Religiosity and the Perception of Interreligious Threats: The Suppressing Effect of Negative Emotions towards God

Abstract

Religiosity has been studied for its impact on other sociological and psychological aspects of society, particularly personal wellbeing and interpersonal relationships. However, it has yet to be studied for its impact on interreligious prejudice as measured by perceptions of interreligious threats. The present study investigates how religiosity (both positive and negative measures) affects perception of threats from other religious groups within the Malaysian context by using the Centrality of Religiosity Scale and the Inventory of Emotions towards God as measures of religiosity. Data collected through questionnaires administered to university students and recent graduates (N = 260) in Malaysia were subjected to bivariate correlation analysis, multiple regression analysis and mediation analysis. Our findings show that the positive and negative measures yielded different effects on the perception of interreligious threats. While the centrality of religiosity and positive emotions towards God have statistically significant negative correlation with perception of interreligious threats, we show that negative emotions towards God suppresses the effect of the positive measures of religion on the dependent variable. The paper discusses the implications of these results within the socio-political context of Malaysia, in which ethnic identity and religious affiliation are closely intertwined.

1. Introduction

Malaysia often prides itself on the cultural and religious diversity that makes up the composition of the nation, but if this diversity is to be celebrated at all, it needs to be more than just a tourism tag line. A predominantly Muslim nation (63.5%), Malaysia also has a significant population of Buddhists (18.7%), Christians (9.1%), Hindus (6.1%), and adherents of other religions (0.9%). Another 1.8% claim to have no affiliation with any religious groups (Department of Statistics Malaysia 2022). A study by the Pew Research Center (2018) indicated that 77% of the population reported religion to be “very important” to them, and, unlike the global trend, there is also no statistically significant difference in religiosity between the old (above 40 years old) and the young (below 40 years old).

While freedom of religion is promised in the constitution, Islam is recognised as the official state religion and is stated in the constitution as a defining characteristic of the Malay ethnicity as well, contributing to the intertwined conceptualisation of ethnicity and religion in the nation. Along with the parallel existence of Civil Law and Syari’ah Law in Malaysia, these have contributed to complicated interreligious relations in the social and political realms. Muzaffar (1996) traced the complex interreligious / interracial relations all the way back to the founding of the nation as an Islamic-Malay polity, at once highlighting the closely intertwined nature of ethnicity and religion in Malaysia. More recently, Tan (2022) pointed to religion as a form of identity politics in the racialised context of Malaysia.

Religion continues to take centre stage in the social and political life of the nation. Indeed, it is the most significant factor that shaped the unexpected outcome of the recently-concluded 15th General Election that saw the conservative Islamic party being the single party that won the highest number of seats in parliament while its political coalition partner played up interreligious sentiments during the election campaigning period (Ng 2022). The fact that this outcome came about after the recent passing of the bill that lowered the voting age to 18 (and thus, for the first time, allowed those aged between 18 to 20 to vote) seems to indicate that religion may serve as a greater factor of consideration for the younger generation. Against this socio-political backdrop, the present study explores the relationship between religiosity, emotions towards God, and perceptions of interreligious threats in Malaysia, and discusses the implications of our findings with regards to recent experiences of interreligious relations in the nation.

1.1. Religiosity as Multidimensional

Religiosity has been studied across various academic disciplines, with many attempts to define the concept and predict its impact on other sociological and psychological aspects of the human experience. While the propositions and findings vary, scholars have generally converged on the multidimensionality of religiosity. Glock and Stark’s (1965) influential work proposed five dimensions of religiosity: experiential, ritualistic, ideological, intellectual, and consequential, which they subsequently revised in 1968 with the removal of the consequential dimension and the splitting of the ritualistic dimension into private practice and public practice (Stark and Glock 1968). Though Glock and Stark’s work was mainly undertaken within the context of North American Christianity, many studies elsewhere involving religiosity adopted a similarly multidimensional view of religiosity. One such is Saroglou (2011), who presented four dimensions that somewhat correspond to Glock and Stark’s proposal: believing, bonding, behaving, and belonging.

