Religion and Loneliness: Investigating Different Aspects of Religion and Dimensions of Loneliness

Abstract

This study explores the relationship between religiosity, spirituality, and loneliness in the United States, using the 2018 General Social Survey to assess their interactions against a backdrop of declining traditional religious affiliation and a rise in “spiritual but not religious” identification. It examines religion and spirituality’s capacity to counteract loneliness, a condition with significant health implications. The analysis uncovers complex relationships between aspects of religious life and loneliness, showing no mitigating role of spirituality when controlling for other factors, with complex and varied negative relationships of religious service attendance and self-rate religiosity to different aspects of loneliness. Yet, any potentially protective effect of religion varies, with minority religious groups reporting feelings of increased loneliness. These findings underscore religion and spirituality’s nuanced roles in emotional well-being, indicating that their benefits (or not) against loneliness are complex, varied, and depend on the aspect of religion or loneliness observed, along with non-religious factors. The paper contributes to the literature on societal loneliness, changing religious and spiritual engagements, and highlighting the need for a comprehensive approach to explore the role of religion and spirituality in understanding loneliness specifically, and personal and social well-being more generally.

1. Introduction

The nexus of religion, spirituality, and loneliness constitutes an important yet incomprehensibly understood and dynamic domain of research, reflecting broader inquiries into how social and spiritual connections influence social and psychological well-being. Religion is considered an important constituent element for life satisfaction, prominently because of the social relationships built within religious congregations and facilitated by religious service attendance within those congregations (Lim and Putnam 2010). As societies grapple with increasing instances of social isolation and loneliness—a condition linked to myriad health risks including depression, anxiety, and cardiovascular disease (Leigh-Hunt et al. 2017; NASEM 2020)—the potential mitigating role of religion and spirituality (or lack thereof) requires attention. Understanding and mitigating loneliness involves exploring the human need for belonging and connection (Ding et al. 2022). Research suggests that feeling part of a community, whether through place or shared religious beliefs, can counteract loneliness (Pospíšil and Macháčková 2021; Antonsich 2010; Berghuijs 2017). This paper seeks to explore the intricate dynamics between various expressions of religiosity and spirituality and their impact on different dimensions of loneliness, employing the robust dataset of the General Social Survey (GSS) to investigate the complexities of these relationships in the context of an evolving religious landscape in the contemporary United States. This includes contrasting these relationships relative to the growing societal group that has no religion, either in identity or practice.

While there is still strong inter-generational transmission of different aspects of religious belief and practice in the United States (Gemar 2023a), recent shifts in the religious and spiritual fabric of society, including a notable decline in traditional religious affiliation and attendance (Hout and Fischer 2002, 2014; Chaves 2011) juxtaposed with an increase in individuals identifying as ‘spiritual but not religious’ (Fuller 2001; Ammerman 2013a, 2013b), underscore the necessity of examining the consequences of these changes for social connectedness. This evolving landscape prompts a reevaluation of the roles that religious congregations and spiritual practices play in fostering community and alleviating loneliness, a condition characterized not just by a lack of social contacts but by the subjective feeling of being disconnected or isolated (Cacioppo and Patrick 2008). Despite the integral role of religious beliefs and practices in shaping cultural contexts, studies directly linking religiosity to loneliness are scarce (Rokach et al. 2012).

Drawing upon the previous work of researchers who highlighted the community and social benefits derived from religious service attendance (e.g., Lim and Putnam 2010; Rote et al. 2013), this study addresses the need for deeper analysis into how different dimensions of religiosity and spirituality correlate with different dimensions of loneliness. This includes responding to calls from scholars like Yang and Nino (2023) for a more nuanced understanding of the relationship between different religious identities and social isolation, along with non-traditional forms of belief amidst the backdrop of decreasing formal religious engagement and the rise of individualized spiritual practices. While there has been increasing focus on loneliness, previous research on loneliness has only sparingly considered religion, focusing instead on elements of age and generation (Rote et al. 2013), race (Yang and Nino 2023), politics (Yang and Nino 2023), and social networks (Hastings 2016), often themselves calling on further investigation into any findings regarding connections of religion and loneliness. In this paper, I seek to provide an updated and more comprehensive assessment of interactions of different aspects of religious belief and engagement with different dimensions of loneliness that might be disparately connected to different elements of religion.

This study leverages data from the 2018 General Social Survey, which notably includes specific variables for assessing different aspects of loneliness, to conduct a thorough analysis of how religious ties, attendance at services, and levels of personal spirituality and religiosity correlate with experiences of feeling excluded, isolated, or lacking companionship. By delving into these relationships, the research contributes to and expands upon existing academic discussions surrounding religion and loneliness. It also uncovers the intricate impact that religious identity and alternative spiritual practices have on different facets of loneliness. Consequently, the outcomes of this study enhance our comprehension of the broader societal effects stemming from the shifting landscape of religious and spiritual practices. In providing an extensive investigation into the complex dynamics among religion, spirituality, and loneliness, the research sheds light on how engagement in religious and spiritual practices may serve as a countermeasure to loneliness in contemporary society, underscoring the need for a holistic examination of the role of spirituality in fostering social well-being.

