The Role of Religion in the Mental Health of Single Adults: A Mixed-Method Investigation

ABSTRACT

The current investigation examined the role of religion in the experiences of single Polish adults and their mental health. Study 1 employed semistructured interviews, while Study 2 involved two quantitative assessments separated by a 1-month interval. Study 1 (N = 4) identified seven themes related to religion in the context of singlehood. Study 2 (N = 661; M = 32.69, SD  = 6.90) revealed that negative religious coping mediated the links between singlehood boundary ambiguity, and anxiety, depression, and emotional and psychological well-being. A follow-up study (N = 319) showed that higher singlehood boundary ambiguity was related to higher negative religious coping, which was related to greater increases in later anxiety and depression. Singlehood boundary ambiguity appears to prompt single people to exhibit negative religious coping, which negatively affects their positive and negative mental health outcomes concurrently and amplifies the changes in their anxiety and depression over time.

The contemporary transformations in the domain of intimate relationships observed in recent decades in the United States of America, Canada, and many Asian and European countries, have led, among other results, to an increase in the number of single individuals (Adamczyk & Trepanowski, 2023; Girme et al., 2022; Himawan, 2020c; MacDonald & Park, 2022). These changes have been accompanied by a significant shift in focus from the merely negative outcomes of singlehood to an exploration of the factors that may mitigate the effects of single status on mental and physical health outcomes (e.g., Adamczyk, 2021; Girme et al., 2022; MacDonald & Park, 2022). However, a review of the literature reveals that religion is not a major factor that has been investigated in the context of singlehood and the associated health outcomes.

This deficit of research and the omission of religious variables in research on singlehood appears to be quite surprising in light of at least two circumstances. First, most people profess a religious faith (Cooper et al., 2019). For many individuals, religion represents a central aspect of their lives that is essential to their functioning in the context of interpersonal and family relationships, including the process of mate selection and the decisions made regarding the structure of their intimate relationships (Braithwaite et al., 2015; Mahoney, 2010). Second, religion has been considered to be a factor that contributes to both mental and physical health as well as an essential element in coping with stress (e.g., Cooper et al., 2019; Garssen et al., 2021; Krok, 2015; Pargament et al., 1988, 1990, 2001; Park, 2005; Park et al., 2018; Zarzycka et al., 2020).

Ambiguous loss and boundary ambiguity among single adults

One of the conceptualizations of singlehood suggested to be useful in explaining the mental health of single individuals, including psychological distress (e.g., anxiety, depression), particularly among involuntary never-married single adults, involves the concept of ambiguous loss (Jackson, 2018). Ambiguous loss and boundary ambiguity are theoretical constructs that refer to the ABC-X model of family stress theory (Carroll et al., 2007). The term ambiguous loss was introduced by Boss in Boss (1977) to refer to two types of losses, namely, 1) the physical presence and psychological absence of a family member and 2) the psychological presence and physical absence of a family member. The ambiguous loss theory has been applied in research involving a variety of populations (e.g., individuals with autism spectrum disorder, military families, couples coping with cognitive impairment, and families with chronically ill children) (Maxwell, 2015).

Ambiguous loss has also been referred to singlehood. For instance, Sharp and Ganong (2007) considered the non-traditional family life trajectories and the lack of assurance as the fundamental rationale for understanding singlehood among women as an ambiguous loss. In addition, the internal construction of the non-materialized anticipated partner/spouse is related to expectations and desires, the uncertainty of whether and when an anticipated partner/spouse would occur, ambivalence, and how to manage single status (Sharp & Ganong, 2007). Furthermore, in a recent conceptualization of singlehood ambiguous loss, Jackson (2018) proposed that singlehood ambiguous loss is related to the lack of clear information on whether an individual will or will not meet a potential spouse and whether he or she will meet the spouse in the very short or in the very long-term periods.

While ambiguous loss indicates an objective situation/event and refers to a stressor situation, boundary ambiguity emphasizes the perception of this loss and the individual’s/family’s response to a situation of ambiguous loss. Boundary ambiguity appears when a person’s thoughts, feelings, or actions become affected by a lack of clarity concerning the presence or absence of a family member (Boss, 2002; Boss et al., 1990). For instance, a single person who does not have a partner/spouse can be said to exhibit a high degree of boundary ambiguity if they are significantly preoccupied with the fact that they do not know whether they will find a partner/spouse (see Maxwell, 2015). Evidence of this preoccupation would be seen in a single person’s expectation or hope to find a partner/spouse to be found, difficulty determining whether or not to continue to search for a partner/spouse, or constant thoughts regarding the possibility of finding a partner/spouse and whether to embrace single status or attempt to alter this situation (see Maxwell, 2015). Furthermore, higher boundary ambiguity has been recognized to be related to greater personal dysfunction and distress, such as higher levels of anxiety, depression, and stress as well as increased individual and family dysfunction (e.g., Boss, 2002; Boss et al., 1990; Maxwell, 2015). Moreover, ambiguous loss among single adults who strongly desire to have a partner and are involuntarily single has been suggested to be associated with ambiguity, distress, anxiety, blame, guilt, sadness, and immobilization (Jackson, 2018).

Religion in the life of single adults

In line with the notion that religion may represent a source of support and consolation for many individuals and a factor related to mental health and personal adjustment (e.g., Zarzycka et al., 2020), past literature also demonstrated the presence of religion in the lives of single individuals. To be precise, in cases featuring a lack of an attachment figure, such as a romantic partner/spouse, people strive to form an affectional bond with an attachment figure who plays a compensatory role, for instance, God (Granqvist & Hagekull, 2000). In line with this compensation hypothesis, Granqvist and Hagekull (2000) found that single Swedish students were characterized by higher levels of religious activity and personal relationships with God as well as with greater religiosity based on affect regulation and a higher significance of religious beliefs.

Attachment to religious values and rituals has also been suggested to constitute a strategy for coping with singlehood (Himawan et al., 2018a, 2018b). Himawan and colleagues (Himawan et al., 2018a, 2018b) indicated that the adaptive function of religion for single individuals might arise from intrinsic religious orientations, lower levels of negative religious coping, attachments to God, and the feeling that the religious community satisfies the individual’s need to belong. Simultaneously, religion may also have a maladaptive effect on single individuals. In particular, religiosity can lead to negative religious coping, causing individuals to concentrate more on religion than on single status or to disregard religion as a source of meaning in life (Himawan et al., 2018a, 2018b).

Notably, the majority of extant research concerning religion and single individuals has been conducted by reference to samples of single Indonesian individuals. Specifically, Himawan (2020a) found that higher religiosity is associated with higher life satisfaction but not with lower loneliness among unmarried Indonesian individuals. In the same study, by reference to qualitative interviews with unmarried adults, Himawan (2020a) identified three themes pertaining to the role of religion in constructing the meaning of singleness, i.e., 1) single status as a temporary period willed by God, 2) single status as a period that can be used to embrace life outside marriage, and 3) single status as a period during which conservative socioreligious norms are negotiated with contemporary values.

Furthermore, by analyzing qualitative interviews conducted with involuntary single Indonesian individuals, Himawan (2020c) identified two themes pertaining to the role of religion in the lives of single people, i.e., 1) the interpretation of singlehood and marriage through a religious lens and 2) the employment of religious practices to cope with involuntary singlehood. Another analysis of interviews conducted with Indonesian single individuals allowed Himawan (2020b) to report that religion provides single people with a sense of belonging, a close spiritual relationship with God and the religious community, and an alternative way of understanding their single status. Finally, a recent study involving 169 involuntary single Indonesian women belonging to both Muslim and non-Muslim religions showed that higher religiosity was related to greater life satisfaction and that religion explained 15.80% of the variance in the life satisfaction of single women (Lianda et al., 2022).

