Investigating a Relationship between Perceived Stress, Religious Coping, and Religiosity in Migrant Muslim Women

The last decade has seen the greatest increase in international migration across the globe (Triandafyllidou, 2018). The United Kingdom (UK) hosts the fifth largest migrant community globally, with over 9 million foreign-born residents (United Nations, 2017). The majority of the UK’s migrant community are from non-EU countries, the highest being from Pakistan, Nigeria, Bangladesh, and South Africa (Migration Observatory, 2015). Muslims make up the second largest group of migrants based on religion (Apostolova & Hawkins, 2017).

Migration in the context of this research is considered as the permanent or semi-permanent move of individuals from one place of residence to another (Bhugra & Ayonrinde, 2004). Previous research highlights the development of stress in the life of a migrant and the potential risk it plays in the development of mental health difficulties (Teodorescu, Heir, Hauff, Wentzel-Larsen, & Lien, 2012). According to the World Health Organisation (WHO; 2013), migrants are often exposed to displacement, difficulties in transit, dangerous travel, immigration uncertainty, and potential hostility which have great effects on stress levels. These stresses can occur at three stages of migration identified by Bhugra and Ayonrinde (2004), pre-migration, migration, and post-migration (also known as, settlement). Each of these stages may constitute factors that could lead to significant perceived stress in the life of a migrant.

The experience of perceived stress may differ depending on the reason for migration. In premigration stages, forced migrants may have been exposed to psychological trauma, physical torture, and adverse life events, which increase stress (Bhugra & Ayonrinde, 2004). Voluntary migrants may find the potential loss of social support, higher incidence of crime, and limited availability of employment in the potential host country to be stressful (Yijala, 2012; Tabor, 2010). During the process of migration, voluntary migrants have the advantage of planning to migrate which may significantly reduce potential stress (Ventriglio & Bhugra, 2015). However, for forced migrants, the process is often unplanned and many experience significant psychological trauma as they embark on a journey to safety in a foreign land. Some lose their lives or loved ones and others are scarred emotionally (Foster, 2001).

Settling into the host country could further expose a migrant to perceived stress; for some migrants, this may be a direct impact of the post-migration stage such as discrimination, culture shock/acculturative stress, immigration status, and lack of social and economic support (Li, 2015; George & Jetner, 2014; van der Ham, Ujano-Batangan, Ignacio, & Wolffers 2014; Yijala, 2012). For others, it may be due to trauma experienced during pre-migration and/or migration processes (Li, 2015). Constant exposure to stress leads to chronic perceived stress which is a risk factor for other mental health conditions such as depression and anxiety (Michl, McLaughlin, Shepherd, & Nolen-Hoeksema, 2013). Considering that most migrants leave their home countries to escape war, political instability, and generally to seek a better life, this stress can be counterproductive as it may mitigate the purpose of migration.

Discrimination may lead to increased perceived stress in migrants (Sellers, Caldwell, Schmeelk-Cone, & Zimmerman, 2003). Muslims in the UK currently face the greatest social mobility restrictions, which have been attributed to discrimination (Stevenson, Demack, Stiell, Abdi, Clarkson, Ghaffar, & Hassan, 2017). Nearly half of the Muslim population in the UK (46%) live in Britain’s most deprived areas; an increase from 33% in 2001. These areas offer poorer choices for education and health care and have higher crime rates. In England, such areas have been found to have a lower life expectancy with some areas as much as 10 years difference (Smith, Noble, Noble, Wright, McLennan, & Plunkett, 2015). Following the UK’s EU referendum, 31 police forces across the UK found religious and racial aggravated crimes increased by 41%, when compared to the previous year (from 3,886 to 5,568; Corcoran & Smith, 2016). Goodwin and Milazzo (2017) found some individuals who voted to leave the EU were largely considered to be motivated by anti-immigration sentiment. This may lead to the perception of increased hostility toward migration. The Muslim community in Britain have also experienced an increase in religious aggravated attacks in the last decade (Mend, 2014). According to Tell MAMA (2018), a project that records anti-Muslim attacks across England, there were 1,201 reports of verified Islamophobic attacks between January and December 2017 with 61% of victims being women, 75% of those women wearing hijab.