While the relationship between religiosity and wellbeing has long been established (Lassi and Mugnaini 2015; Koenig 2012), more recent studies that employ religiosity as a predictor variable tend towards supporting this multidimensional conceptualisation of religiosity. Studying the psychological effects of religion, Baetz et al. (2004) found that frequent attendance at religious gatherings lowered the risk of depression, but that placing high importance on faith and spiritual values contributed to higher levels of depressive symptoms instead. Edara et al. (2021), on the other hand, found that a belief in the existence of the divine leads to increased resilience in individuals. Examining the relationship between religiosity and life satisfaction, Bergan and McConatha (2001) looked specifically at religious affiliation and private religious devotion, and found the former to have a stronger relationship with life satisfaction than does the latter.

1.2. Religiosity and Emotions towards God

As religion often yields psychological impact on individuals, Huber and Richard (2010) argued for the necessity to study emotions associated with religion, which led to their development of the inventory of Emotions towards God (EtG), a short instrument that measures both positive and negative psychological valences associated with religion. In validating the EtG, Huber and Richard (2010) found that the presence of both positive and negative emotions correlated with the experiential, ritualistic (private), and ideological dimensions of religiosity, but not with the purely intellectual dimension, thus highlighting that the presence of these emotions is associated with the experiential construct of religion. Echoing Huber and Richard’s assertion, Ferran (2019) also posited that religious emotions make up a part of the experiential dimension of religion.

Studies have shown a complex relationship between religiosity and the presence of these emotions, and how they affect individual wellbeing and interpersonal relationships. Martínez de Pisón (2022) found that religion both contributes to excessive feelings of guilt and shame while simultaneously also enabling healing from these same negative emotions. While some had found that those with higher levels of religiosity tend to display a greater willingness to forgive others (Edwards et al. 2002; Rye et al. 2001), Huber et al. (2011) pinpointed on the experience of forgiveness by God as facilitating a greater willingness to be forgiving of others, but this relationship is moderated by one’s centrality of religiosity. Though studies have generally found a positive correlation between religiosity and satisfaction with life, Aditya et al. (2020) discovered, in their sample, that the relationship is moderated by feelings of anger towards God.

1.3. Perceptions of Interreligious Threats

The Intergroup Threat Theory (ITT) proposed by Stephan et al. (2009) suggests that humans are “tribal” by nature in that we rally together in groups that share similar elements for identity formation. These groups tend to anticipate threats from other outgroups. The theory is particularly concerned with perceived threats—regardless of whether they are real or not—because even perceived threats can lead to real consequences. The authors proposed two types of perceived threats: realistic threats (threats to physical, socio-political welfare) and symbolic threats (threats to identity and values). Building on the foundation of this theory, Makashvili et al. (2018) found that perceived threats contribute to an increase in prejudice towards outgroups and religiosity significantly moderated that relationship.

It is, thus, worth investigating perception of interreligious threats as an indicator of interreligious prejudice in Malaysia. However, the two types of threats as proposed by Stephan et al. (2009) may not be as clearly distinctive in the context of Malaysia. Due to the intertwined nature of ethnicity and religion, and the affirmative action policy that favours the Muslim majority of the country, the perceived threats from other religious groups may be both realistic and symbolic in nature.

1.4. Research Objectives and Hypotheses

While much of the existing literature has explored the impact of religiosity on individual wellbeing, not much has been explored in terms of perception of interreligious threats. Therefore, the present study seeks to bridge this gap by examining how various measures of religiosity affect the perception of threats between religious groups in Malaysia. The measures of religiosity employed in this study include the centrality of religiosity, and positive and negative emotions towards God.

As the literature has shown that positive aspects of religiosity facilitate a greater satisfaction with life and generally better interpersonal relationships, we formulate the following hypotheses:

Hypothesis 1 (H1).

Centrality of religiosity correlates negatively with perceptions of threat though the level of correlation may vary for the different dimensions of religiosity.

Hypothesis 2 (H2).

Positive emotions towards God correlates negatively with perceptions of threat while negative emotions towards God correlates positively with perceptions of threat.

Since the measures of religiosity used in this study include both positive and negative psychological valences, we hypothesized a complex relationship between the predictor variables (measures of religiosity) and the outcome variable (perception of interreligious threats). Huber and Richard (2010) found that the presence of both positive and negative emotions towards God increase in frequency along with an increase in the centrality of religiosity. Since some literature has shown moderation (Huber et al. 2011; Aditya et al. 2020) and mediation effects (Szcześniak et al. 2019) in similar studies, initial results from the testing of H1 and H2 led us to formulate H3 as follows:

Hypothesis 3 (H3).