The elaborated above studies suggest that also for single individuals as for individuals in general religion may be an essential strategy for coping with major stressful life events, including cancer or trauma, as well as with stressors of less severity (Pargament et al., 1990, 1998, 2001; Park et al., 2018). Positive religious coping has been recognized to reflect a secure and trusting relationship with God and involves strategies such as seeking religious support and making benevolent religious reappraisals (Pargament et al., 1998; Park et al., 2018). In turn, negative religious coping has been recognized to reflect a less confident relationship with God and involves strategies such as religious discontent and making punitive religious reappraisals (Pargament et al., 1998; Park et al., 2018).

Religious coping has been demonstrated in previous research to affect health-related outcomes in different ways. Negative religious coping is strongly associated with poorer mental and physical health (e.g., Park et al., 2018), and positive coping is related to better well-being (Pargament et al., 1998). A meta-analysis of 49 studies showed that in general, religious coping strategies are significantly associated with psychological adjustment to stress, i.e., positive religious coping is related to positive outcomes to stressful events and inversely related to negative psychological adjustment, whereas negative religious coping is positively linked to negative psychological adjustment to stress, but negative religious coping is not inversely associated with positive psychological adjustment (Ano & Vasconcelles, 2005). Moreover, in general, studies that have measured positive and negative religious coping revealed that negative religious coping is more strongly and consistently associated with health outcomes than positive religious coping (e.g., Abu-Raiya & Pargament, 2015; Park et al., 2018). For instance, negative religious coping predicts increased depression and negative affect and reduced self-esteem over a period of 2.5 years, whereas positive religious coping predicts decreased depression and increased positive affect, self-esteem, and life meaning (Park et al., 2018).

Polish religious and marital contexts

Because social and marital contexts affect people’s subjective well-being (Wadsworth, 2016) and because it has been suggested that various religious cultures impact individual well-being in different ways (Tay et al., 2014), the contexts of marriage and religion in Poland should also be mentioned briefly here. The data collected in the nine waves of the European Social Survey (ESS) in the period of 2002–2018 revealed that in Poland never-married single individuals aged 18 years and above constitute 34.24% of the adult population, including 56.40% of women and 43.60% men (Adamczyk & Trepanowski, 2023). Poles have been recognized as having a firm commitment to heterosexual marriage and low approval of alternative forms of marital life (Janicka & Szymczak, 2019; Niehuis et al., 2021). For instance, census data demonstrate that 45% of Poles hold negative attitudes regarding single people (Bożewicz, 2019).

Furthermore, Poland has been traditionally recognized as a country that exhibits high homogeneity in terms of religious affiliation with the Roman Catholic Church (Żemojtel-Piotrowska et al., 2021). This tendency is confirmed by a recent Polish census, which indicated that approximately 91% of Poles declared their affiliation with the Roman Catholic Church (GUS, 2019). Simultaneously, in contemporary Poland, significant changes can be observed in the domain of religion, namely, a decrease in church attendance and religious interest and an increasing individualization toward religion (Zarzycka et al., 2021), which intensified after the COVID-19 pandemic (Bożewicz, 2022). Therefore, we considered the current mixed-method research in such a changing religious context in Poland as an opportunity to provide crucial insights into religion’s role in single adults’ lives.

Finally, it is probable that the marital and religious contexts of Poland may affect the mental health of single adults conjointly and may exaggerate the negative mental health outcomes of single people. On the one hand, unfavorable perceptions of single individuals in Poland, i.e., persons who are not in a serious relationship, particularly a marital union (Janicka & Szymczak, 2019), may contribute to lower perceptions of the availability of social support and greater experiences of negative treatment and discrimination compared to partnered individuals (Girme et al., 2022). On the other hand, religion is a strong predictor of individuals’ lives across various cultures (Tarakeshwar et al., 2003), and some religions may attach a high value to marriage (Himawan, 2020a). As a result of the combined impacts of these marital and predominantly Roman Catholic contexts in Poland, single people in Poland may be particularly inclined to use various religious coping styles. The possibility of employing different religious interpretations of singlehood is confirmed by the study conducted by Himawan (2020a), who found that single Indonesian individuals who belonged to different religions conceptualized their single status in different ways. For instance, single Buddhist participants perceived their single status to be a form of karma, whereas Muslim, Protestant, and Catholic single participants viewed their single status as a temptation from God intended to develop their faith and endurance (Himawan, 2020a).

The present study

Although previous research has provided an excellent understanding of the mental health of the growing population of single adults and the role played by religion in people’s lives in general, the recognition of the role of religion for single people and their mental health remains limited, and research concerning this topic has been scarce (e.g., Granqvist & Hagekull, 2000) and has focused mainly on the cultural and religious contexts of Indonesia (Himawan, 2020a, 2020c; Himawan et al., 2018a, 2018b; Lianda et al., 2022). Therefore, to explore for the first time the role played by religion in the lives of single adults and their mental health in marital and religious contexts other than those of Indonesia, we must employ the open, exploratory approach offered by qualitative studies (Patton, 2015; Słysz & Soroko, 2012), which facilitate the exploration, understanding, and investigation of a specific phenomenon that is understood only to a small degree or that has not been fully examined Rendle et al. (2019). Consequently, we utilized the open nature of qualitative research (Study 1) as a starting point to support subsequent quantitative study (Study 2) and designed our mixed-method investigation to employ an exploratory sequential design in which the collection of qualitative data to explore the phenomenon precedes the collection of quantitative data (Creswell, 2012). It means that the data presented in the present paper were collected in the scope of a larger project of exploratory sequential design devoted to singlehood and its life outcomes.

In Study 1, semistructured interviews were conducted with 40 never-married single adults living in Poland. Among the 40 individuals who participated in the interviews, four Roman Catholic participants spontaneously raised the issue of religion in reference to their singlehood. They provided rich narratives to highlight the essential role that religion plays in their experience of singlehood. Therefore, in line with the open and flexible nature of qualitative research (e.g., Patton, 2015), the present authors decided to follow this unanticipated research topic of religion regarding singlehood that emerged from the interviews.

The exploratory nature of Study 1 allowed us to develop general statements (see Mayring, 2007) concerning religion and singlehood and suggested that religion may be an important part of the lives of some single people, and one of the functions of religion for single adults may involve coping with singlehood. It means that questions in qualitative Study 1 operated as “compasses” (see Noyes et al., 2019), that is, as a starting point from which we could explore further and potentially ask further questions (i.e., What role can religion play for single individuals beyond the sample consisting merely of four single Roman Catholic individuals?) and develop hypotheses to explore the role of religion in single people’s lives.

Therefore, we based Study 2, i.e., the quantitative study, on religious coping theory (e.g., Pargament et al., 1988, 1998, 2001, 2011; Park, 2005; Park et al., 2018), which has previously been investigated with respect to specific stressors such as bereavement, cancer, and other illnesses (see Park et al., 2018) or divorce (Simonič & Rijavec Klobučar, 2017). However, to our knowledge, religious coping theory has not yet been examined in regard to singlehood, although singlehood can be a powerful source of stress and ambivalence, particularly when individuals remain single but desire to have a romantic partner/spouse (Jackson, 2018; Lewis, 1994; Maxwell, 2015; Sharp & Ganong, 2007).