2. Religion and Loneliness

Social isolation, characterized by both the objective physical separation from others and a subjective sense of lacking social relations and companionship, encompasses various forms, including being socially disinterested, socially avoidant, and actively socially isolated, and is linked to risky behaviors among adolescents as well as adverse mental and physical health outcomes such as depression, anxiety, adjustment disorder, chronic stress or illness, insomnia, suicide, and dementia (e.g., Banerjee and Rai 2020; Leigh-Hunt et al. 2017; NASEM 2020; Niño et al. 2016; Yang and Nino 2023). Research has highlighted the importance of social connections in various contexts, emphasizing the benefits of interactions with friends, family, co-workers, and neighbors (Hastings 2016). These connections serve as channels for social support, norm socialization, and access to information and resources, which are particularly crucial during challenging times (Durkheim [1897] 1951; Bellah et al. 1985; Wellman and Wortley 1990; Heckathorn 1988; Podolny and Baron 1997; Granovetter 1973; Hurlbert et al. 2000). Consequently, a strong link has been identified between an individual’s network of social connections and their physical, mental, and emotional well-being (Hastings 2016), elements closely associated with feelings of loneliness. While religious congregations have historically and continue to be centers of social life for millions of Americans, empirical observations consistently indicate a rise in individuals identifying as religiously unaffiliated (Hout and Fischer 2002, 2014; Putnam et al. 2010) alongside a downturn in attendance at religious services (Chaves 2011), potentially signaling a shift in the potency of religious congregations as social meeting places. Findings of the decline in social connectivity, especially through religious communities, raise concerns about its broader implications on societal well-being (Putnam 2000; Fischer 2011). Lim and Putnam (2010), as well as Lewis et al. (2013), have found that active engagement in religious communities, rather than mere personal faith, significantly contributes to life satisfaction and civic engagement, highlighting the community aspect of religiosity over individual belief systems. While religious congregations have historically and continue to be centers of social life for millions of Americans, empirical observations consistently indicate a consistent rise in individuals identifying as religiously unaffiliated (Gullickson 2018; Hout and Fischer 2002, 2014; Putnam et al. 2010; Smith 2021) alongside a downturn in religious observance and attendance at religious services (Brauer 2018; Chaves 2011; Inglehart 2020; Voas and Chaves 2016), potentially signaling a shift in the potency of religious congregations as social meeting places.

Findings of the decline in social connectivity, especially through religious communities, raise concerns about its broader implications on societal well-being (Putnam 2000; Fischer 2011), and for our purpose, feelings of loneliness. Lim and Putnam (2010), along with Lewis et al. (2013), have found that active engagement in religious communities, rather than mere personal faith, significantly contributes to life satisfaction and civic engagement, highlighting the community aspect of religiosity over individual belief systems. These dynamics and the social connection gleaned through religious engagement and connection reflect elements of social capital and have often been studied through the lens of volunteer, civic, and social engagement, with the degree of embeddedness within a social network such as a congregation significantly influencing this engagement for reasons also of interpersonal and institutional trust (Gemar 2024a). While many have found higher levels of religiosity being often linked to higher levels of this engagement (Lewis et al. 2013), others suggest that the accumulation of social capital, as found in religious communities, can sometimes lead to reduced levels of civic engagement (Putnam 1993; Uslaner 2002; Warren 2001). This is especially the case in some denominations that have deep-seated apprehensions about diversity (Porter 2007; Porter and Emerson 2012), while individuals claiming multiple religious identities have been found to have elevated levels of diversity in social connection and cultural engagement (Gemar 2023b, 2024b). Additionally, amounts of religious capital, such as frequent attendance and ritual observance with friends and family (Baker and Miles-Watson 2010), may facilitate belonging and embeddedness within a closed religious community or across communities, if such knowledge and ritual observance extends to other groups in a similar way that multiple religious identities might facilitate. In line with the concept of religious capital, Acevedo et al. (2014) find that religious participation can serve as a resource to buffer the effects of psychological stress, while integration in religious congregations is negatively associated with teen drug use (Bartkowski and Xu 2007).

The relatively divergent findings on the relationship between religious connection and social engagement may be attributed to the varied motivations behind individuals’ engagement in social networks and voluntary associations (Porter and Emerson 2012), which range from the desire for experiential diversity to the pursuit of affiliations with individuals sharing similar values or beliefs (Newton 1999; Paxton 2002), and thus more of a type of network closure. Regarding religious affiliation, religious identification can often reflect ethnic rather than strictly religious ties (Voas and Bruce 2004), a dynamic which may lead to either network closure or a social capital associated with multiple religious identifications from diverse ethnic backgrounds (Gemar 2024b). For instance, Davenport (2016) notes that religious affiliation is a significant factor in how biracial Americans define their racial identity, while Philips and Fishman (2006) observe that among American Jews, a higher level of social capital is associated with a greater likelihood of intermarriage. Understanding these dynamics is essential for examining the broader implications of religious and spiritual involvements on loneliness and social integration, as the nature of social ties established through religious communities or spiritual practices is pivotal in shaping individual experiences of social connectedness and isolation.

While religious institutions may be a prime site of the development of in-group ties (Blanchard 2007; Porter 2007; Porter and Emerson 2012; Putnam 2000), a growing number of Americans identify as spiritual without adhering to a major religious tradition, a phenomenon labeled with various terms such as ‘believing without belonging’ (Davie 1994 and ‘spiritual but not religious’ (Fuller 2001), among others non-traditional forms of religious engagement and belief). Studies have shown that ‘religiosity’ is more connected to institutional trust than ‘spirituality’ (Gemar 2024a), and thus this separation may be less conducive for forming social ties based upon trust, especially in the more public realm. But the line between spirituality and religion has always been fluid, with current understanding emphasizing spirituality as a personal quest for life’s ultimate meaning and an individual experience of the sacred, contrasting with religion’s more public and stable nature (Bender 2010; Oh and Sarkisian 2012). However, these concepts are not mutually exclusive, as many individuals integrate spiritual beliefs and practices within the context of traditional religious participation (Ammerman 2013a, 2013b; Albanese 2001; Mahoney and Cano 2014). The terminology used to describe oneself as ‘spiritual’ or ‘religious’ often varies by context and social appropriateness, reflecting the fluidity between these identities (Bender 2010; Ammerman 2013a, 2013b). Despite this, there is an increasing trend of people identifying as spiritual without being conventionally religious, which has also been identified in the United States (McDowell 2018). While it may seem intuitive that the rise of the spiritual but not religious (e.g., Parsons 2018; Seto 2021) entails solitary faith practices in a type of ‘privatization’ of religious belief and practice (McDowell 2018), research highlights the existence of non-religious collectives engaging in communal spiritual practices (Bender 2010; Oh and Sarkisian 2012). However, given these dynamics, the influence of religion and spirituality on socialization and connection outside traditional congregations generally remains under-explored (Hastings 2016). This is particularly true for feelings of loneliness or lack of connection.