Negative emotions towards God has a suppressing effect (also known as negative mediation effect) on the relationship between religiosity and perception of interreligious threats.

2. Research Methods and Instruments

2.1. Ethics and Consent

The study and research design obtained ethical approval from the University of Nottingham Malaysia Faculty of Arts and Social Sciences Research Ethics Committee. Informed consent was also obtained from the participants before the questionnaire was administered.

2.2. Research Instruments

The research design employed in this study is a descriptive design, whereby the relevant variables are described, highlighting their inter-relationships through hypotheses. A quantitative research methodology was used, with data collected from a questionnaire administered via Qualtrics. While the questionnaire was part of a wider study, data from four sections of the questionnaire were analysed and reported in this study: (1) demographic information, (2) Centrality of Religiosity Scale (CRS), (3) the inventory of Emotions towards God (EtG), and (4) perception of threats from other religious groups (Threat).

The Centrality of Religiosity Scale (CRS) was developed by Huber and Huber (2012). It has been translated into more than 20 languages and administered in more than 25 countries. The instrument was designed with the specific aim of measuring “the centrality, importance or salience of religious meanings in personality” (Huber and Huber 2012, p. 711), and the research team theorised that it is this perception of the importance of religion that may play a role in affecting one’s perception of interreligious threats. The scale was also built upon the five-dimensional view of religiosity as proposed by Stark and Glock (1968), and this element of multidimensionality provides richer analysis for the present study. Considering the multi-religious context of Malaysia, we administered the CRSi-20 that has five additional items catering to the differences in religious constructs and beliefs across different religions, particularly for Islam and Buddhism that have a strong presence in Malaysia. Each of the five dimensions—ideology (IDL), intellectual (ITL), public practice (PUB), private practice (PRI), and experience (EXP)—was measured with three items in the CRS, making a total of fifteen valid responses for each respondent. Acknowledging the sensitivity of religious terminologies, the research team consulted with the developer of the instrument to translate the CRSi-20 into Malay, Chinese, and Tamil for the Malaysian context so that participants had the option of answering this part of the survey in the language they worship in. To ensure clarity and accuracy of the translations, we worked with native speaker translators for all three translations and engaged different native speakers of a different religion to back-translate the instrument into English. Nevertheless, all the valid responses were completed in English, which is the language used in the rest of the questionnaire. The instrument was administered in Likert-scale format, and a final CRS score was calculated by averaging out the respondents’ scores for each item.

Developed by Huber and Richard (2010), the inventory of Emotions towards God (EtG) is a 16-item inventory with nine emotions of positive psychological valence and seven emotions of negative psychological valence. All sixteen items are also mapped to three distinct theological issues: divine holiness, divine providence, and divine justice. Participants were asked to indicate on a five-point Likert scale the frequency with which they experience each emotion towards God, deities or any divine power, with 1 being never and 5 being very often. A higher score would then reflect a more frequent experience of a certain emotion. For the present study, separate average scores were tabulated for positive emotions (EtG-pos) and negative emotions (EtG-neg), and both were treated as individual independent variables in the analysis—the former being a positive measure of religiosity and the latter a negative measure.

We adapted the perception of group threat survey from Kanas et al. (2015) to the Malaysian context as a measurement of perceived interreligious threats. This section of the questionnaire contains twelve items that include both realistic threats (socio-economic opportunities and security, nine items) and symbolic threats (identity and way of life, three items). However, in the analysis, we make no distinction between realistic and symbolic threats for the aforementioned reason, and perception of interreligious threats is treated as a singular outcome variable in our study. Participants were asked to rate all items on a five-point Likert scale (1 = totally disagree; 5 = totally agree), and a higher score reflects a greater perception of interreligious threats. There is also an option of “not applicable” for items that do not yet relate to the participant’s stage of life. For instance, a student who is still studying in university may find the following item to be irrelevant: “I am worried that my chances of getting a job promotion will decline due to the presence of other religious groups.” In such cases, the item will then be excluded from the calculation of the average score for this variable.

