The role of demographic factors on religious beliefs: Evidence from five countries

Demographic factors and religious beliefs

Gender

Numerous studies on the relationship between gender and religion have yielded mixed results. According to one study by the University of California, Los Angeles, women have a stronger belief in God and higher levels of religious participation than men (Bryant, 2007). This finding is primarily due to traditional gender roles that place women in nurturing and caring roles, which can lead to a greater reliance on religious beliefs for comfort and support. Trzebiatowska and Bruce’s (2012) study discovered that men are more likely to identify as religious leaders or hold positions of power within religious organizations. Other research has found that gender differences in religious beliefs and practices are not universal but rather are influenced by cultural and historical factors (Winkel, 2019). For example, a study conducted in Iran discovered that women play a more prominent role in religious practices than men because they are perceived to have a closer connection to God due to their maternal role (Betteridge, 2002).

Age

Research has shown that people's religious views shift over time, often due to personal experiences (Butėnaitė, 2019). Older adults are generally more committed to their religious beliefs than younger adults (Meza, 2020). This finding is frequently due to more time spent reflecting on their beliefs and developing a deeper understanding of their religion (Hayward and Krause, 2015). On the other hand, the relationship between age and religious beliefs is more complex. According to some research, as people age, some become more flexible and open to different religious perspectives, while others become more rigid in their beliefs (Schaie and Willis, 1991).

Countries

Religious beliefs vary significantly between countries and regions (Behere et al., 2013). The levels of religious belief in the various religions vary among the five nations of Malaysia, Indonesia, China, Kenya, and the Philippines. It is necessary to examine nation's demographics, cultural influences, and historical backgrounds to determine which religion tends to have the highest-level religious belief.

Indonesia and Malaysia are located in Southeast Asia, and they share many similarities because of their proximity in distance. Islam is the dominant religion in Malaysia and Indonesia, with most of the population identifying as Muslim (Amran et al., 2017; Lee et al., 2020). Approximately 61% of Malaysians identify as Muslim (Ng et al., 2020), with over 90% of the population identifying as Muslim in Indonesia (Lee et al., 2020). Despite the presence of diverse religions such as Christianity, Buddhism, and Hinduism in these countries, Islam significantly impacts the cultures of Indonesia and Malaysia because people of all religions are expected to adhere to specific Islamic rules. The high levels of religious belief in these countries can be attributed to various factors, including the influence of Islamic culture and tradition, and local communities' firm adherence to religious teachings and practices. Muslims, for example, are encouraged to attend the mosque regularly, observe religious holidays and rituals, and pray daily, all of which contribute to a strong sense of religious identity and belief (Hopkins et al., 2023).

Conversely, China is predominantly irreligious, with only 14% of the population adhering to religious beliefs (Yang and Huang, 2018). Nonetheless, religious freedom is guaranteed by the constitution. The laws protect individuals' right to believe in - or not believe in - any particular faith (Liang, 2021). It is often difficult to pinpoint what a Chinese person believes in religion. Some religions, such as Buddhism and Taoism, have deep roots in Chinese civilization (Luo and Chen, 2021). Because of China's polytheistic tradition, individuals or even believers may have blurred religious boundaries and worshipped both Buddhist and Taoist gods (Zhang and Lu, 2020; Chen, 2001). Apart from that, individuals who attend religious services may still see themselves as atheists as they do not perceive themselves as belonging to any particular religious institutions and never really associate their conduct with the term “religion”. In their study in Taiwan (China), Zhang and Lin (1992) discovered that while 40% of survey participants claimed they did not believe in God, 70% participated in religious events and rituals such as god-worship. According to Marighella (2020), Chinese millennials may be unaffiliated with forms of religion but are engaging with Buddhism in their lives, not as a religious belief but as a spiritual and cultural journey through life under pressure caused by rapid changes.

Kenya is an East African country with the equator running through it. The country shares borders with five other countries: Uganda, Tanzania, Ethiopia, South Sudan, and Somalia. Kenya is governed by the 2010 Constitution, which recognizes the supremacy of the Almighty God and, along with other laws, prohibits religious discrimination and protects religious freedoms. Religion pervades every aspect of African society, with nearly all political, economic, and social activities bearing a religious overtone (Alwala, 2022). Christianity was introduced to Kenya in 1844 by missionaries who settled near Mombasa, and the country now has several Christian denominations (Gathogo, 2020). Islam, on the other hand, is Kenya's second most widely practiced religion (Iyer and Weeks, 2020). Kenyans identify culturally with their tribe or ethnic group, and most members of the same ethnic group share religious beliefs (Lynch, 2006). There are churches in almost every town and village in the country, and most towns have at least one mosque where people can worship. This situation demonstrates the impact and spread of Christianity in Kenya.