For instance, the COVID-19 pandemic has empirically underscored the protective role of spiritual well-being against loneliness, revealing that spiritual and religious practices can provide strength and reduce feelings of isolation (Ausín et al. 2021; González-Sanguino et al. 2020, 2021; Hwang et al. 2020). Studies show that spirituality, particularly during crises like the pandemic, supports coping mechanisms, fosters community connections via virtual religious events, and enhances personal faith to combat loneliness (Roy et al. 2020; Heidari et al. 2020; Hamilton et al. 2022). Therefore, higher levels of spirituality may be associated with lower levels of loneliness, in a deployment of a type of ‘spiritual capital’ which can be understood as a reservoir from which individuals draw to navigate life’s hurdles, such as illness, injustice, moral dilemmas, societal issues, or psychological distress (Acevedo et al. 2014; Baker and Miles-Watson 2010). However, while there are studies that find a mediating effect of spirituality on loneliness in some populations, such as Latinos (Gallegos and Segrin 2019), this beneficial effect is not found amongst other populations, such as Jewish seniors (Springer et al. 2003), or the negative association between spirituality and loneliness is weak (Migdal and MacDonald 2013).

Regarding forms of religious capital, Hastings (2016) finds that regular attenders of religious services exhibit greater social connectedness than both Religious Non-Attenders and those identified as Neither Spiritual Nor Religious, with the latter two groups showing comparable levels of connectedness. Hastings (2016) also finds that spiritual but not religious individuals have similar core discussion networks and numbers of close ties to regular attenders, and in some measures, appear more socially connected than non-spiritual/non-religious and religious non-attendees, though these differences are not always statistically significant. This research implies that heterogeneity exists within the population disengaging from traditional religious affiliations, underscoring the need for a nuanced understanding of their social connectivity (Hastings 2016). Research also indicates that religious attendance acts as a unifying force across the lifespan, particularly benefiting older adults as they transition away from roles linked to work and family (Idler et al. 2003; Krause 2008). Idler et al. (2009) highlight the multifaceted nature of worship services, encompassing rituals, prayers, music, and social support opportunities, making attendance a rich, multidimensional experience. Regular participation in these services has been linked to reduced social isolation and loneliness (Gray 2009; Johnson and Mullins 1989; Kobayashi et al. 2009; Schwab and Petersen 1990; Rote et al. 2013).

The significant impact of religious involvement on increasing overall happiness and reducing feelings of loneliness has been thoroughly recognized and supported by research. Engaging in regular religious services cultivates a community feeling and a sense of belonging among participants by sharing in sacred practices and rituals, which in turn fosters friendships and other connections (Bradley 1995; Ellison and George 1994). This communal bond is further enhanced by the efforts of religious leaders and organizations to create a spirit of community through arranging social gatherings and promoting volunteerism and charity work that solidifies social connections (Rote et al. 2013; Wilson and Janoski 1995). Additionally, being part of a religious community serves as a springboard for wider social and civic participation, broadening one’s social circle and enhancing interactions within families and between spouses. Such dynamics contribute to a supportive community framework, essential for achieving a satisfying life (Ellison et al. 2010; Bradley 1995; Ellison and George 1994; Idler and Kasl 1997; Krause 2006; McIntosh et al. 2002).

Expanding upon these insights, Rote et al. (2013) explore how regular engagement with religious communities, social interconnectedness, and feelings of loneliness interrelate, proposing a model where active religious participation is seen to bolster social connections and cohesion, thereby reducing loneliness. This model suggests that consistent involvement in religious ceremonies is linked with heightened social support and community integration, which, in effect, is negatively correlated with loneliness (Rote et al. 2013). These observations align with prior studies, highlighting the importance of genuine involvement in religious activities in promoting social health across various age groups and demographics. It emphasizes the importance of distinguishing simple religious identity from meaningful participation in religious community life, arguing that such active involvement brings substantial social and communal advantages that surpass mere personal spiritual fulfillment (Verba et al. 1995; Putnam 2000; Mahoney 2010, 2013; Hastings 2016; Rote et al. 2013).

Further elucidating the multifaceted relationship between religiosity and loneliness, research by Burris et al. (1994) and others (Rosmarin et al. 2009; Ressler 1997) highlights the protective role of a deep, personal connection to religion. Burris et al. (1994) specifically point out that intrinsic religiosity, characterized by a personal and committed faith, can significantly mitigate feelings of loneliness when individuals are confronted with their vulnerability to isolation. This effect, they argue, is distinct from extrinsic religiosity or a quest for simple religious answers, suggesting that the depth of religious commitment plays a crucial role in buffering against loneliness. Supporting this notion, studies within Jewish communities have identified a generally inverse relationship between religiosity and loneliness, indicating that not only does religiosity reduce negative emotions and enhance a sense of belonging but that this effect varies across different religious practices and beliefs (Ellison 1983; Kirkpatrick et al. 1999; Paloutzian and Ellison 1982; Johnson and Mullins 1989). Rokach et al. (2012) extend this discussion to the coping strategies of Israeli Jewish groups, revealing how diverse religious observances influence loneliness and suggesting the profound impact of religious orientation on individuals’ experiences of isolation and their coping mechanisms.