The focus of this study is on the experience of migrant Muslim women. Chandra (2011) suggested female migrants have higher stress levels, which were attributed to being less likely to assimilate in the host country. Migrant women are often viewed by their communities as the carriers of culture and traditions and are often expected to adhere to the gender roles expected of them in their home countries (AbdurRaqib, 2006). The migration of Muslim women to the West has been associated with family reunification, as in most cases, Muslim men are the first to migrate before their families join them abroad (Ferrero, 2018). Muslim women in the UK are more likely than all other women in the UK to be economically non-active. Muslim women disproportionally are full-time homemakers; 18% of Muslim women are recorded as looking after the home and family, compared to only 6% in the general population (Muslim Council of Britain, 2015). Low socioeconomic status (SES) has been demonstrated in the literature to have a significant effect on perceived stress levels (Algren, Ekhlom, Nielsen, Ersboll, Bak, & Anderson, 2018) and many migrant Muslim women fall into a low SES group. There is also a struggle with language, as the 2011 census, reported that 22% of Muslim women said they spoke English “not well” or “not at all” (Stokes, 2013). This may affect their ability to access health services, education, and work. Sullivan and Kasubeck-West (2015) found that difficulty with language was linked to higher stress. The challenges faced by female Muslim migrants may lead to increased stress and a greater increase in perceived stress. Perceived stress is the level at which an individual deems situations in their life to be stressful (Lazarus & Folkman, 1984). Perceived stress refers to the degree to which one identifies one’s life situation as stressful (Cohen, Kamarck, & Mermelstein, 1983).

Muslim women in Britain appear to be exposed to various risk factors that may lead to increased perceived stress. However, Muslims in the UK have one of the lowest rates of accessing mental health services (Joint Commissioning Panel for Mental Health, 2014). The reason why appears to be inconsistent across current research. Various studies have explored the role of stigma in preventing Britain’s Muslim community from accessing services (REACH Community Health, 2008; Knifton, 2012). However, the results of these studies are not representative of the diversity of Britain’s Muslim population. Other studies (Laird, Amer, Barnett, & Barnes, 2007) have tackled the disparities in health care systems and suggest Muslims do not necessarily have access to health care services. Such studies have also looked at discrimination within the health care system.

There is growing research on the role of religion in preserving good mental health. Ellison and Levin (1998) proposed the Religion-Health Connection theory which suggests that behavioral and psychosocial constructs of religion lead to positive mental and physical health outcomes. When exposed to stress, individuals often adapt means or strategies which could reduce how stress is perceived, known as ‘coping’ (Cohen, Kessler & Gordon, 1997). The research on coping and its positive effects on perceived stress is increasing, with investigation into methods of coping including meditation and mindfulness (Biegel, Brown, Shapiro, & Schubert, 2009). Recent research has explored the role religion plays in reducing perceived stress. It is possible that the use of religious coping may be a preventive measure used by female Muslims to reduce perceived stress. Further consideration of religious coping would be productive.

Lambert and Dolhaite (2006) define religiosity as “a person’s spiritual beliefs, religious practices, and involvement with a faith community”. Religious coping on the other hand, is defined by Pargament, Smith, Koenig and Perez (2000) as responses to stress that are spiritually and religiously based and, can be cognitive, behavioral, and interpersonal. Religious coping could be positive or negative. Positive religious coping is the expression of spirituality with self and others, connection with God, and the belief that there is a meaning to life. It may include benevolent religious appraisals, collaborative religious coping, seeking help from God, religious forgiveness, and support from clergy and members (Pargament et al., 2000). Negative religious coping, on the other hand, is described as the expression of a less secure attachment with God, an ominous view of the world, and a struggle in the search for significance. Negative religious coping strategies often include passive religious deferral, spiritual discontent, appraisal of punishment from God, self-directing religious coping (seeking control through personal initiative), interpersonal religious discontent, and pleading for direct intercession (Pargament et al., 2000). Pargament and colleagues (2000) found positive religious coping to be associated with lower symptoms of psychological distress and higher levels of stress-related growth, while negative religious coping was associated with higher levels depression and lower quality of life.

Eliassen, Taylor, and Lloyd (2005) found higher levels of stress in individuals with moderate religiosity compared to those who were highly religious. They attributed higher stress to individuals having difficulty in their understanding of, commitment to, and contentment with religion, possibly leading them to adopt negative religious coping strategies. Research has shown that for many people of faith, religion acts as a source of comfort in stressful times (Van Dyke & Elias, 2007; Pargament, 2002). Many studies have been conducted looking at the role of religious coping in managing or coping with stressful events. A meta-analysis of 49 studies with a total of 13,512 participants by Ano and Vasconcelles (2004) found that positive religious coping was significantly related to positive adjustment to stress. However, their analysis was based on a predominantly Christian population. Only 15% of the samples were from non-Christian faiths. Gardner, Krageloh, and Henning (2013) explored religious coping, quality of life, and perceived stress in Muslim international students and local students in New Zealand and found that international Muslim students used more positive religious coping and had less perceived stress than local students.