2.3. Participants

Participants were recruited for the study via convenience sampling through the researchers’ institutions of higher education and networks. The participants were either students at the institutions or recent graduates. This group of participants provides an interesting target group to study perceptions of threats as they are competing for opportunities in educational institutions and the marketplace while going through a transitional phase of identity formation. Considering that religion seems to have played a significant role in the recently-concluded general election that saw those between ages 18 to 20 voting for the first time, it is also particularly valuable to study how this young group of individuals perceive interreligious threats. The call for participation was sent out via the universities’ communication channels and personal correspondence with alumni. After data cleaning to remove incomplete and irrelevant records, 260 valid responses (166 female, 63.4%) were retained for analysis. The distribution of respondents based on demographic information is shown in

Table 1. Demographic data of the respondents.

Category		Frequency (f)	Percentage (%)
Age	18–25	238	91.5
	26–35	18	6.9
	36–45	3	1.2
	46–55	0	0.0
	>55	1	0.4
Total		260	100
Religion	Islam	67	25.8
	Christianity (Protestants and Catholics)	64	24.6
	Chinese religions (Buddhism and Taoism)	73	28.1
	Indian religions (Hinduism and Sikhism)	10	3.8
	No religion	45	17.3
	Others	1	0.4
Total		260	100
Ethnicity	Chinese	157	60.4
	Indian	23	8.8
	Malay	45	17.3
	Indigenous people in Peninsular Malaysia	3	1.2
	Others	32	12.3
Total		260	100

.

2.4. Statistical Analysis

The data were analysed using IBM SPSS Statistics (version 28) with the confidence level set at 95% (p < 0.05). Preliminary analyses, including a test of normality and descriptive statistics, were run to provide an overview description of the data collected. Bivariate correlational analysis was carried out to examine the relationship between the variables of the study: CRS, EtG-pos, EtG-neg, and Threat. The initial findings from the correlational analysis then led us to pursue further analysis of the data with a hierarchical multiple linear regression and mediation analysis. The mediation analysis was carried out using the PROCESS macro (version 4.2) extension in SPSS to examine if any mediation effect exists among the independent variables (Hayes 2013).

3. Results

3.1. Preliminary Analyses

Descriptive statistics were generated for the four variables: CRS, EtG-pos, EtG-neg, and Threat, as shown in

Table 2. Descriptive statistics for all variables.

Scales	Min	Max	M	SD	Skewness	Kurtosis	α	No. of Items
1.CRS	1.1333	5.0000	3.3269	1.0156	−0.236	−1.094	0.952	15
2.EtG-pos	1.0000	5.0000	3.3641	1.2201	−0.564	−0.748	0.971	9
3.EtG-neg	1.0000	5.0000	2.3242	0.9636	0.581	−0.227	0.893	7
4.Threat	1.0000	5.0000	2.1451	0.9565	0.716	0.050	0.931	12

Centrality of Religiosity Scale (CRS); Positive Emotions towards God (EtG-pos); Negative Emotions towards God (EtG-neg); Perception of Interreligious Threats (Threat).

.

The Kolmogorov-Smirnov test of normality shows that all the variables are not distributed normally (p < 0.001). However, for the purpose of regression and mediation analyses, we referred to the Central Limit Theorem and assumed that, due to the sample size exceeding 200 (N = 260), the distribution of residuals in the data will approximate normality (Altman and Bland 1995; Ghasemi and Zahediasl 2012).

3.2. Bivariate Correlational Analysis

Table 3. Correlation matrix for all variables.

	1	1a	1b	1c	1d	1e	2	3	4
1.CRS	1
1a IDL	0.719 ***	1
1b ITL	0.589 ***	0.402 ***	1
1c PUB	0.740 ***	0.571 ***	0.506 ***	1
1d PRI	0.786 ***	0.616 ***	0.488 ***	0.618 ***	1
1e EXP	0.745 ***	0.626 ***	0.410 ***	0.549 ***	0.658 ***	1
2.EtG-pos	0.700 ***	0.625 ***	0.437 ***	0.582 ***	0.659 ***	0.659 ***	1
3.EtG-neg	0.276 ***	0.260 ***	0.138 **	0.247 ***	0.280 ***	0.300 ***	0.268 ***	1
4.Threat	−0.130 **	−0.144 ***	−0.07	−0.13 **	−0.137 **	−0.085	−0.136 **	0.051	1