Lastly, Yang and Nino’s (2023) research introduces a compelling counterpoint to common assumptions about religious affiliation and social isolation, finding that individuals affiliated with religions outside of Protestantism or Catholicism experience greater social isolation than those with no religious beliefs at all. This finding challenges the notion that absence from religious congregations inherently leads to greater social isolation among non-believers, signaling a crucial area for further research. It underscores the necessity of more systematic investigations into how various religious affiliations impact feelings of loneliness and social isolation, suggesting that the relationship between one’s religious identity and their experience of loneliness is more complex than previously understood. Similarly, Bartkowski et al. (2017) shed light on the nuanced effects of religious practices on psychological well-being, noting that while individual prayer may decrease anxiety, communal prayer can, paradoxically, increase it, hinting at the complexity of how religious practices affect mental health. Importantly, they emphasize that regular attendance at worship services significantly reduces anxiety, reinforcing the idea that active engagement in religious communities can serve as a powerful antidote to anxiety and, by extension, loneliness (Bartkowski et al. 2017).

Research Questions and Hypotheses

Given the conceptualizations and empirical results discussed earlier, I ask the following research questions and form the following hypotheses:

• R1: What is the nature and strength of the relationship between religion (or non-religion), spirituality, and loneliness?
• H1: Elements of both religion and spirituality will be negatively associated with loneliness compared to no religion or spirituality, even controlling for non-religious factors of social connectivity.
• R2: Is this relationship consistent across different aspects of religion and different dimensions of loneliness, or do they differ?
• H2: Spirituality and religious identity will show a weaker negative relationship to loneliness than religious service attendance, and the loneliness variables of feeling ‘left out’ or ‘isolated’ will be more strongly and differentially impacted than feeling a ‘lack of companionship’, which may be filled by non-religious factors, such as family.

Given the foundational role of social connections within religious settings in enhancing life satisfaction and mitigating loneliness, as evidenced by Lim and Putnam (2010) and further supported by Hastings (2016), it is hypothesized that active participation in religious congregations is inversely related to feelings of loneliness. This is predicated on the understanding that religious involvement, particularly through regular attendance at worship services and engagement in community activities, fosters a network of social support, norm socialization, and access to resources which are essential in reducing the subjective and objective aspects of social isolation.

Considering the findings of Yang and Nino (2023) and the differentiation between intrinsic and extrinsic religiosity as discussed by Burris et al. (1994), it is hypothesized that the relationship between religion and loneliness varies significantly across different aspects of religion. Specifically, intrinsic feelings of religiosity, characterized by a deep, personal commitment to faith, are likely to have a stronger inverse relationship with loneliness compared to extrinsic religiosity or non-traditional spiritual affiliations. This variance stems from the different levels of social integration and community support experienced by individuals within these religious or spiritual categories.

Additionally, given the nuanced understanding that religious and spiritual practices, especially during crises like the pandemic, have protective roles against loneliness (Ausín et al. 2021; González-Sanguino et al. 2020, 2021; Hwang et al. 2020), it is proposed that this relationship also varies across different dimensions of loneliness. For instance, feelings of being ‘left out’ or ‘isolated’ may be more significantly impacted by the frequency of religious service attendance and the nature of one’s spiritual beliefs and practices than the feeling of ‘lack of companionship’. This hypothesis recognizes the multifaceted nature of loneliness and suggests that the protective effects of religious and spiritual engagement are not uniform across all its dimensions.

3. Materials and Methods

This study relies on data from the General Social Survey (GSS) conducted by the National Opinion Research Center (NORC) at the University of Chicago, a nationally representative survey that captures the attitudes, behaviors, and demographics of the US population. Initiated in 1972, the GSS has been administered 34 times up to 2022, adapting its questions over the years to reflect the evolving societal concerns and academic interests. The 2018 GSS iteration is the only one that includes the loneliness variables that I use in this paper to capture different dimensions of loneliness in relation to religion. In this survey wave, the variables related to loneliness, which are central to our analysis, were answered by just over half of the respondents, resulting in a subsample of 1,185 responses for this study. The specific question used to construct these three distinct loneliness variables was, ‘How often in the past 4 weeks have you felt that…?’ with options including ‘You are left out’, ‘You are isolated from others’, and ‘You lack companionship’. Response options included ‘never’, ‘rarely’, ‘sometimes’, ‘often’, and ‘very often’. For our analysis, ‘often’ and ‘very often’ have been coded together.

Table A1. Relative frequencies of dependent and independent variables in the sample used for the analysis of this paper.

