the perception of personal discrimination was negatively related to individual self-esteem, while the perception of group or systemic discrimination was positively associated (although weakly) with individual self-esteem.Furthermore, a body of research that focuses on Muslim people has highlighted the negative effects of perceived discrimination on a number of psychological variables.Thus, in a study on Muslim participants, Every and Perry (2014) found a negative link between interpersonal discrimination and self-esteem.Symmetrically, they identified a positive (although weak) link between systemic discrimination (e.g., in the news media) and self-esteem.In addition, Jasperse et al. (2012), in New Zealand, highlighted that the more Muslim women perceive personal discrimination against their religion, the lower their life satisfaction will be.In the same vein, Ysseldyk et al. (2010) proposed that when discrimination targets their religious identity, Muslim people tend to perceive more religious discrimination and, correlatively, more psychological distress.Finally, in their research on Turkish Muslim minorities in The Netherlands, Verkuyten and his collaborators (Verkuyten 2007;Verkuyten and Yildiz 2007) showed that these people do not remain passive when perceiving intergroup hostility.On the contrary, they develop coping strategies such as identification.This strategy helps to contain the deleterious effects of this perceived hostility.These different studies are, therefore, emblematic of a current of research that is aimed at studying the fate of Muslim people who perceive discrimination.However, this literature review led us to observe that: (1) Despite the fact that the attacks committed in France since 2015 have led to an upsurge in hostility toward people of Muslim faith and toward their religion (Collectif Contre l'Islamophobie en France 2016; Commission Nationale Consultative des Droits de l'Homme 2016; Ipsos 2015), there is, to our knowledge, no francophone equivalent of the studies mentioned.For example, the Collective Against Islamophobia in France (CCIF) recorded 905 attacks against Muslims or their religion during the year 2015, which represents an increase of 18.5% compared to 2014 (764).(2) The contribution of the news media in the development and trivialization of discrimination against Muslims is not sufficiently considered by social psychologists.Indeed, several reports and articles initiated by researchers from other disciplines have focused on the image of Islam and its practitioners in the news media (Commission Fédérale contre le Racisme 2006; European Monitoring Center on Racism and Xenophobia 2006;In France: Amiraux 2003;Asal 2014;Geisser 2012;Hajjat and Mohamm 2016;Liogier 2015).According to these works, news media discourse participates more or less consciously in the construction of an "imaginary Islam" (Deltombe 2005) and in the "demonization" of Muslim people (Parini et al. 2012;Yavari-D'Hellencourt 2000).This news media construction thus maintains the conflation between "Islam" and "Islamism" and allows the spread of stereotypical images or speeches (Mishra 2007;Parini et al. 2012).This presentation of Islam and its believers could explain the negative perception that the general public has of them.This recurrent news media stigmatization is not without effect on the everyday experience and feelings of Muslim people.It should thus be thoroughly studied.However, it has been little investigated by psychosocial research, especially in francophone countries.To our knowledge, only one Norwegian research attempted to assess the consequences of the perception of religious discrimination in the media.Indeed, Kunst et al. (2013) focused on examining the relationship between the perceptions of discrimination and perceived stress.They showed the existence of a significant link between the perception of discrimination in the media and an increase in perceived stress.The present study aims at highlighting this type of link and should thus be considered as a contribution to the field of perceived religious discrimination.
The aim of this research is to study the mediating role of group identification in the relationship between perceived religious discrimination in the news media and individual variables such as perceived stress and self-esteem.The results from the literature mentioned above lead us to consider the negative tone of the news media discourse concerning Muslims and its amplification (Poole 2011;The Runnymede Trust 1997;Saeed 2007).Indeed, news media discourse, more or less consciously, commonly associates Muslim people with references such as "terrorism" or "aggression".Moreover, this discourse tends to magnify the supposed incompatibility between Muslim culture and the dominant culture.The result is a negative representation of Muslim individuals, which leads to the legitimization of discrimination against Muslims.Finally, this study relies on the rejection-identification theory (Branscombe et al. 1999;Jetten et al. 1996Jetten et al. , 2001;;Schmitt and Branscombe 2002).This theory suggests that a high level of psychological identification with their group helps to mitigate the negative effects of perceived discrimination on the well-being and self-esteem of stigmatized group members.In fact, when individuals overinvest in their group, they tend to feel less isolated in the face of stigma (Bourguignon et al. 2006).