Recognizing the fact that in times of stress, people use various resources, including religion, to manage stressful life events (Pargament et al., 2001), we intended to assess the mediating roles played by positive and negative religious coping in the link between singlehood boundary ambiguity and mental health, which were measured in terms of negative indicators (anxiety and depression) and positive indicators (emotional and psychological well-being). In past literature, predominantly involuntary singlehood was conceptualized as ambiguous loss and boundary ambiguity (Lewis, 1994; Sharp & Ganong, 2007), and individuals who desired to have a partner/spouse were suggested to experience ambiguous loss (Jackson, 2018). At the same time, previous studies showed that belonging to one of the groups (single out of choice vs. single out of necessity) is more fluid and is not rigid, and this fluidity may concern one’s understanding of choice and coercion (Reynolds et al., 2007).

Therefore, in the present investigation, we acknowledged that single individuals differ in the degree to which singlehood is perceived as voluntary vs. involuntary, and we considered singlehood boundary ambiguity not in dichotomous terms (the presence vs. lack of boundary ambiguity) but rather as a continuum alongside which single individuals may vary.

Finally, in Study 1, a 1-month interval between two assessments was employed to determine whether the mediating role played by religious coping also emerged over a shorter interval rather than employing a longer interval, for instance, such as the 6-month interval used in the study conducted by Rosmarin and colleagues (Rosmarin et al., 2017) or the 2.5-year interval applied in the study conducted by Park and colleagues (Park et al., 2018) because outcomes of the coping process may be not only long-term, but also immediate, or short-term (Lazarus & Folkman, 1984).

We thus proposed the following hypotheses:

H1:

T1 positive religious coping mediates the link between T1 singlehood boundary ambiguity and T1 positive and negative mental health outcomes, i.e., higher T1 singlehood boundary ambiguity is related to higher T1 positive religious coping, which translates into higher T1 positive and lower negative mental health outcomes.

H2:

T1 negative religious coping mediates the link between T1 singlehood boundary ambiguity and positive and negative T1 mental health outcomes, i.e., higher T1 singlehood boundary ambiguity is related to higher T1 negative religious coping, which translates into lower T1 positive and higher negative mental health outcomes.

H3:

T1 positive religious coping mediates the link between T1 singlehood boundary ambiguity and changes in positive and negative mental health outcomes over the course of one month, i.e., higher T1 singlehood boundary ambiguity is related to higher T1 positive religious coping, which translates into higher changes in positive mental health outcomes and lower changes in negative mental health outcomes over a 1-month interval.

H4:

T1 negative religious coping mediates the link between T1 singlehood boundary ambiguity and changes in positive and negative mental health outcomes over the course of one month, i.e., higher T1 singlehood boundary ambiguity is related to higher T1 negative religious coping, which translates into lower changes in positive mental health outcomes and higher changes in negative mental health outcomes over a 1-month interval.

Study 1

Method

Participants and procedure

The participants included in Study 1 were recruited by an advertisement posted on the official and public Facebook profile of the website of the current project. The participants were eligible if they were at least in the period of young adulthood, i.e., in the ages 18–40 (Staudinger & Bluck, 2001). In Study 1, the sample consisted of four never-married, childless, heterosexual Roman Catholics (two women and two men) aged 23 to 43. Three participants (two women and one men) reported that they would like to have a romantic partner but is not a priority for them, whereas one man had a very strong desire to a have romantic partner and finding a partner was very important to him. Among three of these participants, the duration of singlehood ranged from 75 months to 11 years (M = 8 years), whereas one person indicated that he had never been in a relationship. The detailed characteristics of the four participants are provided in Table S1 in the online supplemental materials, which is accompanied by considerations related to power analysis for Study 1.

The authors met all ethical requirements during data collection. Ethical approval for the study was obtained from the Research Ethics Committee at the Faculty of Psychology and Cognitive Sciences, Adam Mickiewicz University in Poznań, Poland (Decision number: 2/07/2020).

Data were collected via in-depth semistructured interviews that were conducted between December 2020 and May 2021. Due to the COVID-19 pandemic, all interviews were conducted online or using mobile applications. All participants received compensation for their participation in the form of vouchers to an online bookstore with a value of 90 PLN (20.23 USD). All participants provided signed informed consent regarding their participation in the interview, recording of the interview via voice recorder, data processing, and the use of their data in publications prior to participating in the interviews.

In Study 1, the interviews were conducted based on an interview guide that was designed and pretested using the bracketing-interview procedure and a pilot study. This guide was adjusted to suit the unique experiences of each participant during the interview. In each interview, experiences related to singlehood were evoked by the following prompt: “Please tell me about your life as a single person [what is it like for you to be single?], including how it used to be, how it is now, and how you see your future.” This approach offered participants the freedom to tell their stories and discuss the issues that were important to them. In response to this stimulus question and subsequent questions, four participants spontaneously raised the issue of religion with respect to their singlehood. The duration of the interviews ranged from 67 to 103 minutes, and each interview was recorded and transcribed verbatim.

Researcher reflexivity

To limit the influence of potential bias and accomplish the credibility of the results, we considered and declared potential biases and employed several reflective strategies to mitigate the research team interference in the course of the data collection and data analysis (Levitt et al., 2017; Rendle et al., 2019). All the potential biases resulting from our relationship status and religion have been openly discussed and acknowledged. All the research team members have been raised in Poland and are familiar with the Polish culture and Roman Catholic religion, which is deeply inherent in the Polish tradition (Zarzycka et al., 2017). The primary researcher has an extensive, 17-year engagement with the topic of singlehood investigated from the quantitative and qualitative perspectives. The third author is also a certified specialist in clinical psychology and a psychotherapist of the Polish Psychiatric Association in the process of certification. The fourth author specializes in developmental and narrative psychology, and one of his research areas concerns the role of narrative activity in emerging adults’ identity development and mental health. The present authors also trained their experience in performing qualitative studies and data analysis under the supervision of one of the most prominent experts in Poland in the field of qualitative research. We also used a bracketing interview procedure and open-ended questions to limit the effects of our perspectives on the data collected. We also closely followed the participants and asked them to consider what had not been asked, and we considered how the relationship between the researchers and the interviewees might have impacted the quality of data (see Levitt et al., 2017).

Data analysis

The data collected from the interviews were subjected to qualitative content analysis (QCA) in line with the theoretical propositions of Graneheim and Lundman (2004), who indicated that qualitative content analysis may reach the level of interpretation reflected in themes which answer the question “How” and which are “a thread of underlying meaning” (p. 107). Our qualitative content analysis represents a conventional content analysis which is generally applied in exploratory work in less explored areas (Green & Thorogood, 2004; Hsieh & Shannon, 2005), and is based on participants’ unique perspectives and grounded in the actual data (Hsieh & Shannon, 2005), and allows for the simple reporting of common issues present in data (Green & Thorogood, 2004).

The entire analysis was conducted according to the following six steps as proposed by Graneheim and Lundman (2004): (1) multiple readings of transcribed interviews by the authors to understand the entirety of the interview’s content; (2) extraction of texts concerning experiences of singlehood and religion and their incorporation into one text to create a unit of analysis; (3) division of the text into meaning units that were subsequently condensed; (4) abstraction of the condensed meaning unit and labeling using a code; (5) discussion of and reflection on the meaning units with codes to achieve consensus regarding the classification of the codes; and (6) formulation of the themes to reflect the latent constructs of categories.