Barhem, Younies, and Muhamad (2009) describe Islam as being a complete way of life. Many Muslims adopt the continuous remembrance of Allah, to deal with stressful life events (Mujahid, 2006). This is known as ‘dhikr’ and constitutes supplication, recitation of the Qur’an, and praising God which are prescribed as methods to deal with stressful times. As stated in Qur’an 13: 28 (Oxford World’s Classics edition) “truly it is in the remembrance of God that hearts find peace” and Qur’an 2: 286 (Oxford World’s Classics edition), “God does not burden any soul with more than it can bear”. It is evident that Islam aims to provide direction for its followers in dealing with distress and reducing perceived stress.

Although previous studies have explored the relationship between religiosity, religious coping, and stress, few have explored a relationship with ‘perceived stress’ or its impact on migrant Muslims. There appears to be a gap in the literature when looking toward the role religious coping and religiosity play in the perceived stress of migrant Muslims. Considering the growing hostility toward migrants, and Muslim migrant women specifically, who may be exposed to heighted stressful events. Thus, based on previous findings and the Religion-Health Connection theory by Ellison and Levin (1998), the current research investigates the relationship between migration and perceived stress. It further explores the mediating effect of religiosity on the relationship between religious coping and perceived stress in migrant Muslim women. The following are predicted:

1. There will be an association between migration and perceived stress.

2. Participants with higher religious coping will have lower levels of perceived stress.

3. Religiosity will mediate the relationship between religious coping and perceived stress.

Method

Participants were recruited for the study through a non-profit community organization, which provides support and advice to minority ethnic groups in an English city (population 250, 000). The charity provides support to approximately 500 people through its community outreach projects and 251 people through its support and advice center. Around 70% of center participants are women, two-thirds of whom are Muslim women (A. Sanneh, personal communication, May 11, 2018). Through the center, the researcher was able to recruit an ethnically diverse sample. Participants could only take part in the survey if they identified as Muslim women, were over the age of 18, and had been born outside the United Kingdom (UK). In return for completing the study, participants had a choice to enter a draw to win one of five £10 Amazon vouchers.

Participants

Forty-one women aged 18 to 57 years (M=30.44, SD = 8.99, Mode =33 years) were recruited. The majority of participants (n=31) had indefinite leave to remain (ILR) in the UK or were British citizens by naturalization; ten were temporary residents in the UK. The sample is a diverse representation of participants who accessed the charity for support with over half being of South Asian descent (n=22), Middle Eastern (n=6), West-African (n=6), East African (n=6), and North African (n=1). The majority of participants were married (n=28), eleven were single, one was divorced, and one was widowed. Most participants had received at least a secondary-school education (n=30), and many had children (n=26).

Only sixteen participants had previously lived in a different country other than their home country prior to migrating to the UK. Two had lived in other Western countries, twelve in non-Western countries, and two had lived in both other Western and non-Western countries. Over half of participants (n=23) had lived in the UK for at least five years, while eighteen had lived in the UK for fewer than five years.

Materials

Perceived Stress Scale

To measure the perceived stress of respondents, the Perceived Stress Scale (Cohen, Kamarck, & Mermelstein, 1983) was used. The Perceived Stress Scale measures the degree to which an individual views a certain situation as stressful. The scale is widely used and has demonstrated satisfactory reliability and validity in non-Western samples (α =.80; α = .89) (Almadi, Cathers, Mansour, & Chow, 2011; Mimura & Grifiths, 2008). The original scale is a 14-item questionnaire on a 5-point Likert scale (0- never through 4- very often). The ten-item shortened version (PSS-10) was used in the current study, with the total scores achievable on the perceived stress scale range from a minimum of 0 to a maximum of 40. Scores 0–13 indicate low perceived stress, 14–26 indicates moderate perceived stress, and 27–40 indicate high perceived stress.