	Relative Frequency
Dependent Variables
Feeling ‘left out’
Often or very often	5.7%
Sometimes	12.2%
Rarely	26.0%
Never	56.0%
Feeling ‘isolated’
Often or very often	8.6%
Sometimes	14.4%
Rarely	19.8%
Never	57.3%
Feeling ‘lack of companionship’
Often or very often	10.3%
Sometimes	17.3%
Rarely	24.4%
Never	48.0%
Independent Variables
Religious affiliation
Protestant (unidentified)	15.9%
Catholic (Hispanic)	8.2%
Catholic (non-Hispanic)	15.3%
Jewish	1.6%
Other	3.7%
Black Protestant	6.5%
Mainline Protestant	10.2%
Evangelical Protestant	15.0%
No religion	23.6%
Religious attendance
Every week or more	22.9%
Almost every week	12.4%
Once a month	6.4%
Several times a year	10.0%
Once a year or less	18.5%
Never	29.8%
Religiosity
Very religious	15.5%
Moderately religious	37.7%
Slightly religious	25.2%
Not religious	21.6%
Spirituality
Very spiritual	29.2%
Moderately spiritual	35.4%
Slightly spiritual	23.7%
Not spiritual	11.7%
Sport or cultural org.
Multiple times/year	53.0%
Once in past year	11.3%
Never	35.8%
Political org.
Multiple times/year	10.1%
Once in past year	12.0%
Never	35.8%
Charity/religious org.
Multiple times/year	40.2%
Once in past year	19.7%
Never	40.1%
Occupational network diversity
0–2 different occupations	9.2%
3–5	24.9%
6–8	45.4%
9–10	19.9%
Employment status
Full-time employment	49.7%
Not employed full-time	50.3%
Years of education
0–12	12.9%
12	27.4%
13–14	21.0%
15–16	24.4%
More than 16	14.3%
Household income
$130,000+	18.0%
$90,000–129,999	14.5%
$60,000–89,999	18.5%
$40,000–59,999	15.4%
$25,000–39,999	11.6%
Less than $25,000	22.0%
Sex
Male	45.5%
Female	54.5%
Race
Other	12.8%
Black	14.8%
White	72.4%
Hispanic	16.9%
Non-Hispanic	83.1%
Age
18–24	11.8%
25–34	19.0%
35–49	25.8%
50–64	24.9%
65+	18.3%
Number of children
0	29.2%
1	13.9%
2	26.9%
3+	30.0%
Marital status
Married	49.3%
Widowed	5.8%
Divorced	13.7%
Separated	2.5%
Never married	28.7%

displays the relative frequencies of the subcategories for these dependent variables.

This paper examines religious affiliation, engagement, and belief through four specific questions addressing religious preference, religious service attendance, and the extent of respondents’ self-rated religiosity and spirituality. By applying Steensland et al.’s (2000) classification to religious identification, I further distinguish among Protestant denominations (‘Black’, ‘Mainline’, and ‘Evangelical’) and separate ‘Hispanic’ from non-Hispanic Catholics, enabling a nuanced analysis of religious identification and possible relationships to loneliness. I therefore also respond to the call of Yang and Nino (2023) for a more comprehensive understanding of the relationship between identity and loneliness, and how it may differ across identities. Because responses capturing private religious practices, such as prayer and scripture reading, did not come from the same subsample as those answering the questions on loneliness, they were unfortunately unable to be used in our analysis, and thus our research questions may not be able to be as comprehensively answered as I would desire. Therefore, only certain elements of more public religious engagement, such as religious service attendance, along with self-reported aspects of religious identity, religiosity, and spirituality can be assessed. I argue that these are still useful measures for the purposes of this analysis to answer our research questions.

To explore how our independent religious variables relate to the three dependent loneliness variables, I conducted multinomial logistic regression analyses. These analyses were carried out separately for each loneliness variable to identify potential variations in the impact of religious factors on different aspects of loneliness. In addition to the four religious variables, I also incorporate variables of participation in sports or cultural organizations, participation in political organizations, participation in charitable or religious organizations, diversity of occupational networks, employment status, years of formal education, household income, sex, race, Hispanic ethnic status, age, number of children, and marital status as control variables in the regression models. I calculate occupational network diversity from the GSS survey question that asked, ‘Do you know a woman or man who is a(n) X?’. This is the only survey year that the GSS has included a question of this sort. The occupations that the GSS includes bus driver, cleaner, police officer, hair-person, executive, personnel manager, lawyer, car mechanic, nurse, and schoolteacher. Therefore, there are only ten occupations included, but I include this type of measure to capture social network diversity. Table A1 displays the relative frequencies of the subcategories for these independent variables, and the results of the regression analyses are presented in

Table 1. Results from regression analysis of frequency of feeling ‘left out’ ^a.

Feel ‘Left Out’	Rarely	Sometimes	Often/Very Often
Religious preference
Protestant (unidentified)	0.245	0.525	1.322 *
Catholic (Hispanic)	0.467	−0.217	0.676
Catholic (non-Hispanic)	−0.015	−0.171	0.149
Jewish	0.768	−0.765	2.627 **
Other	−0.416	1.428 **	2.724 ***
Black Protestant	0.014	0.067	1.186
Mainline Protestant	0.041	0.660	0.581
Evangelical Protestant	−0.036	0.342	0.666
No religion	0	0	0
Religious attendance
Every week or more	−0.558	−0.625	−0.811
Almost every week	−0.509	0.215	−0.487
Once a month	−0.755	−1.002	0.165
Several times a year	0.069	−0.131	0.196
Once a year or less	0.051	−0.647	−0.681
Never	0	0	0
Religiosity
Very religious	−0.890 *	−0.179	0.208
Moderately religious	−0.268	−0.374	−0.093
Slightly religious	−0.107	−0.022	0.462
Not religious	0	0	0
Spirituality
Very spiritual	0.453	−0.169	0.047
Moderately spiritual	0.027	0.132	0.291
Slightly spiritual	0.340	−0.112	0.107
Not spiritual	0	0	0
Sport or cultural org.
Multiple times/year	0.195	−0.312	−0.085
Once in past year	0.236	−0.350	−0.857
Never	0	0	0
Political org.
Multiple times/year	0.401	0.844 *	−1.283
Once in past year	0.306	0.824 *	−0.532
Never	0	0	0
Charity/religious org.
Multiple times/year	0.532 *	0.391	0.202
Once in past year	0.504 *	0.289	−0.147
Never	0	0	0
Occupational network diversity
0–2 different occupations	−0.481	−0.971	−0.385
3–5	0.013	0.357	−0.040
6–8	−0.010	−0.079	−0.358
9–10	0	0	0
Employment status
Full-time employment	−0.069	0.032	0.031
Not employed full-time	0	0	0
Years of education
0–12	0.067	0.670	1.281 *
12	−0.299	0.514	0.293
13–14	−0.105	0.658	0.611
15–16	−0.123	0.367	0.726
More than 16	0	0	0
Household income
$130,000+	−0.308	−0.341	−0.698
$90,000–129,999	0.217	−0.016	−0.791
$60,000–89,999	0.505	0.268	−0.999
$40,000–59,999	0.210	−0.275	−0.203
$25,000–39,999	−0.028	0.118	−0.577
Less than $25,000	0	0	0
Sex
Male	−0.188	0.035	−0.036
Female	0	0	0
Race
Other	−0.235	−0.250	−0.868
Black		−0.266	−0.095
0.176
White	0	0	0
Hispanic	−0.320	0.410	0.886
Non-Hispanic	0	0	0
Age
18–24	0.014	−0.307	−0.021
25–34	0.207	−0.222	0.511
35–49	0.328	−0.188	−0.197
50–64	0.125	−0.037	0.094
65+	0	0	0
Number of children
0	−0.063	0.429	−0.032
1	−0.151	0.773 *	−0.267
2	−0.085	0.388	−0.856
3+	0	0	0
Marital Status
Married	−0.460	−0.991 **	−1.093 *
Widowed	0.228	−0.614	−0.059
Divorced	−0.158	−0.511	−0.090
Separated	0.362	0.434	0.540
Never married	0	0	0
Nagelkerke R-squared	0.244