Participants
Eighty-eight Muslim participants with an average age of 31.51 years (SD = 9.53) agreed to participate in this research.There were 47 women (M = 30.15,SD = 10.38) and 41 men (M = 33.07,SD = 8.30).Fifty-two of these participants were born in France and 36 were born in other countries (Turkey, Morocco and Algeria).

Materials
Participants were invited to answer a self-report questionnaire composed of a number of tools:

•
A scale measuring the perception of religious discrimination in the news media.To our knowledge, there was no scale that specifically measured the perception of religious discrimination in images broadcasted by the news media.That is why we designed this tool by drawing on items developed by Kunst and his colleagues (Kunst et al. 2013;Kunst et al. 2012).Our scale consists of 6 items (e.g., "In the media, Islam is often presented as a threat to French culture").Three items were put in the negative form so as not to induce bias in participants' responses ("French media do not portray Muslims as dangerous people").Participants were asked to rate them on a seven-point Likert scale ranging from 1 (strongly disagree) to 7 (strongly agree).Thus, a high score on this scale indicates a strong perception of religious discrimination in the news media.In this study, Cronbach's alpha is 0.70.

•
An identification scale for the Muslim group consisting of 6 items from Verkuyten's research (2007).Item 2, "I strongly identify with Muslims", is an example of these items.This is a seven-point Likert scale ranging from 1 (strongly disagree) to 7 (strongly agree).In this study, Cronbach's alpha is 0.90.
A French version of this scale was validated by Vallières and Vallerand (1990).This tool consists of 10 items and participants were invited to answer on a Likert scale ranging from 1 (strongly disagree) to 4 (strongly agree).Item 2, "I think I have a number of qualities", is an example of these items.In this study, Cronbach's alpha for this study is 0.74.

•
A perceived stress scale (PSS 10) by Cohen et al. (1983).It was validated in French by Bellinghausen et al. (2009).This tool consists of 10 items including 4 negative items.The participants were asked to evaluate each proposal on its frequency of occurrence during the last month.Item 3, "Last month, how often did you feel nervous and stressed?" is an example of these items.Participants used a Likert scale ranging from 1 (never) to 5 (often).In this study, Cronbach's alpha is 0.89.

•
Sociodemographic variables: participants were asked to indicate their religion so that we could make sure that we would only integrate participants who self-identified as Muslims.Age, gender, and place of birth of participants were also taken.

Procedure
Some of the participants were contacted in the entourage of the experimenters and others were recruited by mail in an association and a mosque in the Pays de la Loire region.
To facilitate the completion of this survey, both paper and online versions were made available.Regardless of the version of the questionnaire chosen, anonymity and confidentiality were guaranteed to participants.

Operational Hypotheses
Hypothesis 1 (H1).The high score of perceived religious discrimination toward Muslims in the news media predicts a high level of stress (Kunst et al. 2013).
Hypothesis 2 (H2).The high score of perceived religious discrimination toward Muslims in the news media predicts a positive self-esteem score (Bourguignon et al. 2012;Every and Perry 2014).

Hypothesis 3 (H3).
Identification with the Muslim group plays a mediating role in the relationship between perceived religious discrimination and self-esteem (Branscombe et al. 1999;Jetten et al. 1996Jetten et al. , 2001;;Schmitt and Branscombe 2002).
Hypothesis 4 (H4).Identification with the Muslim group plays a mediating role in the relationship between perceived religious discrimination and perceived stress (Branscombe et al. 1999;Jetten et al. 1996Jetten et al. , 2001;;Schmitt and Branscombe 2002).