**—p ≤ 0.01; ***—p ≤ 0.001; Centrality of Religiosity Scale (CRS); Ideology (IDL); Intellectual (ITL); Public Practice (PUB); Private Practice (PRI); Experience (EXP); Positive Emotions towards God (EtG-pos); Negative Emotions towards God (EtG-neg).

shows the bivariate correlation matrix for all the variables. Confirming H1, centrality of religiosity has a statistically significant negative correlation with perception of interreligious threats (p < 0.001). This confirms that individuals for whom religion occupies a more central position tend to demonstrate lower perception of threats from other religious groups. However, out of the five dimensions that the CRS measures, only three—ideology (IDL), public practice (PUB), and private practice (PRI)—show statistically significant negative correlation with perception of threats. The intellectual (ITL) and experiential (EXP) dimensions have no statistically significant correlation with perception of threats.

As hypothesised in H2, positive emotions towards God has a statistically significant negative correlation with perception of threats. It also demonstrates a strong and statistically significant positive correlation with CRS, confirming Huber and Richard’s (2010) findings. Negative emotions towards God also correlates significantly and positively with both CRS and positive emotions towards God, though the strength of correlation is weaker. However, what is of greater interest is that it has no statistically significant correlation with perception of interreligious threats.

3.3. Regression Analysis

Linear regression analysis was performed to test if the two measures of religiosity with statistically significant correlation with perception of interreligious threats (CRS and EtG-pos) also significantly predicted the outcome variable. It was found that CRS significantly predicts perception of threats at β = −0.161 (p = 0.006), while EtG-pos significantly predicts perception of threat at β = −0.142 (p = 0.003).

At the second stage of regression analysis, negative emotions towards God was added in the second block of the hierarchical multiple linear regression as a second predictor to test if it improves the predictive ability of the models. The new models were statistically significant for both CRS + EtG-neg (R² = 0.061, F(2, 257) = 8.388, p < 0.001) and EtG-pos + EtG-neg (R² = 0.067, F(2, 257) = 9.235 p < 0.001), as shown in

Table 4. Hierarchical multiple linear regression analysis.

Predictors	β (Model 1)	β (Model 2)
CRS	−0.161 **	−0.236 ***
EtG-neg		0.194 **
R2	0.029	0.061
ΔR2		0.032 **
EtG-pos	−0.142 **	−0.208 ***
EtG-neg		0.202 **
R2	0.033	0.067
ΔR2		0.034 **

**—p < 0.01; ***—p < 0.001; Centrality of Religiosity Scale (CRS); Positive Emotions towards God (EtG-pos); Negative Emotions towards God (EtG-neg); R² is R-squared value; ΔR² is the change in R-squared value between model 1 and model 2.

.

With the addition of EtG-neg, both cases report statistically significant improvements in R², thus affirming that EtG-neg significantly affected the predictive ability of both independent variables. However, it is worth noting that the effect of EtG-neg carries a positive sign, which is the opposite of that for CRS and EtG-pos.

3.4. Mediation Analysis

The following observations led us to formulate H3—that negative emotions towards God acts as a suppressor (negative mediator) in the relationship between the positive measures of religiosity (CRS and EtG-pos) and perceptions of threat:

• EtG-neg has a statistically significant bivariate correlation with both CRS and EtG-pos, but it has no statistically significant bivariate correlation with the outcome variable.
• The inclusion of EtG-neg into the regression models resulted in an improved model with statistically significant change in R².
• The effect of EtG-neg bears an opposite sign from that of CRS and EtG-pos.

A suppressor variable (or negative mediator, because it has the opposite effect from a regular mediator) has also been called an “enhancer” (McFatter 1979), due to its ability to enhance the predictive power of the model through its inclusion. Pandey and Elliott (2010) described suppressors as a third variable that may not be significantly correlated with the outcome variable but increases the magnitude of the regression coefficient of other predictor variables and improves the predictive power when added to a model because it correlates strongly with the predictor variable(s). This description seems fitting to the observations made above, and hence, we further performed a mediation analysis to investigate this hypothesis. This relationship is conceptually visualised in Figure 1.