^a Reference group = never; * p <0.05; ** p < 0.01; *** p < 0.001.

,

Table 2. Log odds ratios from regression analysis of frequency of feeling ‘isolated’ ^a.

Feel ‘Isolated’	Rarely	Sometimes	Often/Very often
Religious preference
Protestant (unidentified)	0.242	0.449	1.128 *
Catholic (Hispanic)	0.359	0.260	0.114
Catholic (non-Hispanic)	−0.119	0.044	0.404
Jewish	0.901	0.283	2.017 *
Other	1.500 *	0.207	0.838
Black Protestant	−0.706	0.483	0.778
Mainline Protestant	0.042	0.147	0.620
Evangelical Protestant	−0.471	−0.332	0.247
No religion	0	0	0
Religious attendance
Every week or more	0.334	−0.277	−0.719
Almost every week	0.298	−0.306	−0.922
Once a month	0.234	−0.286	−0.910
Several times a year	0.477	−0.163	−0.434
Once a year or less	0.008	−0.397	−1.126 **
Never	0	0	0
Religiosity
Very religious	−0.602	−0.856	−0.310
Moderately religious	−0.362	−0.556	0.079
Slightly religious	−0.307	−0.173	0.493
Not religious	0	0	0
Spirituality
Very spiritual	0.333	0.406	0.814
Moderately spiritual	0.240	0.264	0.245
Slightly spiritual	0.407	0.387	0.645
Not spiritual	0	0	0
Sport or cultural org.
Multiple times/year	−0.115	−0.192	−0.594
Once in past year	−0.161	−0.185	−0.267
Never	0	0	0
Political org.
Multiple times/year	0.267	0.247	−0.386
Once in past year	0.244	0.354	−0.072
Never	0	0	0
Charity/religious org.
Multiple times/year	0.457	0.365	0.068
Once in past year	0.238	0.691 *	0.223
Never	0	0	0
Occupational network diversity
0–2 different occupations	−0.090	−0.130	−0.419
3–5	0.208	0.526	0.828
6–8	0.024	0.128	0.433
9–10	0	0	0
Employment status
Full-time employment	−0.201	−0.194	−0.508
Not employed full-time	0	0	0
Years of education
0–12	−0.274	−0.182	0.488
12	−0.053	−0.281	−0.496
13–14	0.012	0.365	−0.067
15–16	−0.227	0.048	−0.399
More than 16	0	0	0
Household income
$130,000+	−0.944 *	−0.774	−0.240
$90,000–129,999	−0.166	−0.139	−0.692
$60,000–89,999	0.034	−0.431	−0.288
$40,000–59,999	0.115	0.062	0.102
$25,000–39,999	0.349	−0.656	0.290
Less than $25,000	0	0	0
Sex
Male	−0.010	−0.031	0.022
Female	0	0	0
Race
Other	0.434	0.039	0.254
Black		−0.273	−0.340
0.363
White	0	0	0
Hispanic	−0.808	−0.117	0.496
Non-Hispanic	0	0	0
Age
18–24	0.419	0.426	0.763
25–34	−0.089	0.613	1.596 *
35–49	0.167	0.603	1.435 *
50–64	−0.312	0.397	1.212 *
65+	0	0	0
Number of children
0	0.554	−0.178	0.690
1	0.408	−0.021	−0.411
2	0.036	−0.009	−0.230
3+	0	0	0
Marital status
Married	−0.075	−0.690 *	−1.451 **
Widowed	−0.295	0.901	0.683
Divorced	0.537	0.464	0.542
Separated	1.065	0.441	1.311 *
Never married	0	0	0
Nagelkerke R-squared	0.262

^a Reference group = never; * p < 0.05; ** p < 0.01.

and

Table 3. Log odds ratios from regression analysis of frequency of feeling ‘lack of companionship’ ^a.