Religion is deeply ingrained in the culture and daily practices of many Filipinos in the Philippines. The Filipino people are the tenth most religious in the world, and nearly 87% place a high value on religion (Bagaoisan, 2016). This is because Spain exerted colonial rule over the country for three centuries during which time the dominant religion was Catholicism. The Philippines is one of only two Asian countries with a predominantly Catholic population (Hoh, 2018). Many Filipinos have long relied on various religious beliefs, practices, and attitudes to assist them in their daily lives or to help them cope with the difficulties and problems they face. Gonzalez (2009) stated that many people attend church and other religious events to live morally. Some people turn to religion for assistance, guidance, and miracles (Lagman et al., 2021). Pilgrimages, saints, and charismatic gatherings can also provide access to divine power (Brion et al., 2018). For example, it has become a tradition in many parts of the Philippines to celebrate patron saint feast days and Catholic holidays to celebrate bounty and protection, and to bring families and friends together. According to Bagaoisan (2016), many Filipinos use their religion to cope with issues and problems. When Filipinos have no other option but to turn to God, they usually say, “Ipagpasa-Diyos na natin,” or “they turn over their situation to God” (Cariñgal et al., 2022).

In conclusion, religious beliefs vary significantly across countries, but Islam and Christianity are more prevalent in Malaysia, Indonesia, and Kenya, while Catholicism is the dominant religion in the Philippines. Religious beliefs are generally weaker in China, with many people viewing religion as a personal choice rather than a necessary part of their daily lives. Finally, the strength of religious beliefs is determined by a complex combination of cultural, historical, and personal factors that cannot be readily determined by looking at statistics.

Employment status and household income

Studies indicate that unemployed or underemployed people are more likely to have a stronger sense of religious beliefs (Christy et al., 2017). This could be because unemployment or underemployment can cause feelings of powerlessness, uncertainty, and existential anxiety, leading people to seek comfort and stability in religion (Zheng et al., 2020). On the other hand, those who are highly employed and have a higher income may be less likely to participate in religious activities and have a weaker sense of religious beliefs (Lipford and Tollison, 2003). This is because financial security and success can reduce the need for religious comfort and stability (Barber, 2013). Studies conducted during the COVID-19 pandemic revealed that Filipinos relied heavily on their religious beliefs as a means of coping. del Castillo et al. (2021) observed the practice of ‘Dungaw’, or the veneration of sacred images and placing them outside the window of a house or a church, as a coping strategy for difficulties. ‘Dungaw’ in Tagalog means ‘to look’ or ‘to peep out’, and in the context of religious belief, it is often used to describe the idea that saints or Jesus himself is watching over and protecting you from a house or window. Tanucan and Bojos (2021) also noted how Filipinos in slum areas use religion as a source of strength to survive the pandemic's hardships. It is important to note that these are general trends and that each person's experiences and beliefs are unique. However, these findings suggest that an individual's employment status and household income can significantly impact religious beliefs and behaviours.

Marital status

Although the effects vary depending on the individual and their cultural context, marital status can significantly impact religious beliefs. Marriage can increase religious involvement in some cases, as couples may attend religious services together and participate in faith-based activities (Kim and Dew, 2019). As partners support and encourage each other in their beliefs, this can form a shared religious identity (Alghafli et al., 2014). Nonetheless, there is evidence that marriage can have the opposite effect, leading to a decline in religious activity when spouses come from divergent religious backgrounds (Elmali-Karakaya, 2022). In these situations, the couple may compromise or take a more eclectic approach to their beliefs. Being single may impact on religious beliefs because one can have more time and freedom to explore different life perspectives or question their beliefs (Apostolou and Christoforou, 2022). Single people may be more likely to seek out supportive communities, including religious communities, to meet their social and spiritual needs (Himawan et al., 2018a, 2018b). Divorce can also significantly impact on religious beliefs because individuals may experience guilt, loss, and anger (Tuttle and Davis, 2015). To sum up, being married can profoundly affect religious beliefs, though these can be nuanced and unique to each individual. It is critical to consider the impact of various life events on our beliefs and be open to new perspectives and experiences.