To establish trustworthiness during each phase of analysis, we employed several means recommended in the literature (see Nowell et al., 2017). To be precise, in the first phase, we familiarized ourselves with the data through prolonged engagement, documentation of theoretical and reflective thoughts, and ideas about potential codes/themes. Secondly, we generated initial codes by peer debriefing, researcher triangulation, audit trails of code generation, and documenting team meetings and peer debriefings. Thirdly, we searched for themes by researcher triangulation, diagramming to make sense of theme connections, preparing notes about development, and drawing up hierarchies of concepts and themes. Fourthly, we reviewed themes by researcher triangulation and vetting themes and subthemes by team members. Fifth, we defined and named themes by researcher triangulation, peer debriefing, and team consensus on themes. Finally, we produced the report by member checking and peer debriefing, describing the processing of coding and analysis in detail, providing thick descriptions of context, descriptions of the audit trail, and reasons for theoretical, methodological, and analytical decisions made throughout the entire investigation. As an additional means to establish the credibility and robustness of the current study and its results, the procedure of conducting interviews, data analysis, and reporting and interpreting the obtained results was done in consultation with an independent researcher who was not involved in any of the steps of the present investigation. This was done to introduce an analytic perspective of an independent, external expert (see Robinson, 2014).

Results

The content analysis in line with the procedure proposed by Graneheim and Lundman (2004) identified seven major themes. A detailed description of each theme with quotations from the participants is included in the online supplemental materials.

The first theme, Mental Health Outcomes of Singlehood, pertained to the negative and positive mental health outcomes of singlehood. The participants reported negative mental health outcomes such as loneliness, depression, negative moods, a reluctance to live, a lack of purpose in life, suicidal thoughts, and the feeling of being excluded from various social activities that are perceived as reserved mainly for couples (e.g., going to the cinema or going to a restaurant). Simultaneously, the participants also indicated some positive factors associated with singlehood, such as a sense of freedom and independence, satisfaction with single status, and various domains, including family, friends, and work. The second theme, Religion as a Criterion in Selecting a Partner/Spouse and a Guide in the Relationship Domain, reflected the participants’ perceptions of religion as a provider of scenarios that define the manner in which relationships ought to develop, i.e., from initial dating, through engagement, to a sacramental marriage. This theme also pertained to the desire to find a partner/spouse who would be as religious as the participants were and for whom religion was a salient aspect of life.

With respect to the theme Church Community as a Potential Community for Finding a Partner/Spouse, the issue of searching for and selecting a suitable partner/spouse was emphasized. Specifically, the participants considered church communities and attendance masses for single individuals as a potential field for finding a partner/spouse because such religious communities may allow people with similar religious values to associate with one another. The fourth theme, Giving Up Control of Singlehood to God and Finding a Partner/Spouse, addressed the belief that God has a plan concerning one’s ability to find a partner/spouse. This belief was associated with relying on and trusting God to guide participants to an appropriate partner/spouse. This theme also reflected the experience of turning to God through prayer, attending mass, confession, and attending retreats and church services. The theme Redefining Singlehood through Religion as a Benevolent and Advantageous Situation reflected the participants’ active attempts to reframe singlehood in terms of positive and advantageous life situations, for instance, by recognizing singlehood as part of God’s plan with the aim of allowing an individual to have more time for family and friends or to serve as a good daughter, sister, or friend rather than a wife. The sixth theme, Internal Struggling Between Religious Values and Desires in the Domain of Romantic Relationships, pertained to the issue of the conflict and internal struggle between religious values and life paths in the domain of marital and family life that are acceptable by the Catholic Church and the individual’s desires in the context of romantic relationships, including doubts concerning whether the individual wanted and was able to follow the Catholic model of marriage. Finally, the theme Confusion and Dissatisfaction reflected experiences of confusion and dissatisfaction with the individual’s relationship with God in light of a period of prolonged singlehood and ineffective attempts to find a partner/spouse.

Discussion

Study 1 demonstrated that single individuals may experience both negative and positive outcomes in the domain of mental health and that single people may position themselves with respect to singlehood via religion, which can be an essential part of their entire experience of singlehood. Study 1 suggested that religion may have various functions in single individuals’ lives, including coping and non-coping functions. Based on these findings, in Study 2, we focused on the coping role of religion (Pargament et al., 1988, 1998, 2011; Park, 2005) since some individuals may experience singlehood in terms of ambiguous loss, boundary, and stress, which may incline individuals to start the process of coping (Jackson, 2018).

Study 2

Method

Participants and procedure

The participants in Study 2 were recruited via an advertisement posted on the official and public Facebook profile of the current project website. Ethical approval was obtained from the Research Ethics Committee at the Faculty of Psychology and Cognitive Sciences, Adam Mickiewicz University in Poznań, Poland (Decision number: 2/07/2020).

Study 2 was conducted in the form of an online survey in the Polish language distributed via the Qualtrics platform between January and March 2022, which featured two assessments over a 1-month interval. The participants were offered the possibility of winning lottery vouchers to the online bookstore with values ranging from 20 PLN (4.50 USD) to 100 PLN (22.47 USD) at T1 and from 30 PLN (6.74 USD) to 100 PLN (22.47 USD) at T2 as compensation for their participation in the study.

Initially, at T1, 1243 individuals participated in the survey; however, only 661 respondents were included in the final analyses (detailed information regarding the recruitment of the samples at T1 and T2 is provided in the online supplemental materials).

The total sample at Time 1 consisted of 661 single participants who identified themselves as Poles, including 370 females (56.00%), 285 males (43.10%), and six persons who indicated another gender (0.90%) with a mean age of 32.69 years (SD = 6.90) (ranging from 18 to 56 years). The mean length of singlehood was 10.15 years (SD = 11.91). With respect to perception of singlehood as voluntary vs. involuntary, 26.60% of individuals strongly disagreed and 31.90% of individuals rather disagreed with the statement that status is determined by their choice and decision.

In terms of religious affiliation,¹ 337 participants (51%) were not religious, 275 were Catholic (41.60%), 10 were Protestant (1.50%; 2 were Baptist (0.30%), six were Lutheran (0.90%), and two were other Protestant (0.30%)), one was a Conservative Jew (0.20%), 10 were Buddhist (1.50%), one was a Muslim (0.20%), three were New Age believers (0.50%), and 20 declared that they belonged to some “other religion” (3.00%).

Among the 661 eligible respondents at Time 1 (T1), in the second assessment (T2), 324 individuals dropped out of the study, and 319 were retained at T2 (for an attrition rate of 51.74%). The detailed characteristics of the sample at T1 and T2 as a function of attrition rate are provided in Table 1. The dataset referenced by Study 2 is available from the Open Science Repository (OSF) at https://osf.io/vb6kz/?view_only=2c96e532f6ef48ff95b0872c5184814a (Adamczyk & Trepanowski, 2022).

Table 1. Sample characteristics of the total sample at Time 1 and Time 2 and as a function of attrition rate between time 1 and time 2.