Results as shown in

Table 5. The role of negative emotions towards God (EtG-neg) in the relationship between Centrality of Religiosity Scale (CRS), positive emotions towards God (EtoG-pos), and perception of threats (Threat).

Paths	a		b		c		c’		ab
	β	SE	β	SE	β	SE	β	SE	β	SE
1. CRS—EtG-neg—Threat	0.3849 ***	0.5400	0.1942 **	0.6560	−0.1613 **	0.0578	−0.236 ***	0.0623	0.0747	0.0286
2. IDL—EtG-neg—Threat	0.3062 ***	0.0479	0.1779 **	0.0648	−0.135 **	0.0505	−0.1895 ***	0.0536	0.0545	0.0222
3. ITL—EtG-neg—Threat	0.2040 **	0.0644	0.1111	0.0626	−0.0747	0.065	−0.0974	0.0660	0.0227	0.0163
4. PUB—EtG-neg—Threat	0.2509 ***	0.0438	0.1742 **	0.0632	−0.1359 **	0.0453	−0.1796 ***	0.0574	0.0437	0.0181
5. PRI—EtG-neg—Threat	0.3108 ***	0.0432	0.2054 **	0.0653	−0.145 **	0.0462	−0.2089 ***	0.0497	0.0638	0.0239
6. EXP—EtG-neg—Threat	0.3155 ***	0.0443	0.1633 *	0.0666	−0.0831	0.0479	−0.1346 **	0.0518	0.0515	0.0235
7. EtG-pos—EtG-neg—Threat	0.3261 ***	0.0448	0.2019 **	0.0657	−0.1419 **	0.048	−0.2077 ***	0.0519	0.0658	0.0239

*—p < 0.05; **—p < 0.01; ***—p < 0.001; Centrality of Religiosity Scale (CRS); Ideology (IDL); Intellectual (ITL); Public Practice (PUB); Private Practice (PRI); Experience (EXP); Positive Emotions towards God (EtG-pos); Negative Emotions towards God (EtG-neg). a = effect of the predictor on the mediator; b = effect of the mediator on the outcome; c = total effect of the predictor on the outcome; c’ = direct effect of predictor on the outcome while controlling for the mediator; ab = indirect effect of the predictor on the outcome through the mediator.

were obtained through bootstrap sampling (5000) with a 95% confidence interval, and they show that negative emotions towards God has a significant mediating effect on the relationship between both positive measures of religiosity (CRS and EtG-pos) and the outcome variable with the following indirect effects: (1) CRS—Threat, β (SE) = 0.0747 (0.0286), 95% CI [0.0225, 0.1344]; (2) EtG-pos—Threat, β (SE) = 0.0658 (0.0239), 95% CI [0.0222, 0.1167]. In the case of CRS, negative emotions towards God also has a significant effect on the three dimensions (IDL, PUB, and PRI) that have statistically significant correlation with perception of threats.

However, instead of observing a decrease in absolute value from c to c’—as would be the case for a regular mediator—we observed a consistent increase in absolute value from c to c’ across all statistically significant cases above, which is due to the β coefficient of the direct effect and that of the indirect effect having opposite signs. This leads us to conclude that negative emotions towards God functions as a classic suppressor (Pandey and Elliott 2010) in the relationships between the positive measures of religiosity and perception of interreligious threats.

4. Discussion

The study had aimed to investigate the relationship between different measures of religiosity (centrality of religiosity and emotions towards God) and perception of interreligious threats. The data show that the relationship is a complex one and that the positive and negative psychological valences associated with religion make the greatest difference.

That the centrality of religiosity has a significant negative correlation with perception of threats from other religious groups is supported by much of the existing literature that finds religiosity to correlate positively with individual wellbeing (Edara et al. 2021; Baetz et al. 2004) and life satisfaction (Aditya et al. 2020; Bergan and McConatha 2001; Szcześniak et al. 2019). It may be inferred that individuals who feel greater life satisfaction are less likely to perceive external threats to their identity and wellbeing. However, of the five dimensions, only ideology, public practice, and private practice are significantly correlated with perception of interreligious threats. The intellectual and experiential dimensions are, arguably, personal and inward-looking by nature—the former is purely cognitive while the latter is concerned with feeling the divine’s presence and intervention in one’s life—and so they have no significant impact on one’s perception of external threats.