Self-perceived adult status

Adulthood is a period in life when people develop a sense of self-identity, self-sufficiency, and personal responsibility (Arnett, 1998; lowe et al., 2013). Perceiving oneself as an adult can significantly impact religious beliefs. Adults are more likely to question their previous beliefs, develop new spiritual perspectives, and engage in various religious experiences (Levenson et al., 2005). Through this process, they may gain a better understanding of their own beliefs, which can shape their religious identity and influence their future religious beliefs. Adults are more confident in their decision-making and less susceptible to external influences, which can significantly impact their religious beliefs (Arnett and Jensen, 2002). Individuals are more likely to make decisions as adults based on their own experiences and understanding of the world rather than blindly following their parents' or peers' beliefs and traditions (Barry and Nelson, 2005). Individuals may question the validity of their previous beliefs or embrace new ones that align more closely with their own values and experiences, which can lead to changes in religious beliefs. One way that seeing oneself as an adult can influence religious beliefs is by fostering greater independence from one's family or community (Bengtson, 2017).

Objectives

The purpose of the present research is to investigate the demographic factors (gender, age groups, religion, employment status, household income, marital status, self-perceived adulthood, and country of origin) that may influence the five variables of religious beliefs (religious certainty, religious culture, religious importance, religious practice, and belief in God).

Ethical statement

The research was conducted following in accordance with the ethical principles outlined in the Helsinki Declaration. The research was conducted with the approval of the Scientific and Ethical Review Committee at INTI International University (ref no: INTI/UEC/2021/002). Written informed consent was obtained from all participants before taking part in the study.

Study design and participants

This study used a cross-sectional design and survey methods to investigate the relationship between demographic factors and religiosity variables such as religious certainty, religious culture, religious importance, religious practice, and belief in God.

Respondents for the study came from Malaysia, the Philippines, China, Kenya, and Indonesia. The inclusive criteria for the study include being at least 18 years old and residing in one of the five countries, while the exclusive criteria are respondents under 18, those who do not live in the mentioned countries, those who do not provide informed consent, and those who do not complete the survey. Convenience sampling was used to collect data for the study, as respondents were selected based on their availability and willingness to participate. Co-authors who collected the data were trained to explain the study's purpose and confidentiality, and to respond to questions from respondents. To ensure objectivity and anonymity, the researchers contacted the co-authors to discuss questionnaire administration before data collection. Co-authors were also required to let participants know they could drop out of the study if they did not feel comfortable answering any questions or providing personal information. The sample size was determined based on the guidelines proposed by VanVoorhis and Morgan (2007), which recommend a sample size of at least 30 for each group being compared. The questionnaire was distributed in the local language of each country and was translated as needed. Completed questionnaires were collected via an online survey platform, and missing variables were managed through mean replacement (Hair et al., 2021). Anonymisation of participants was ensured by assigning unique identification numbers to each respondent and using them to link their responses to the data set.

Instruments

We adapted the religious belief questionnaires from Barry and Nelson (2005). The only adjustments implemented for the questionnaires were translations into Chinese and Indonesian, catering to participants originating from China and Indonesia, respectively. The English proficiency levels of citizens from Malaysia, Kenya, and the Philippines are relatively high, therefore participants from these countries were given the English version of the questionnaire. The questionnaire was initially translated from English to Chinese and then back-translated to English by three university lecturers proficient in English from Malaysia and China, ensuring consistency in content and terminology. The same process was applied to the Indonesian version of the questionnaire. The translation method employed aligns with the approach recommended by the World Health Organization for this purpose (WHO, 2020). Participants rated four items for religious culture on a scale of ‘not at all important’ (1) to ‘very important’ (4): ‘Suppose someone wanted to know all about you, how important would it be for them to know about your religious beliefs?’, ‘How important is it to you to have friends with the same religious background as you?’, ‘How important is it to you to date people with the same religious beliefs as you?’, and ‘How important is it to you to marry someone who shares your religious beliefs?’. These items were highly correlated (r’s ranging from.60 to.62, p < .01) and were combined to form an internally consistent composite (α = .87).

For religious practices, participants were asked to rate one item (‘How important is it to you to attend religious services regularly?’) on a scale of ‘not at all important’ (1) to ‘very important’ (4), and then another item (‘How often do you attend religious services?’) on a scale of from ‘about once or twice a year or less’ (1) to ‘at least three or four times a month’ (4). The items were significantly correlated (r = .56, p < 0.01), so they were combined to form an internally consistent composite (α = .71). Participants rated religious certainty on a scale of ‘very uncertain’ (1) to ‘very certain’ (4). Participants rated the following two items for religious importance on a scale of ‘not at all important’ (1) to ‘very important’ (4):’How important are your religious beliefs to you?’ and ‘How important is religious faith in your daily life?’. The religious belief items were significantly correlated (r = .83, p < 0.01) and then averaged to form an internally consistent composite (α = .91).