Feel ‘Lack Companionship’	Rarely	Sometimes	Often/Very often
Religious preference
Protestant (unidentified)	−0.180	0.136	1.213 *
Catholic (Hispanic)	1.127 *	0.416	1.880 **
Catholic (non-Hispanic)	−0.616	0.388	−0.043
Jewish	1.090	2.100 **	−0.265
Other	−0.346	−0.249	1.717 **
Black Protestant	−0.158	−0.219	1.850 *
Mainline Protestant	0.244	0.505	1.144 *
Evangelical Protestant	0.077	0.138	0.412
No religion	0	0	0
Religious attendance
Every week or more	0.035	0.970 *	0.402
Almost every week	0.362	0.545	0.103
Once a month	0.039	0.701	−0.345
Several times a year	−0.175	1.065 **	−0.453
Once a year or less	−0.318	0.356	0.088
Never	0	0	0
Religiosity
Very religious	−0.283	−0.871	−1.618 *
Moderately religious	0.410	−0.173	−0.503
Slightly religious	0.065	−0.148	−0.629
Not religious	0	0	0
Spirituality
Very spiritual	−0.641	−0.596	0.446
Moderately spiritual	−0.779 *	−0.259	−0.192
Slightly spiritual	−0.410	−0.203	−0.110
Not spiritual	0	0	0
Sport or cultural org.
Multiple times/year	0.250	0.074	0.202
Once in past year	−0.066	0.493	0.532
Never	0	0	0
Political org.
Multiple times/year	0.224	0.837 *	0.527
Once in past year	0.161	−0.524	−0.376
Never	0	0	0
Charity/religious org.
Multiple times/year	−0.198	−0.026	−0.254
Once in past year	−0.353	−0.549	0.100
Never	0	0	0
Occupational network diversity
0–2 different occupations	−0.488	0.380	−0.719
3–5	−0.211	0.732 *	0.188
6–8	−0.271	0.202	0.117
9–10	0	0	0
Employment status
Full-time employment	−0.048	0.058	0.160
Not employed full-time	0	0	0
Years of education
0–12	0.228	0.844	0.722
12	−0.498	0.271	−0.018
13–14	0.282	1.114 **	0.010
15–16	−0.009	0.504	0.026
More than 16	0	0	0
Household income
$130,000+	0.065	−0.530	−0.485
$90,000–129,999	0.274	−0.824 *	−0.564
$60,000–89,999	0.391	−0.400	−0.393
$40,000–59,999	0.203	0.040	−0.153
$25,000–39,999	−0.674	−0.162	−0.995 *
Less than $25,000	0	0	0
Sex
Male	−0.412 *	−0.159	0.248
Female	0	0	0
Race
Other	−0.149	−0.269	−0.386
Black	0.008	0.137	−1.145
White	0	0	0
Hispanic	−0.341	−0.763	−0.298
Non-Hispanic	0	0	0
Age
18–24	0.651	0.789	0.761
25–34	0.416	0.668	0.841
35–49	0.252	1.280 ***	1.472 **
50–64	0.612 *	0.893 *	1.414 **
65+	0	0	0
Number of children
0	−0.164	0.169	0.514
1	−0.309	0.259	0.420
2	−0.524 *	0.276	−0.470
3+	0	0	0
Marital status
Married	−1.039 ***	−1.363 ***	−2.760 ***
Widowed	−0.169	0.394	0.928
Divorced	−0.558	−0.931 *	0.121
Separated	−0.789	0.486	0.804
Never married	0	0	0
Nagelkerke R-squared	0.365

^a Reference group = never; * p < 0.05; ** p < 0.01; *** p < 0.001.

.

4. Findings

In examining the association between religious affiliation, attendance, religiosity, spirituality, and demographic variables with feelings of being left out, isolated, or lacking companionship, several statistically significant results emerged across three regression models, as displayed in each of Table 1, Table 2 and Table 3. For all three measures of loneliness, ‘never’ feeling in that manner is the reference category.

Firstly, when exploring the feeling of being ‘left out’, individuals identifying as Jewish and those with ‘Other’ religious preferences reported significantly higher odds of feeling left out ‘often/very often’ than those of no religion, relative to never feeling left out. There are notably no statistically significant relationships between religious service attendance and feelings of being left out, although non-statistically significant results suggest mostly negative associations for those attending services more often. While a ‘very religious’ orientation was significantly associated with a lower likelihood of feeling rarely left out compared to the non-religious, both religiosity and spirituality showed weak and non-statistically significant relationships to feeling left out. Engagement in political organizations and charitable or religious organizations showed positive associations with feeling left out ‘sometimes’ and ‘rarely’, respectively, relative to never feeling left out. Finally, a lower number of years of formal education was positively predictive of feeling left out often or very often, while being married compared to never married was negatively predictive of feeling left out sometimes or often/very often.

The analysis of feeling ‘isolated’ revealed that ‘unidentified’ Protestants and Jewish respondents experienced higher odds of feeling ‘often/very often’ isolated. Attendance at religious services ‘Once a year or less’ showed lower odds of feeling ‘often/very often’ isolated than feeling ‘never’ isolated. There were no statistically significant results for either religiosity or spirituality, although non-statistically significant results show positive relationships between higher levels of spirituality and feelings of isolation. The final two variables showing statistically significant relationships to feelings of isolation are age and marital status. Compared to those over the age of 65, those between 25–64 were more likely to feel isolated often/very often than to never feel isolated. Married respondents were again less likely than never-married respondents to feel isolated sometimes, or especially often/very often, while respondents who report being separated were more likely to feel isolated often/very often than never isolated, again relative to those who were never married.

For our third dependent variable and dimension of loneliness, in considering feelings of ‘lack of companionship’, results underscore the significant impact of religious affiliation, with unidentified Protestants, Hispanic Catholics, Black Protestants, Mainline Protestants, and those with ‘other’ religious identifications more likely to report ‘often/very often’ feeling a lack of companionship. Regular religious attendance was not statistically significant in the regression model, although elevated attendance showed non-statistically significant negative relationships to increased feelings of lacking companionship. Those who reported being ‘very religious’ were much less likely to feel that they lacked companionship often/very often. While there were some statistically significant categories at lower levels feelings of lacking companionship, age and marital again stand out as other strongly predictive variables. Compared to those over 65, those 35–64 were more likely to feel that they lacked companionship sometimes or often/very often, while married respondents again were much less likely to feel that they lacked companionship than those who were never married.