Nevertheless, Makashvili et al. (2018) had observed that those who were more religious tended to have a lower level of acceptance of outgroups. This suggests that it is not the level of religiosity that entirely predicts the perception of threats, but rather that the level of religiosity is indicative of the presence of certain emotions towards the divine. Individuals for whom religion holds a more salient position tend to experience both positive and negative emotions towards God at a higher frequency than those for whom religion holds a less salient position. As Huber and Richard (2010) pointed out, the existence of positive and negative emotions towards God do not cancel each other out. Instead, they are meaningfully linked. For instance, the feeling of forgiveness (release from guilt) is contingent upon one feeling guilty in the first place. Perceiving God as holy may also evoke both reverence (positive emotion) and fear (negative emotion) at the same time. When these emotions—whether positive or negative—are subsequently projected to others (even outgroups), they either contributed to or allayed perception of threats.

That which warrants greater in-depth discussion is the role that negative emotions towards God play in shaping the perception of interreligious threats in Malaysia. Though many studies looked specifically at feelings of anger towards God (Exline et al. 2011; Exline and Rose 2013; Aditya et al. 2022), there is very little existing literature on negative emotions towards God in general, and their psychological and sociological impacts. While both positive and negative emotions towards God tend to co-exist at a higher frequency in individuals for whom religion occupies a more salient position, they yield opposite effects on one’s perception of threats. While the positive emotions may allay the perception of threats from other religious groups, the presence of negative emotions may instead increase the perception of threats. However, it is uncertain if the existence of negative emotions towards God leads to greater perception of threats or if the presence of these negative emotions results from one’s perception of threats. For instance, it may be that one projects feelings of anger towards God to others from whom they perceive a threat, yet it may also be that one’s perception of threats from others translates into doubt or dissatisfaction towards God’s justice. There is the possibility of a bilateral affective relationship between these two variables.

The impact of religiosity on perception of interreligious threats is particularly important in the context of multiethnic and multireligious Malaysia, where religion also plays an immense role in the social and political life of the nation. While the Pew Research Center (2018) finds that the younger generation tends to be less religious in most parts of the world, this is not the case in Malaysia, where the young and old seem to be equally religious. If the recently concluded general election is any indication at all, it may even seem that the younger Muslims in Malaysia are more religiously zealous than their elders. Due to the intertwined nature of ethnicity and religion in Malaysia (and the practical benefits of identifying with a certain group), this phenomenon may have political implications and implications on ethnic relations in the country. In the weeks of campaigning leading up to the 15th General Election, Muhyiddin Yassin, former Prime Minister and the prime ministerial candidate for one of the political coalitions, made allegations that a rival party with an ethnic Chinese majority was working closely with a foreign Jewish group to realise a Christianisation agenda in Malaysia (The Star 2022) while his political partner, a leader of the conservative Islamic party, warned Muslims not to vote for any of the other political parties/coalitions, lest they go to hell (Palansamy 2022). These recent incidents at once highlight the complex intersection of religion, ethnicity, and politics in Malaysia, and exemplify the targeted use of interreligious threats as weapons to further certain political agendas.

In the case of Malaysia, the perception of interreligious threats seems to be built on the negative emotion of fear: fear of losing one’s identity and even the fear of divine punishment. Both incidents above seem to capitalise on these variations of the emotion of fear—the former sowing fear that Muslims may lose their religious identity if the Christianisation agenda were to be realised in Malaysia, and the latter evoking a sense of fear towards God who may mete out the punishment of hell if one did not behave (or, in this case, vote) appropriately—thus contributing to some form of perception of threats from other religious groups. The presence of this fear can even be traced all the way back to the founding of the nation. Malhi (2015) discussed the conflation of the 3 Rs—race, religion, and royalty—in Malaysia, and traced its historical roots back to the birth of the idea of a “Malay Muslim” and the establishment of the royal institution as “guardians of Malay religion and custom” (p. 474). The idea that the Malay religion and custom needed to be ‘safeguarded’ reveals certain degrees of perception of interreligious threats.