Participants rated a single religious belief item on a scale of ‘definitely do not believe this’ (1) to ‘definitely believe this’ (4): ‘To what extent do you believe that God or some higher power watches over you and guides your life?’

Data analysis

The data’s normality was first examined by referring to skewness and kurtosis. Kim (2013) proposed that skewness 2 and kurtosis 7 be absolute values. The collected data were analyzed using the IBM SPSS Statistic version 25.0 program (https://www.ibm.com/spss). GNU PSPP (https://www.gnu.org/software/pspp/), a freely accessible software, is also able to run the same analysis undertaken in this study. The statistical significance level was set at p < 0.05. Means, standard deviations, and ranges of religiosity variables were used to describe descriptive data. The independent t-test and one-way analysis of variance (ANOVA) were used to determine group differences in religious variables. In ANOVA, we test whether there is a significant difference between the means of three or more groups. If the F-test is significant, indicating that there is a significant difference between the group means, we use post-hoc tests like Tukey's Honestly Significant Difference (HSD) to determine which specific pairs of means are significantly different. Tukey's HSD test is used in ANOVA to provide additional information on where the significant differences between group means lie, allowing for more specific interpretations of the results. In addition, we defined household income as referring to the total incomes received (accumulated) by household members, whether in cash or in kind, that occur repeatedly within the reference period (within a year or more frequently). As household income is a specific measure and can vary widely within a country, and even within a particular region of a country, each co-author was advised to follow the latest government statistics guidelines from their own respective countries. For example, the most recent data on Malaysian household income is from the Department of Statistics' Household Income and Basic Amenities (HIS/BA) survey of 2019. This survey classifies Malaysian households into three income groups: Bottom 40% (B40; lower income), Middle 40% (M40; medium income), and Top 20% (T20; upper income). The B40 group consists of households with an income below RM4,850, while the M40 group includes households with an income ranging from RM4,851 to RM10,970. The T20 group represents households with an income exceeding RM10,970 (DOSM, 2020).

Participant characteristics

A total of 885 participants took part in this study. Of the 885 participants, 52.2% were female, and 47.8% were male; most participants were between 18 and 29 (58.5%). The majority of participants were Buddhist (28.1%), followed by Christians (20.9%), Roman Catholics (17.4%), Muslims (16.5%), Hindus (5.0%), atheists (1.5%), and those who preferred not to answer (10.6%). Regarding employment, 45.6% were full-time students, 44.9% were full-time employees, and 9.5% were others. Regarding household income, most respondents, (61.1%) had a medium income, followed by a lower income (29.6%), and 9.3% had an upper income. In terms of marital status, the majority were single (50.2%), followed by married (27.5%), in-relationship (15.6%), cohabiting (5.1%), and divorced (1.7%). 68.7% of the respondents perceived themselves to be adults. In terms of country of origin, the majority of respondents were from Malaysia (43.1%), followed by the Philippines (20.2%), China (13.9%), Kenya (12.9%), and finally, Indonesia (9.9%). Table 1 summarises the study’s participants’ characteristics. The complete dataset can be found in the underlying data (Wider, 2023).

The level of religiosity variables

The descriptive statistics and normality results for the religiosity variables are shown in Table 2. The skewness and kurtosis values were less than two, indicating that the normality assumption was met for all religious variables. Religious practice had the lowest mean score among the five religiosity variables, followed by religious culture and religious certainty. At the same time, respondents reported high religious importance and belief in God. The mean item scores on our questionnaire ranged from 2.73 to 3.17, with corresponding standard deviations ranging from 0.88 to 0.99. The mean scores indicated that respondents’ religious beliefs were moderate.

The differences of religiosity variables based on gender

The t-test analysis in Table 3 revealed that female respondents rated higher than male respondents in all religious belief variables, including religious certainty, religious culture, religious importance, religious practice, and belief in God.

The differences of religiosity variables based on age groups

The one-way ANOVA results in Table 4 revealed that only religious culture [F(2,882) = 6.98, p < 0.05] and religious practice [F(2,882) = 12.895, p < 0.01] differed across age groups. However, there were no age differences in religious certainty, religious importance, or belief in God. A post-hoc test using Tukey’s HSD revealed that respondents aged 30-45 perceived a higher level of religious culture and practice than those aged 18-29.

The differences of religiosity variables based on religions

Table 5 shows that religious certainty [F(6,878) = 55.96, p < 0.01], religious culture [F(6,878) = 43.21, p< 0.01], religious importance [F(6,878) = 90.35, p < 0.01], religious practise [F(6,878) = 90.67, p < 0.01], and belief in God [F(6,878) = 98.83, p < 0.01] differed across religions.