The COPE Inventory

The COPE inventory (Carver, Scheier, & Weintraub, 1989) is a multidimensional scale assessing people’s responses to stress. The COPE Inventory consist of fifteen subscales; the current study used the ‘Religion’ and ‘Emotional Support’ subscales. The religion sub-scale has had good reliability and validity when used for non-Western samples (α =.81) (Yusoff, Low, & Yip, 2010). These subscales assessed the use of religion as a means of coping with distress. Each scale consisted of 11 items on a 4-point Likert scale (1- ‘I usually don’t do this at all’ to 4- ‘I usually do this a lot’). For the Religious subscale participants could score a minimum of 4 and a maximum of 16. For Emotional Support, participants could score a minimum of 7 and a maximum of 28. Overall, participants could have a minimum score of 11 and maximum score of 44. Scores on all items are summed to give a total COPE score, which if high, indicates high religious coping and if low, indicates low religious coping.

The Centrality of Religiosity Scale

Religiosity was measured using the Centrality of Religiosity Scale (Huber & Huber, 2012). The scale measures the centrality and importance of religion to participants by measuring religious practices and engagement. The scale measures intellect, ideology, public practice, private practice, and experience of religion. It has been modified for use on the major world religions. It was developed in German and the scale has been translated into 19 different languages. This study made use of the 15-item English scale as it allows for the measurement of core dimensions of religiosity and has an excellent reliability when used in cross-cultural contexts (α =.96; Huber & Huber, 2012). Responses are on a 5-point Likert scale (1- ‘Not at all’ to 5- ‘very much’). To calculate centrality of religiosity, participants responses on each question is summed to produce a total score. Participants scores can range from a minimum of 15 to a maximum of 75. Higher scores indicate higher central religiosity, and lower scores indicate a lower central religiosity. If using grouping methods, each participant’s total score is divided by number of questions on the scale to achieve an average score. Participants who score between 1.0–2.0 are considered ‘not religious’, 2.1–3.9 are considered ‘religious’, and 4.0–5.0 are considered ‘highly-religious’. The centrality of religiosity is simply referred to as ‘religiosity’ throughout this paper.

To check reliability of the scales used with the current sample, the Cronbach’s alpha of each scale was analyzed. For Religious coping, internal reliability was good (α = .86), for Centrality of Religiosity reliability was excellent (α = .95), and the Perceived Stress Scale also had an excellent reliability (α = .94).

Procedure

The current study was granted ethical approval from the University of Southampton’s Ethics and Governance Committee (ERGO). Data collection was conducted over the course of three months. Permission was granted by the community center for the study to be advertised to its service users. The researcher attended services provided by the charity where the study was verbally advertised to center participants. Those who wished to take part were provided with a participant information sheet, a consent form, and a four-part questionnaire (demographics, centrality of religiosity, perceived stress scale, and COPE inventory - a total of 48 questions), which were all paper-based. Five participants required assistance filling in questionnaires due to difficulty understanding the wording of some questions and, this assistance was provided by volunteers at the center. After completion of the questionnaire, participants were provided with a debrief sheet.

Statistical Analysis

Descriptive statistics were conducted on all measures. Cronbach’s alpha was used to test reliability and validity of all scales. A one-way ANOVA was used to test for the effect of migration on perceived stress. This was calculated individually for time spent in the UK, migration status, and whether participants had lived in another country other than the UK and their home country.

To test the main hypothesis of whether the relationship between religious coping and perceived stress was mediated by religiosity, a mediation analysis was used. The mediation model proposed was done using PROCESS Macro.

Mediation

The proposed mediation model was analyzed using PROCESS with bootstrapping.

Figure 1

The proposed mediation model

Note. Figure 1 shows the proposed mediation model. It is predicted that there will be an indirect effect between religious coping and perceived stress when mediated by religiosity. The colored line indicates direct effect. Black lines indicate indirect effect as a result of mediation.

Results

Migration and Stress

Tests of differences were used to investigate the relationship between participants’ migration and total Perceived Stress scores. The assumption of independence was met and data was at the interval level. Due to having a small sample size, the Levene’s test for homogeneity of variance was adopted and assumption of variance was met. Through the visual inspection of box plots, there were no outliers in the data and perceived stress scores across the groups seemed to slightly deviate from normality. However, the Shapiro-Wilks test was not significant, and as the deviations did not seem extreme from visual inspection, normality was assumed. Therefore, a one-way ANOVA was used.

Table 1

Mean and standard deviation of total perceived stress scores across migration groups.

	Perceived Stress
Time in UK	Mean	Standard Deviation
Fewer than 5 Years	24.67	4.53
More than 5 years	22.11	5.99
Lived in Other Countries
Yes	24.92	5.78
No	22.82	5.16
Immigration Status
ILR/Citizenship	23.38	5.99
Temporary Visa	23.57	4.18

Note. The table above indicates the mean scores and standard deviation for perceived stress in participants migration groups.