Various incidents in the past two decades also reveal a perpetuation of the same perception of threats from non-Muslims, particularly Christians, in Malaysia. In reviewing the freedom of religious expression in Malaysia, Mohd Sani and Abdul Hamed Shah (2011) examined a few prominent instances of interreligious conflicts in the nation, including the use of religion in politics; controversies surrounding the use of the word “Allah” by non-Muslims (particularly Christians) to refer to God, and the publication and distribution of the Bible in the national language; and repeated allegations of attempts by a certain Chinese-majority political party to ‘Christianise’ Malaysia. Arguably, all these instances of interreligious conflict stemmed from the emotion of fear. Use of the word “Allah” among the Christian community and the distribution of the Bible in the Malay language received opposition from the Islamic authorities and Islamic groups who feared that these may cause confusion among the Muslims and lead them away from the faith, reflecting a fear of divine punishment should Muslims question, or worse yet abandon, their faith. Likewise, the use of religion in politics often involves cautioning the majority Malay-Muslim community against the Christianisation agenda of certain groups, which capitalises on the fear of losing one’s Malay-Muslim identity. As such, considering the socio-political context of Malaysia, the findings of our study show that negative emotions towards God play the most significant role in shaping one’s perception of interreligious threats and offer us a better understanding of interreligious and interethnic conflicts in Malaysia.

5. Limitations

One limitation of the present study is the sample size and representation. With a sample size of 260, it is by no means representative of the population in Malaysia, though the respondents represent a specific subset of the society. The distribution of ethnicities and religions is also not representative of the population in Malaysia. A study of larger scale will need to be conducted in order to draw generalisable conclusions about the perception of threats between the different religious groups in Malaysia.

The present study is also a general description of the perception of threats in Malaysia and did not make comparisons between the religious groups. We decided to report on the general description instead of intergroup comparisons because some groups in our sample are too small for any meaningful comparison. Additionally, both the Muslim majority and non-Muslims alike have expressed that they feel threatened by other religious groups, and this is particularly clear in the recent political discourse of the nation, where both sides have used fear of other religious groups as a campaigning point (AR 2019). Therefore, we decided to pursue a broad analysis of the relationship between measures of religiosity and perception of interreligious threats. Nevertheless, considering that interreligious conflicts in Malaysia most often involve Muslims and Christians, a more targeted study exploring the perception of interreligious threats between these two groups may be beneficial in the near future. A qualitative study may offer a more explicit description of the link between negative religious emotions and perception of interreligious threats, and perhaps even reveal the factors that contribute to such a perception.

While we have statistically demonstrated the relationship between positive and negative measures of religiosity, and their effect on perception of interreligious threats, the role of negative emotions towards God is of peculiar interest, especially in how it shapes perception of interreligious threats. As such, it would be of value to further examine the impact of specific negative emotions towards God and the specific ways in which they interact with perception of interreligious threats, including exploring the possibility of a reciprocal relationship.

Although the present study has established a link between CRS and perception of interreligious threats, and between positive emotions towards God and perception of threats, it does not offer an elaborate model that captures the different dimensions and elements of religiosity that predict one’s perception of interreligious threats. This study merely highlights that negative emotions towards God, in general, suppress the effect of the positive measures of religiosity on perception of interreligious threats. It would perhaps be beneficial to conduct a richer analysis to better understand the interaction between different dimensions of religiosity and their collective impact on the outcome variable.

6. Conclusions

The current study contributes to a better understanding of the role that religiosity plays in predicting perception of interreligious threats. While positive measures of religiosity, like centrality of religiosity and positive emotions towards God, have a statistically significant negative correlation with perception of threats from other religious groups, negative measure of religiosity (in this case, the presence of negative emotions towards God) facilitates an opposite effect on that relationship. As religiosity is a complex and multidimensional concept, both positive and negative measures jointly offer a stronger ability to predict perceptions of threats from other religious groups. In discussing the findings within the socio-political context of Malaysia, the present paper shows how religion complicates the interethnic and interreligious relationships in the nation through its entanglement with politics. While we offer an interpretation of how negative religious emotions may have contributed to perceptions of interreligious threats in Malaysia, we intend to investigate this interpretation further through qualitative approaches, with the aim of eliciting more descriptive data to explain interreligious relations in Malaysia and the specific ways in which negative emotions contribute to the perception of interreligious threats.