5. Discussion and Conclusions

During a time of significant transformations in the religious and spiritual landscape of the United States, this research delves into the nuanced interactions among religion, spirituality, and the experience of loneliness. The objective is to deepen the understanding of how different forms of religious expression and spiritual practice may influence feelings of loneliness, an issue that carries important implications for both societal and individual health and connection. The study reveals complex connections among facets of religious experience and feelings of social isolation, or loneliness. While certain elements of religious engagement seem to offer protection against feelings of exclusion, isolation, or absence of companionship, others, especially specific religious affiliations and spirituality, seem either unprotective or correlate with a higher likelihood of experiencing these negative feelings. These results highlight the subtle but significant impact that religion may have on individuals’ social connections and mental health, even controlling for other factors.

In answer to our first research questions, but slightly contradicting our hypothesis, I do not find statistically significant negative associations between higher levels of religious service attendance and loneliness, when controlling for other factors. Rather, while I do find non-statistically significant results to this effect, the only statistically significant result is for occasional attendance leading to lessened feelings of being ‘left out’. While this may generally underscore some protective role of engagement in religious congregations against loneliness, it suggests that more occasional attendance may provide some level of protection against loneliness. This relationship nuances the theory posited by Lim and Putnam (2010) regarding the personal and social benefits derived from religious service attendance, as well as empirical research to this effect (e.g., Acevedo et al. 2014; Bartkowski et al. 2017; Hastings 2016; Rote et al. 2013).

A primary finding in answer to the first research question reveals that individuals identifying with minority religious affiliations, especially Judaism and ‘other’ religious identities, but also Black Protestants, Hispanic Catholics, and lesser minority affiliations of ‘unidentified’ and Mainline Protestants exhibit higher instances of feeling loneliness in one or more of the loneliness dimensions explored than those of no religion, even controlling for other factors. The two notable exceptions to this among our categories, then, are Evangelical Protestants and non-Hispanic Catholics. This finding suggests a nuanced layer of social isolation that may stem from the marginalization or a lack of representation within broader societal contexts of minority religious identities (Yang and Nino 2023), as well as potentially suggesting a lack of embeddedness among some Protestant groups.

In answering our second research question, the relationship between religion and loneliness indeed proves to be complex, multifaceted, and variable across religion variables as indicated by our findings. Particularly, the lack of a negative association between spirituality and increased feelings of isolation, as found in some other studies, underscores the potentially unique position of spirituality outside the communal constructs of organized religion. This aspect of our findings resonates with the discussions by Bender (2010) and Ammerman (2013a, 2013b), who highlight the individualized nature of spirituality in contrast to the communal engagement fostered by traditional religious practices. Such solitary spiritual pursuits, while deeply meaningful, may lack the social support mechanisms inherent in congregational settings, potentially lessening any mitigating effects for feelings of loneliness. While self-reported religiosity shows similarly muted results when controlling for other factors, strong religiosity was notably strongly negatively associated with feeling a lack of companionship often or very often.

Again, both the strongest and most varied aspect of religion for these measures of loneliness was religious affiliation. Jewish and ‘unidentified’ Protestant respondents were positively associated with the highest levels of all three measures of loneliness, while ‘other’ religious identities were positively associated with the highest levels of feeling ‘left out’ and ‘lacking companionship’, but not feeling ‘isolated’, at least not to statistically significant levels accounting for other factors. The heightened levels of loneliness among minority religious groups may reflect embeddedness challenges within groups or beyond groups in social integration and visibility within broader societal and religious landscapes, echoing Yang and Nino’s (2023) findings on the social isolation of individuals affiliated with religions outside of (Evangelical) Protestantism or (non-Hispanic) Catholicism. For instance, it may be the case that members of minority religious groups may feel less lonely within their own congregations and social circles, but still have a feeling of being ‘left out’, for instance, from broader society. However, except Evangelical Protestants and non-Hispanic Catholics, those who affiliate with a religious tradition are more likely to have feelings that they often or very often ‘lack companionship’, a dynamic that should be further explored and conceptualized if found to the same degree in future research.

The analysis further reveals little evidence of a mediating effect of civic and social engagement for loneliness but does find strong relationships to demographic factors, notably age and marital status, for experiences of loneliness when considering these aspects of religion. This intersectionality suggests that the impacts of religious and spiritual engagement on loneliness may not be uniformly experienced across the population but are influenced by a confluence of societal, cultural, and personal factors (NASEM 2020). Given the limitation of a single cross-sectional dataset that incorporates these measures of religion and loneliness, only so much can be strongly concluded from these results.

In conclusion, I leverage a rich and diverse dataset to investigate the complex dynamics between religiosity, spirituality, and loneliness across the United States. Through our examination of how religion and spirituality may alleviate feelings of loneliness, I uncover intricate connections between various dimensions of religious life and experiences of loneliness. Notably, our analysis reveals that spirituality alone does not mitigate loneliness when other variables are accounted for, presenting complex and divergent associations between religious service attendance, self-reported religiosity, and different facets of loneliness. Moreover, I observe significant positive correlations between religious affiliation and increased instances of loneliness, particularly among minority religious groups who report heightened feelings of isolation. These insights highlight the subtle and multifaceted roles that religion and spirituality play in emotional health, suggesting that their impact on loneliness is intricate, varies across different religious aspects and dimensions of loneliness, and is influenced by factors beyond religion itself. Consequently, this research calls for further investigation into religious identity and its potentially differential effects on loneliness, aiming to shed light on why individuals affiliated with some religious traditions may experience greater or lesser feelings of loneliness than those of other faith traditions or without such affiliations.