A post-hoc test using Tukey’s HSD revealed that Islam groups rated themselves significantly higher than Buddha groups, and Atheist groups and preferred not to answer groups in all religiosity variables, including religious certainty, religious culture, religious importance, religious practice, and belief in God. In terms of religious culture and practice, Christian groups rated themselves higher than all other religions except Islam. On all religiosity variables, Roman Catholic groups rated themselves higher than Buddha groups. Hindu groups rated themselves higher than Atheist groups and preferred not to answer groups on all religious variables. Regarding religious importance, religious practice, and belief in God, Buddhist groups ranked higher than Atheist groups.

The differences of religiosity variables based on employment status

The one-way ANOVA results in Table 6 revealed that only religious practice [F(2,882) = 5.98, p < 0.05] and belief in God [F(2,882) = 4.60, p < 0.05] differed by employment status. However, there were no differences in religious certainty, religious culture, or religious importance based on employment status.

A post-hoc test using Tukey’s HSD revealed that full-time students perceived themselves to have less religious practice than full-time employees and others. Meanwhile, other groups rated their belief in God as higher than the full-time employed groups.

The differences of religiosity variables based on household income

Table 7 shows that religious certainty [F(2,882) = 3.78, p < 0.05], religious culture [F(2,882) = 7.65, p < 0.05], religious importance [F(2,882) = 9.07, p < 0.01], religious practise [F(2,882) = 7.21, p < 0.05], and belief in God [F(2,882) = 6.61, p < 0.05] differed by household income.

A post-hoc test using Tukey’s HSD revealed that lower-income groups rated themselves higher than middle- and upper-income groups on all religiosity variables, including religious certainty, religious culture, religious importance, religious practice, and belief in God. Meanwhile, upper-income groups rated themselves lower on all religiosity variables than middle-income groups.

The differences of religiosity variables based on marital status

The one-way ANOVA results in Table 8 showed that religious culture [F(4,880) = 2.52, p < 0.05], religious practice [F(4,880) = 2.92, p < 0.05], and belief in God [F(4,880) = 3.14, p < 0.05] differed by marital status.

A post-hoc test using Tukey’s HSD revealed that married groups rated themselves higher on religious culture and practice than single and in-relationship groups. Meanwhile, single people rated their belief in God higher than cohabiting people and married people.

The differences of religiosity variables based on self-perceived adulthood status

The t-test analysis in Table 9 revealed that respondents who perceived themselves adults rated higher religious culture and religious practice than respondents who perceived themselves not yet adults. In the meantime, there were no significant differences in religious certainty, religious importance, or belief in God based on perceived adulthood.

The differences of religiosity variables based on country of origin

According to the results of the one-way ANOVA in Table 10, religious certainty [F(4,880) = 78.88, p < 0.01], religious culture [F(4,880) = 48.78, p < 0.01], religious importance [F(4,880) = 120.24, p 0.01], religious practice [F(4,880) = 80.39, p < 0.01], and belief in God [F(4,880) = 139.94, p < 0.01].

According to a post-hoc test using Tukey’s HSD, respondents from Indonesia and Kenya perceived themselves to have higher religious certainty, religious culture, religious importance, religious practice, and belief in God than respondents from the Philippines, Malaysia, and China. However, for all religiosity variables, there was no significant difference between respondents from Indonesia and Kenya. Furthermore, Chinese respondents scored the lowest in all religiosity variables.

Implications

This study proves that religiosity is a complex phenomenon influenced by various demographic factors. The study’s findings show that gender, age, religion, employment status, household income, marital status, and self-perceived adult status all contribute to one’s level of religiosity. The study’s results imply that religious beliefs are influenced by more than just an individual’s set of convictions and the quality of their spiritual experiences. The findings of this study have important practical implications for religious studies and religious communities. The study’s findings, for example, suggest that religious communities should consider the demographic factors that influence religiosity when developing programmes and initiatives to engage people in religious activities. Religious communities may need to tailor their program to appeal to specific age or gender groups. Furthermore, religious communities may need to consider the impact of employment status and household income on religiosity in order to design programmes that are accessible and appealing to people from various socioeconomic backgrounds. Important implications for policymakers and governments can also be drawn from the study’s finding. For example, governments may need to consider different demographic groups’ religious attitudes and practices when policies affect religious communities. Governments may need to consider the impact of education and employment policies on religious practices and ensure that policies do not discriminate against specific religious groups.