Descriptive Analysis
The 88 participants obtained an average of 5.92/7 (SD = 0.94) on the scale of religious discrimination perceived in the news media.These results reveal that they strongly perceive discrimination against their group and their religion, respectively.Indeed, more than half of the participants agreed (30%), or strongly agreed (21%) that the news media portrays Muslims or their religion as a threat or a danger to French culture.In addition, participants scored an average of 6.07/7 (SD = 1.11) in the identification scale and 73% of them agreed (49%) or strongly agreed (24%) with the statement "I strongly identify with Muslims".These scores indicate that Islam plays an important role in the identity of Muslim people.Moreover, participants had a high self-esteem, with an average score of 3.49/4 (SD = 0.37).Finally, the results show that participants perceive moderately high stress (M = 2.56/5, SD = 0.70).

Preliminary Analyses
The correlational analyses are summarized in Table 1.They reveal that perceived religious discrimination toward Muslims in the news media is positively related to group identification (r = 0.34; p < 0.001) and that group identification is negatively related to perceived stress (r = −0.46;p < 0.001).On the other hand, discrimination is not associated with self-esteem or perceived stress.This also applies to identification that is not associated with self-esteem.

The Mediating Role of Identification with the Muslim Group in the Relationship between Perceived Religious Discrimination and Individual Variables
In order to examine the impact of perceived religious discrimination toward Muslims in the news media on self-esteem and perceived stress, and test the mediating role of group identification in the link between perceived religious discrimination and individual variables (self-esteem and perceived stress), we used model 4 of the PROCESS procedure for SPSS (Hayes 2013).This procedure allows us to directly obtain the direct effect and the indirect effect of the independent variable on the dependent variable.
In the model, self-esteem and perceived stress are the dependent variables, while perceived religious discrimination toward Muslims in the news media is the independent variable.To make this link evident, group identification corresponds to the mediator variable.
Perceived religious discrimination toward Muslims in the news media has no direct effect on perceived stress (β = −0.10;p = 0.204) or on self-esteem (β = 0.07; p = 0.121).Hypotheses H1 and H2 are not validated.Furthermore, group identification has no direct effect on self-esteem (β = 0.02; p = 0.522).As a result, the mediating effect could not be tested and Hypothesis 3 cannot be validated.
Nevertheless, the mediating role of group identification in the relationship between perceived religious discrimination toward Muslims in the news media and perceived stress could be tested.The regression analysis shows a direct effect of perceived discrimination in the news media on identification with the Muslim group (β = 0.40, F(1, 86) = 11.38,p < 0.001).Thus, perceived discrimination explains 12% of identification with the Muslim group.To the extent that R = 0.34, the model shows a moderate fit with the data (Table 2).Regression analysis indicates a direct effect of group identification on perceived stress (β = −0.29,F(2, 85) = 11.39,p < 0.001).Thus, identification with the Muslim group explains 21% of perceived stress.Finally, to the extent that R = 0.46, we can say that the model fits well with the data (Table 3).The direct effect of perceived religious discrimination in the news media on perceived stress is not significant (Table 4), which is confirmed by the p value of 0.82 and by the limits of the confidence interval, which allow this effect to have a value of zero (Preacher and Hayes 2008).In contrast, the indirect effect of perceived religious discrimination in the news media on perceived stress through identification with the Muslim group is significant (indirect effect, IE = −0.12,SE = 0.05, 95 %, [CI] = [−0.24,−0.04]) (Table 4).These results validate the existence of a total mediation of identification in the indirect relationship between perceived religious discrimination toward Muslims in the news media and perceived stress.The mediation model is summarized in Figure 1.Hypothesis H4 is validated.