		W1-W2 Attrition Comparisons
Variable	Total sample at T1 (N = 661)	Participants who were at T1 and T2 (n = 319)	Participants who were at T1 and dropped at T2 (n = 342)	Difference
Demographic variables
Age, years				p = .307
Range	18–56	18–56	19–51
M (SD)	32.69 (6.90)	32.95 (7.06)	32.40 (6.73)
Gender, n (%)				p = .095
Female	370 (56.00%)	192 (60.20%)	160 (46.80%)
Male	285 (43.10%)	125 (39.20%)	178 (52.00%)
Other	6 (0.90%)	2 (0.60%)	4 (1.20%)
Place of residence, n (%)				p = .024
Village	66 (10.00%)	27 (8.50%)	39 (11.40%)
City <10, 000	23 (3.50%)	7 (2.20%)	16 (4.70%)
City 11,000–50,000	75 (11.30%)	28 (8.80%)	47 (13.70%)
City 51,000–100,000	48 (7.30%)	21 (6.60%)	27 (7.90%)
City 101,000–499,000	134 (20.30%)	65 (20.40%)	69 (20.20%)
City >500,000	315 (47.70%)	171 (53.60%)	144 (42.10%)
Highest education, n (%)				p = .714
Primary education	5 (0.80%)	3 (0.90%)	2 (0.60%)
Lower secondary education	1 (0.20%)	-	1 (0.30%)
Vocational education	5 (0.80%)	2 (0.60%)	3 (0.90%)
High school	128 (19.40%)	56 (17.60%)	72 (21.10%)
B.A. degree	115 (17.40%)	52 (16.30%)	63 (18.40%)
M.A. degree	370 (56.00%)	187 (58.60%)	183 (53.50%)
Ph.D. and higher	37 (5.60%)	19 (6.00%)	18 (5.30%)
Relationship status, n (%)				p = .429
Never-married	592 (89.60%)	289 (90.60%)	303 (88.60%)
Divorced	65 (9.80%)	28 (8.80%)	37 (10.80%)
Separated	2 (0.30%)	2 (0.60%)	-
Widowed	2 (0.30%)	-	2 (0.60%)
Length of singlehood (in years), M (SD)	10.15 (11.91)	10.70 (11.40)	9.63 (12.36)	p = .247
Having children, n (%)				p = .260
Yes	56 (8.50%)	23 (7.20%)	33 (9.60%)
No	605 (91.50%)	296 (92.80%)	309 (90.40%)
Religion follower, n (%)				p = .470
Yes	324 (49.00%)	161 (50.50%)	163 (47.70%)
No	337 (51.00%)	158 (49.50%)	179 (52.30%)
Religion, n (%)				p = .364
Roman Catholic	275 (41.60%)	139 (43.60%)	136 (39.80%)
Protestant	10 (1.50%)	4 (1.30%)	6 (1.80%)
Baptist	2 (0.30%)	2 (0.60%)	-
Lutheran	6 (0.90%)	2 (0.60%)	4 (1.20%)
Other	2 (0.30%)	1 (0.30%)	1 (0.30%)
Conservative Judaism	1 (0.20%)	1 (0.30%)	-
Buddhism	10 (1.50%)	3 (0.90%)	7 (2.00%)
Islam	1 (0.20%)	1 (0.30%)	-
New Age	3 (0.50%)	1 (0.30%)	2 (0.60)%
Atheism	185 (28.00%)	90 (28.20%)	95 (27.80%)
Agnosticism	146 (22.10%)	65 (20.40%)	81 (23.70%)
Other	20 (3.00%)	10 (3.10%)	10 (2.90%)
Psychological variables
	M (SD)	M (SD)	M (SD)
Singlehood boundary ambiguity	3.01 (1.05)	2.99 (1.05)	3.03 (1.05)	p = .582
Religious positive coping	0.65 (0.76)	0.66 (0.75)	0.64 (0.77)	p =.828
Religious negative coping	0.43 (0.52)	0.47 (0.54)	0.39 (0.49)	p = .061
Religious participation	0.70 (1.13)	0.71 (1.14)	0.68 (1.12)	p = .680
Anxiety	1.35 (1.01)	1.38 (1.00)	1.31 (1.02)	p = .348
Depression	1.34 (0.10)	1.34 (0.95)	1.34 (1.04)	p = .932
Emotional well-being	2.28 (1.29)	2.38 (1.26)	2.19 (1.31)	p = .058
Psychological well-being	2.51 (1.31)	2.53 (1.29)	2.51 (1.33)	p = .796

As Table 1 indicates, the 319 respondents who participated at both T1 and T2 and the 324 respondents who dropped out of the study after T1 did not differ in regard to the sociodemographic variables, with the exception of place of residence (Cramer’s V = .14, p = .024). To be precise, respondents who participated at both T1 and T2 lived in cities containing more than 500,000 residents more often than respondents who dropped out of the study after T1 who, in turn, tended to live in villages and locations with a lower number of residents. These differences in place of residence, however, were weak in terms of magnitude. Regarding psychological variables, no significant differences emerged between the two groups of respondents.

Measures

Singlehood boundary ambiguity

To assess boundary ambiguity at T1 and T2, the 10-item Boundary Ambiguity Scale (BAS; Maxwell, 2015) (Polish adaptation – Adamczyk et al., 2021) was used. The BAS was initially created for single women; however, in its Polish adaptation, the scale also includes single men (Adamczyk et al., 2021). The BAS was developed as a modification of the Boundary Ambiguity Scale (initially known as the Psychological Presence Scale) elaborated by Boss (1977). Example items include “I feel like it will be difficult, if not impossible, to carve out a good life for myself without a partner.” The items included in the measure are rated on a five-point Likert scale ranging from 1 (strongly disagree) to 5 (strongly agree).

Religious coping

Participants’ religious coping was assessed at T1 and T2 using the 14-item Brief RCOPE scale (Pargament et al., 1998, 2011) (Polish adaptation – Jarosz, 2011). The positive religious coping subscale includes seven items related to religious forgiveness, seeking spiritual support, collaborative religious coping, spiritual connection, religious purification, and benevolent religious reappraisal (e.g., “Looked for a stronger connection with God”). The negative religious coping subscale includes seven items related to spiritual discontent, punishing God reappraisals, interpersonal religious discontent, demonic reappraisal, religious focus, and reappraisal of God’s powers (“Wondered whether God had abandoned me”). Participants rated the extent to which they engaged in each aspect of coping with regard to singlehood using a four-point scale ranging from 1 (not at all) to 4 (a great deal).

Religious participation

A single item, “How often do you attend religious services?”, which was rated on a four-point Likert scale ranging from 1 (never or almost never) to 4 (at least one time in a week), was used at T1 and T2 to assess religious participation as a covariate. This item has also been utilized in previous studies, e.g., Fincham and Beach (2014).

Anxiety and depression

To assess anxiety and depression at T1 and T2, the Polish version of the Patient Health Questionnaire-4 (PHQ-4; Kroenke et al., 2009) was used (the Polish translation is available at https://www.phqscreeners.com/select-screener). The PHQ-4 consists of two items assessing anxiety (e.g., “Feeling nervous, anxious or on edge”) and two items assessing depression (e.g., “Little interest or pleasure in doing things”) over the past two weeks. Participants rate items on a four-point Likert scale ranging from 0 (not at all) to 3 (almost every day).

Emotional and psychological well-being

To assess emotional and psychological well-being at T1 and T2, the emotional and psychological subscales of the 14-item Mental Health Continuum – Short Form (MHC – SF; Keyes, 2002) (Polish adaptation – Karaś et al., 2014) were used. Emotional well-being was assessed using three items (e.g., “happy”), and psychological well-being was assessed using six items (e.g., “your life has a sense of direction or meaning to it”), which were rated on a five-point scale ranging from 0 (never) to 5 (every day).