Time in the UK and Perceived Stress

A one-way ANOVA was used to compare the effect of (IV) ‘time in the UK’ on (DV) ‘perceived stress’ in women who had lived in the UK for fewer than 5 years, and women who had lived in the UK for more than 5 years. There was no significant effect of time in the UK on perceived stress, F (1, 39) =.017, p =.898. This suggests that participants perceived stress did not differ based on the time they had spent in the UK.

Migration Status and Perceived Stress

A one-way ANOVA was conducted to compare the effect of (IV) ‘migration status’ on (DV) ‘perceived stress’ in women who had acquired British citizenship or indefinite leave to remain (ILR), and women who were in the UK on a temporary visa. There was no significant effect of migration status on perceived stress, F (1, 39) = 1.177, p =.285., suggesting that participants perceived stress did not differ based on their migration status.

Previous Migration and Perceived Stress

A one-way ANOVA was conducted to compare the effect of previous migration on ‘perceived stress’ in women who had previously lived in a country other than their home country and the UK, and women who had only lived in their home country and the UK. There was no significant effect of previous migration on perceived stress, F (1, 39) =.567, p =.456. This suggests that participants perceived stress did not differ based on whether they had previously lived in a country other than their home country and the UK.

When comparing the type of countries participants had previously migrated to, there was no significant difference between total perceived stress of participants who had previously lived in Western, non-Western countries, or those who had previously lived in both, F (2, 13) =2.384, p =.131. In general, perceived stress scores did not differ across participants’ migration status, time in the UK, or previously migrating to other countries before the UK.

Religious Coping, Perceived Stress and Religiosity

Data was explored to check if it met assumptions for correlational analysis. Visual inspections of scatterplot indicated a linear relationship between variables. There were no univariate outliers and the plots showed absence of bivariate outliers as assessed using Mahalanobis distance. Measures in the study were assumed to be independent. Histograms showed that religiosity, perceived stress, and religious coping scores were not distributed normally across participants. The Shapiro-Wilks tests for all variables were significant further suggesting deviations from normality. However, further analysis involved bootstrapping which is robust to problems with normality.

Mediation Analysis

Table 2

The mean and standard deviations of perceived stress, religious coping, and religiosity.

	Perceived Stress		Religious Coping
Religiosity Groups	Mean	Standard Deviation	Mean	Standard Deviation
Religious	30.85	6.64	32.23	6.56
Highly Religious	14.04	7.42	39.18	3.93

Note. The table above shows means and standard deviations of participants’ perceived stress and religious coping scores based on religiosity groups.

A mediation analysis was carried out using PROCESS Macro v3.1 (Model 4; Hayes 2013) to test whether religiosity mediated the effect of religious coping on perceived stress. The total effect of religious coping on perceived stress was negative and significant, F (1, 39) = 22.73, p < .001, R2 =.37, b = −1.11, SE = 0.23, 95% CI [−1.58, −.64]; the higher religious coping participants reported, the lower the perceived stress. There was a significant positive relationship between religious coping and religiosity, F (1, 39) = 34.02, P < .001, R2 = .47, b =1.33, SE = .23, 95% [.87, 1.80]. This suggests that the higher religious coping participants reported, the higher degrees of religiosity. The relationship between religiosity and perceived stress was negative and significant, b=−0.55, SE=.14, 95% CI [−0.83, −0.27], suggesting that higher religiosity rates reported lower perceived stress. There remained a significant negative relationship between religious coping and perceived stress, F (2, 38) = 23.64, p<.001, R2 = .55, b= −.55, SE = 0.14, 95% CI [−.83, −0.27], when controlling for the mediator, religiosity. However, when the effect of religiosity was removed, the relationship between religious coping and perceived stress was no longer significant, b=−.37, SE = .27, 95% [−.92, .18]. This suggests that the effect of religious coping was fully mediated by religiosity. To check if the indirect effect was significant, bootstrapping estimation with 1000 samples was used. As the confidence intervals did not cross 0, the results showed significant indirect effect, suggesting that use of religious coping may affect levels of perceived stress via religiosity, b = −.74, β = 0.40, bootstrapped SE = .23, bootstrapped 95% CI [−1.19, −0.29]. Overall, religious coping and religiosity explained 55% of the variance in perceived stress.

Figure 2

The mediation model

Note. Figure 2 shows the relationship between religiosity and perceived stress, showing unstandardized regression coefficients. * p<.001. 95% confidence interval are bootstrapped based on 1000 resamples. C stands for of the direct effect mediator.