Discussion
The aim of this research was to study the mediating role of group identification in the relationship between perceived religious discrimination toward Muslims in the news media and individual variables such as perceived stress and self-esteem.This aim is only partially achieved insofar as the mediation hypothesis between discrimination and perceived stress has been validated.
The lack of a significant relationship between self-esteem and the theoretical constructs classically associated with it is surprising.The meta-analysis by Schmitt et al. (2014) helps us to explain and understand this lack of connection.In fact, this meta-analysis shows that (1) the effects of perceived discrimination are greater when the latter concerns interpersonal discrimination than when group discrimination is at stake, and (2) that the link between perceived discrimination and the positive variables (particularly self-esteem) is generally weaker than with the negative variables (e.g., depression).The lack of effect on self-esteem may also be due to the fact that religiosity moderates the relationship between discrimination and self-esteem (Ghaffari and Ç iftçi 2010).Finally, the work of Bourguignon and his colleagues showed that the perception of personal discrimination is negatively linked to individual self-esteem.Conversely, the perception of group discrimination is weakly but positively associated with individual self-esteem (Bourguignon et al. 2006(Bourguignon et al. , 2012)).In our study, the perception of group religious discrimination was assessed.Thus, it is likely that Muslim people perceived discrimination against their ingroup while considering that their personal fate was better than the fates of other members of their group.Therefore, their self-esteem would be preserved.
Despite this, the mediation analysis validates Hypothesis 4 and confirms the existence of a total mediation of identification in the indirect relationship between religious discrimination perceived in the news media and perceived stress.Specifically, the more Muslim people perceive this

Discussion
The aim of this research was to study the mediating role of group identification in the relationship between perceived religious discrimination toward Muslims in the news media and individual variables such as perceived stress and self-esteem.This aim is only partially achieved insofar as the mediation hypothesis between discrimination and perceived stress has been validated.
The lack of a significant relationship between self-esteem and the theoretical constructs classically associated with it is surprising.The meta-analysis by Schmitt et al. (2014) helps us to explain and understand this lack of connection.In fact, this meta-analysis shows that (1) the effects of perceived discrimination are greater when the latter concerns interpersonal discrimination than when group discrimination is at stake, and (2) that the link between perceived discrimination and the positive variables (particularly self-esteem) is generally weaker than with the negative variables (e.g., depression).The lack of effect on self-esteem may also be due to the fact that religiosity moderates the relationship between discrimination and self-esteem (Ghaffari and Çiftçi 2010).Finally, the work of Bourguignon and his colleagues showed that the perception of personal discrimination is negatively linked to individual self-esteem.Conversely, the perception of group discrimination is weakly but positively associated with individual self-esteem (Bourguignon et al. 2006(Bourguignon et al. , 2012)).In our study, the perception of group religious discrimination was assessed.Thus, it is likely that Muslim people perceived discrimination against their ingroup while considering that their personal fate was better than the fates of other members of their group.Therefore, their self-esteem would be preserved.
Despite this, the mediation analysis validates Hypothesis 4 and confirms the existence of a total mediation of identification in the indirect relationship between religious discrimination perceived in the news media and perceived stress.Specifically, the more Muslim people perceive this discrimination, the more they will identify with their ingroup and the less they perceive stress.Thus, this identification with the ingroup has beneficial effects on psychological well-being in the sense that the group offers social support and makes it possible not to feel alone in the face of this feeling of discrimination (Bourguignon et al. 2006;Cameron and Lalonde 2001).
These results confirm previous research that addressed the issue of group identification in the context of perceived discrimination (Branscombe et al. 1999;Jetten et al. 2001;Schmitt and Branscombe 2002).They also call for further research on the social fact of Islamophobia in France.
To conclude, this study is important in the context that much of the research into news media representations of Muslims has focused on analyzing content, rather than responses to that content by Muslims.That makes this a useful study with potential policy and practice implications.The study by Kunst et al. (2016) demonstrates that the greater the fear of Islam, the greater the demand that minorities assimilate-the greater the perceived Islamophobia, the greater the separation of minorities and the greater the fear of them.It is the vicious circle of Islamophobia.This points to the responsibility of media and policy discourse to integrate minority members, and the need to decrease religious prejudice.

Table 4 .Figure 1 .
Figure 1.Negative indirect effect of perceived religious discrimination in the news media on perceived stress through Muslims' identification with their group.

Figure 1 .
Figure 1.Negative indirect effect of perceived religious discrimination in the news media on perceived stress through Muslims' identification with their group (Note *p < 0.001; IE = indirect effect; CI = Confidence Intervals).

Table 1 .
Correlations between discrimination, identification, self-esteem and perceived stress.

Table 2 .
Direct effect of discrimination on identification.

Table 3 .
Direct effect of identification on perceived stress.