Demographic information

Respondents were asked to indicate their age, the gender with which they identified most, their place of residence, their highest level of education, their civil status, their duration of singlehood, whether or not they had children, and the religion they followed.

Results

The means and standard deviations of the study variables at T1 and T2 stratified by religious and nonreligious status; internal reliability using McDonald’s omega, test-retest stability, and the matrix correlations are presented in Tables S2 – S5 in the online supplementary materials.

Individuals who declared belonging to a religion or denomination reported higher positive religious coping at T1, F(1,619) = 391.37, p < .001, η² = .39 and at T2, F(1,311) = 213.10, p < .001, η² = .41, and lower negative coping at T1, F(1,618) = 30.52, p < .001, η²= .05 and at T2, F(1,305) = 18.90, p < .001, η² = .06 than individuals who declared that they did not belong to a religion or denomination; the effect size of these differences ranged from weak to moderate or large in magnitude. Individuals who declared belonging to a religion or denomination also reported higher religious participation than individuals who declared that they did not belong to a religion or denomination at T1, F(1,659) = 329.34, p < .001, η² = .33 and T2, F(1,373) = 145.76, p < .001, η²= .32, and the effect sizes of these differences were moderate in magnitude. No differences were observed between these two groups of individuals in the domain of mental health and singlehood boundary ambiguity (see Table S2).

All the mental health indicators demonstrated very large stability across the 1-month interval, with r values ranging from .58 (for depressive symptoms) to .72. (for psychological well-being). Each T1 mental health indicators also exhibited significant cross-lagged correlations with T2 mental health indicators that ranged from moderate to large and very large in magnitude (see Tables S3 – S5).

SEM analyses

The substantial SEM analyses were conducted using R script with lavaan (R Core Team, 2020; Rosseel, 2012). The calculations were performed in line with Baron and Kenny (1986) and mediation recommendations for lavaan (Rosseel, 2012). We expected the occurrence of partial mediation in the tested models. The analyses were preceded by a set of confirmatory factor analyses (CFAs) that used the nonparametric weighted least squares mean estimator (WLSM) since the variables lacked normality and were categorical. As reported in Table S6 in the online supplemental materials, all variables had acceptable factor loadings, with the exception of item no. 14, which was included in the negative religious coping subscale of the Brief RCOPE measure (T1_initial). This item was removed, and the CFA was recalculated (T1_final). Subsequently, the divergent validity of the measures was assessed using the Fornell-Larcker criterion (Fornell & Larcker, 1981; AVE should not exceed the squared correlation between any two given variables) and the heterotrait-monotrait ratio of correlations (Hensler et al., 2015; values should not exceed .85, or .90 if the measured constructs were similar). This analysis confirmed divergent validity, demonstrating that each instrument measured different distinctive constructs. The AVE and HTMT values for Models 1, 2 are reported in Tables S7 and S8 in the online supplemental materials.

In the first step, we assessed the concurrent mediating role of religious coping at T1 in the link between singlehood boundary ambiguity at T1 and mental health outcomes at T1 by conducted SEM analysis on the data collected at T1 (Model 1) (see Figure 1). Previous studies investigating religious coping and well-being indicators have included in their analyses covariates such as age, gender, and education (see Park et al., 2018). These variables were also included in the current analyses in connection with religious participation as a covariate since religious participation has been found to be related to high levels of well-being (see Brown et al., 2021). For the purpose of the analyses, gender was coded in the form of a binary variable by removing the six people who reported “other” gender. Direct, indirect and total effects (standardized) for Model 1 are reported in Table S9 in the online supplemental materials.

Figure 1. Model 1 including positive and negative religious coping at Time 1 as mediators in the links between singlehood boundary ambiguity and anxiety, depression, emotional and psychological well-being at Time 1.

Note. Only significant effects are shown in the figure.

Model 1 included four exogenous observable variables (religious participation, gender, age, and education), one exogenous latent variable (singlehood boundary ambiguity at T1), and four endogenous latent variables (positive and negative religious coping, singlehood boundary ambiguity, anxiety, depression, psychological, and emotional well-being at T1). Error covariance was estimated between (1) anxiety and depression; (2) positive and negative religious coping; and (3) psychological and emotional well-being. Model 1 exhibited a good fit to the data (see Table 2). The analyses demonstrated that negative religious coping at T1 partially mediated the links between singlehood boundary ambiguity at T1, on the one hand, and anxiety at T1 (β = .069), depression at T1 (β = .073), emotional well-being at T1 (β = −.068) and psychological well-being (β = −.059) at T1, on the other.

Table 2. Fit indices for models including four latent variables in study 2.

Fit indices	Model 1	Model 2
χ2(df)	2396.37***	1206.84***
Tucker Lewis Index (TLI)	.976	.971
Comparative Fit Index (CFI)	.974	.968
Root Mean Square Error of Approximation (RMSEA)	.045	.036
Standardized Root Mean Square Residual (SRMR)	.051	.055

Note. Robust values are reported.

***p < .001.

In the second step, to explore the mediating function of religious coping with respect to the link between singlehood boundary ambiguity at T1 and changes in anxiety, depression, emotional and psychological well-being between T1 and T2 (Model 2; Figure 2 and Table 2), we used a latent difference score model (see Castro-Schilo & Grimm, 2018). Difference scores were computed in terms of the differences between T1 and T2 scores (see Kisbu-Sakarya et al., 2013) of anxiety, depression, emotional and psychological well-being. The latent difference score variable was used as the outcome in the SEM analysis (Castro-Schilo & Grimm, 2018) to test Model 2, which was identical to Model 1 aside from the expectation of outcome variables, which in Model 3 were deltas. The results of this analysis indicated a significant full indirect effect (full mediation) of singlehood boundary ambiguity at T1 via negative religious coping at T1 on changes in anxiety (β = .096) and depression (β = .094) over the course of one month. Direct, indirect and total effects (standardized) for Model 2 are reported in Table S9 in the online supplemental materials.

Figure 2. Model 2 with positive and negative religious coping at T1 as mediators in the link between singlehood boundary ambiguity at T1 and changes in anxiety, depression, emotional and Psychological Well-being between Time 1 and Time 2.

Note. Only significant effects are shown in the figure.

Discussion

The results obtained in Study 2 demonstrated that greater singlehood boundary ambiguity at T1 was related only to higher negative religious coping at T1, which affected both positive and negative indicators of mental health concurrently by enhancing anxiety and depression as well as by reducing emotional and psychological well-being. Negative religious coping at T1 mediated the link between singlehood boundary ambiguity and changes in anxiety and depression that occurred over a 1-month interval. To be precise, higher singlehood boundary ambiguity at T1 was associated with higher negative religious coping at T1, which translated into greater increases in anxiety and depression over the course of one month.

General discussion

The qualitative content analysis of data collected from the interviews conducted for Study 1 revealed that single individuals may experience their single status in terms of boundary ambiguity. That is, they may become preoccupied with the lack of information regarding whether and when they may be able to find a partner/spouse and experience doubts concerning whether to continue searching for a partner/spouse or give up this search (Jackson, 2018; Maxwell, 2015).

The qualitative findings obtained in Study 1 corroborate the conclusions of previous studies on religion and singlehood in single Indonesian individuals and reveal that some single people may also position themselves with respect to their single status by using religion to understand their single status and cope with it, for instance, by relying on religious interpretations of singlehood in terms of a temporary period willed by God or a period aimed at embracing life outside marriage (Himawan, 2020a, 2020c; Himawan et al., 2018a, 2018a, 2018b, 2018b; Lianda et al., 2022). Notably, the small sample of four participants derived from the total sample of 40 participants who spontaneously raised the issue of religion in reference to singlehood may reflect the declining role of religion in the Polish society (Bożewicz, 2022; Zarzycka et al., 2021).

The results of Study 1, which were obtained due to the open and flexible nature of qualitative research (Patton, 2015; Słysz & Soroko, 2012), operated as a compass (Noyes et al., 2019) for the subsequent quantitative Study 2 and suggested that religion may play various functions in single individuals’ lives, including both coping and non-coping functions. Regarding that singlehood for some individuals may be a source of stress ambiguous loss and boundary ambiguity (Jackson, 2018), in subsequent quantitative Study 2, we focused on the coping functions of religion for single adults’ mental health. This aim was also prompted by the recognition that Poland still may be in general considered as a country in which religion is strongly and positively valued by Poles (Żemojtel-Piotrowska et al., 2021), despite an observed a decline in church attendance and religious interest and increased individualization toward religion (Bożewicz, 2022; Zarzycka et al., 2021).

Before further discussing the results obtained in Study 2, it should be acknowledged that power analysis for indirect effects through positive coping at T1 and T2 fell below an optimal value of < 0.8 (Serdar et al., 2021), which could hinder the emergence of the significance of indirect effects for positive religious coping. Therefore, our results concerning positive religious coping should be interpreted with caution, however, bearing at the same time in mind that power values below < 0.8 do not immediately derail the value of the study (Serda et al., 2021).

The analyses conducted in Study 2 did not support H1, as positive religious coping at T1 was not demonstrated to mediate the concurrent links between singlehood boundary ambiguity and the positive and negative indicators of mental health. That is, positive religious coping was neither capable of enhancing single people’s emotional well-being (i.e., being happy, satisfied, and interested in life) and psychological well-being (e.g., being good at managing the responsibilities of daily life or having a sense of direction or meaning in life) concurrently nor of reducing their feelings of anxiety and depression.

This pattern of results appears to contradict the conclusions of previous research, which has claimed that, in general, positive religious coping is related to improved well-being (Pargament et al., 1998; Park et al., 2018) as well as to lower anxiety (Abu-Raiya & Pargament, 2015), depressive symptoms (see Abu-Raiya & Pargament, 2015; Park et al., 2018), and in general less negative psychological adjustment to events (Ano & Vasconcelles, 2005). In addition, Park and colleagues (Park et al., 2018) also demonstrated that positive religious coping was better able to predict positive affect, while only negative religious coping predicted negative affect. Simultaneously, our findings are congruent with certain other results, for instance, with those obtained by Rosmarin and colleagues (Rosmarin et al., 2017) by reference to a sample of American and Canadian Jewish individuals, among whom positive religious coping predicted higher life satisfaction but not levels of depression or anxiety.

Simultaneously, the present findings suggest that the feelings of boundary ambiguity, i.e., the overwhelming experience of the lack of information regarding whether and when a single person may be able to meet a partner/spouse, uncertainty regarding whether he or she should continue or cease to search for a partner/spouse, or whether he or she should embrace his or her single status or attempt to alter it (Jackson, 2018; Maxwell, 2015), may indeed lead to adverse effects such as the blocking of coping processes (Jackson, 2018), including the blocking of positive religious coping. Such a failure to engage in positive religious coping, in turn, hinders the possibility of experiencing boundary ambiguity in more benign ways and feeling empowered to cope with this ambiguity or accept the ambiguous situation of one’s single status, thereby leading to more positive emotional and psychological well-being and reduced symptoms of anxiety and depression.

It is also probable that the strong adherence to marital and family values and lower acceptance of relationships alternatives in Poland (Janicka & Szymczak, 2019; Niehuis et al., 2021) may enhance the single individuals’ sense of deviating from the socially expected path of life in a formal relationship what, in turn, does not contribute to positive, but rather more negative approach toward singlehood (see Girme et al., 2022) and discourage single individuals from employing positive religious coping. Finally, it is also probable that the mediating role of positive religious coping did not emerge to be significant due to the transformations of religious lives in Poland, which intensified during and after the COVID-19 pandemic, and which are primarily seen in the departure from the church of those who previously practiced regularly, and the increase in the percentage of respondents declaring that they never go to church (Bożewicz, 2022).

Study 2 also examined the mediating effect of religious coping on prospective mental health outcomes measured one month after the first assessment. The analyses that utilized the difference score approach demonstrated that negative religious coping operated as a mediator in the link between singlehood boundary ambiguity and changes in anxiety and depression over the course of one month, thus partially supporting H3 and H4. That is, the more single individuals were preoccupied with their lack of knowledge regarding whether they would find a partner/spouse, expected or hoped to find a partner/spouse, oscillated between continuing or ceasing their search for a partner/spouse, hesitated to decide whether to embrace or change their single status (Jackson, 2018; Maxwell, 2015), and employed negative religious coping to a higher degree, the more they exhibited greater changes in anxiety and depression over the course of one month.

These findings imply that the role of negative religious coping may extend beyond their concurrent adverse function to encompass a temporal, albeit short-term perspective on changes in the negative indicators of single people’s mental health. Once again, this pattern of results confirms, in line with the conclusions of previous research, that negative religious coping is related to poorer mental health (Ano & Vasconcelles, 2005; Park et al., 2018) and that negative religious coping may exhibit stronger and more consistent relationships with health outcomes than positive religious coping (e.g., Abu-Raiya & Pargament, 2015; Park et al., 2018). Our finding is also congruent with the conclusions of previous studies that have shown that negative religious coping may predict greater depression (Park et al., 2018; Rosmarin et al., 2017) and elevated anxiety (Rosmarin et al., 2017) over longer periods of time.

For instance, negative religious coping was found to more strongly predict the negative aspects of well-being (e.g., depressive symptoms, negative affect) 2.5 years after the first assessment (Park et al., 2018).

The significance of negative religious coping revealed in Study 2 might also be related to the fact that a large number of individuals at T1 (58.50%) consisted of single individuals who, to a lesser degree, perceived their single status as resulting from the choice and personal decision. This group of singles might be considered involuntary singles who may have to deal with a sense of failure in achieving the goal of having a partner and may attribute less agency to themselves than those who choose to remain single (Reynolds et al., 2007) and may experience greater boundary ambiguity (Jackson, 2018). As a result, these singles may be more likely to use negative rather than positive religious coping.

Finally, the significance of negative religious coping revealed in Study May 2, be analogical as indicated in regard to positive religious coping related to Poles’ religious lives marked by declining religiosity and participation in religious services (Bożewicz, 2022; Zarzycka et al., 2021). This trend of decreasing religiosity appears to be reflected in the distribution at T1 of individuals who did not belong to a religion or denomination (51%) compared to 49% of individuals who belonged to a religion or denomination. As a result, a higher number of single individuals who did not belong to a religion or denominations might be more likely to employ more negative than positive religious coping. The support for this assumption provides the results from the study investigating the coping strategies with COVID‑19 in which nonreligious people were less likely to engage in more adaptive coping strategies, although there was no difference in the use of dysfunctional coping across (non)religiousness/ (non)spirituality, however higher dysfunctional coping was related to were higher psychological distress among people with higher levels of (non)religiousness and (non)spirituality (Abbott & Franks, 2021).

In summary, we provided unique qualitative insights from the perspective of single individuals regarding the role of religion in the experience of singlehood in connection with a quantitative assessment of the mediating concurrent and temporal functions of both positive and negative religious coping in the link between singlehood boundary ambiguity and both negative and positive indicators of mental health (anxiety, depression, and emotional and psychological well-being). The present study thus suggests that religion and religious coping may be an essential area for research and may have clinical relevance with respect to single individuals. Our exploratory investigation is providing important insight into the role of factors associated with adult singlehood and as a stimulation for further research to further recognize the role of religion for single people. At the same time, we strongly encourage researchers to replicate in future studies our findings obtained in the current mixed method investigation of exploratory nature since replication can increase certainty when findings are reproduced (Open Science Collaboration, 2015).

Limitations and future research directions

The results of this investigation should be considered in light of several limitations of these studies. First, Study 1, which was based on interviews, involved only four never-married single Roman Catholic adults. Although “it is typical in qualitative research to study a few individuals or a few cases” (Creswell, 2012, p. 209), Study 1 does not allow for the generalization of laws and rules and reference of its findings to a wider population but allows for more restricted and modest generalization in the form of context-specific statements (Mayring, 2007), which describe the associations between singlehood and religion in the conditions specified for four never-married single Roman Catholic Poles.

Second, the samples in Study 2 also predominantly consisted of Roman Catholic Poles, thereby increasing the ambiguity of these studies concerning the extent to which the role of religion in the experience of singlehood may be generalizable to members of other religious groups, ethnicities, and nations. Therefore, future research would benefit from performing comparisons across cultures (Tarakeshwar et al., 2003) and religions since individuals who belong to various religions may experience singlehood in different ways (Himawan, 2020a). Furthermore, participants both in Study 1 and 2 predominantly belonged to the periods of young established and middle adulthood, which calls for the need to include in future research individuals in old adulthood, what, in turn, would enable us to determine the role of religion for single adults’ mental health as a function of age.

Third, in Study 2, the distribution of participants who belonged and did not belong to a religion or denomination should be acknowledged. To be precise, our sample in Study 2 in the baseline assessment consisted of 49% of religion or denomination followers and 51% of individuals who declared that they do not belong to any religion or denomination. This share of individuals who did not belong to any religion or denomination might have affected the patterns of results unveiling the role of negative but not positive religious coping with singlehood boundary ambiguity. In connection with this limitation remains the fact that participants in Study 1 and Study 2 were recruited via Facebook, which might affect the distribution of religious and nonreligious individuals in the investigation since social media platforms such as Facebook tend to assemble people with lower levels of religious identification (e.g., Davidson & Farquhar, 2014).

Fourth, in Study 2, we noted a relatively high attrition rate between the first and second assessments. Although the differences between the participants who continued or dropped out of the study pertained only to their place of residence and were small in magnitude, the attrition rate of 51.74% contributed to the notable reduction of the sample size at T2 and may have affected the power analysis required to detect the indirect effects of positive religious coping for single adults’ mental health.

Fifth, Study 2 utilized a relatively short-term interval of one month. Therefore, future research might consider employing longer intervals to determine the associations from the long-term perspective (see Park et al., 2018; Rosmarin et al., 2017). Sixth, Studies 1 and 2 were conducted during the COVID-19 pandemic, which has been shown to affect mental health negatively (Kranz et al., 2020). Therefore, the performance of this study during the COVID-19 period might have exaggerated the role of negative religious coping with singlehood.

Sixth, in Study 2, we did not perform a prospective power analysis when choosing the sample size, nor did we conduct a post-hoc power analysis given its limited usefulness (e.g., Dziak et al., 2020; Heckman et al., 2022; Lakens, 2022). Future studies planning to verify the model tested in Study 2 would benefit from conducting a priori power analyses to determine the optimal sample size.

Finally, Study 2 focused merely on selective indicators of negative and positive mental health. Hence, future research would benefit from including other negative indicators of mental health, such as loneliness, positive indicators, such as subjective well-being, satisfaction with one’s single status, other life domains, as implied by the results of Study 1, and physical health outcomes or lifestyle behaviors. Furthermore, in Study 2, we focused on the coping function of religion implied by Study 1. However, future studies would benefit from also investigating non-coping functions of religion, such as the influence of religion on the partner selection strategies revealed by Study 1 (see Braithwaite et al., 2015). In addition, future research would also benefit from analyzing other aspects and dimensions of religion and religiosity beyond religious coping, for instance, the role of prayer (e.g., Cooper et al., 2019; Puchalska-Wasyl & Zarzycka, 2020), the meaning-making function of religion and spirituality (e.g., Gallegos & Segrin, 2019; Krok, 2015; Park, 2005), and the role of religion as a source of internal struggles (e.g., Zarzycka et al., 2020).

Ethical approval

All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards.

The research was positively evaluated by the Research Ethics Committee at the Faculty of Psychology and Cognitive Sciences, Adam Mickiewicz University in Poznań, Poland (Decision number: 2/07/2020).

Informed consent from participants

Informed consent was obtained from all individual participants included in the study.

Author contributions

Katarzyna Adamczyk: in Study 1 and Study 2 - Conceptualization, Methodology, Investigation, Formal analysis, Data Curation, Writing – Original Draft, Writing – Review & Editing, Supervision, Visualization.

Radosław Trepanowski: in Study 2 - Investigation, Data Curation, Formal analysis, Software, Writing – Original Draft, Writing – Review & Editing, Visualization.

Marta Mrozowicz-Wrońska: in Study 1: Conceptualization, Methodology, Investigation, Formal analysis, Data Curation, Writing – Original Draft, Writing – Review & Editing.

Kamil Janowicz: in Study 1 - Conceptualization, Methodology, Investigation, Formal analysis, Data Curation, Writing – Original Draft.

Acknowledgments

We thank our participants for their trust and participation.

We thank Emilia Soroko, Ph.D. associate professor at AMU in Poznań, for her invaluable advice and guidance on qualitative content analysis.

We thank Monika Zielona-Jenek, Ph.D. assistant professor at AMU in Poznań, for her invaluable advice and guidance on the ethical aspects of our study.

Disclosure statement

No potential conflict of interest was reported by the authors.

Data availability statement

The data that support the findings of Study 1 are not publicly available to honor the participants’ individual privacy but are partially available from the corresponding author, KA, on reasonable request.

The data that support the findings of Study 2 are openly available in the Open Science Framework at https://osf.io/vb6kz/?view_only=2c96e532f6ef48ff95b0872c5184814a

Supplementary material

Supplemental data for this article can be accessed online at https://doi.org/10.1080/10508619.2023.2265275

Additional information

Funding

The study was funded by grant no. UMO-2019/34/E/HS6/00164 issued by the National Science Center in Poland awarded to Katarzyna Adamczyk. The funding agency had no role in the design or execution of the study, data collection, and management, analysis, interpretation of data, or preparation, review, or approval of the manuscript.

Notes

¹ Regarding that religious coping has been investigated in samples including both religious and nonreligious individuals (see Pargament et al., 2011, for a review of studies) and that distinction between positive religious coping and negative religious coping has been demonstrated to be valid among individuals from various religious traditions (e.g., Abu-Raiya & Pargament, 2015), we decided to perform the analyses utilizing the data form religious and nonreligious participants and participants representing various religions.