Religious / Spiritual Struggles and Life Satisfaction among Young Roman Catholics: The Mediating Role of Gratitude

: An extensive review of the psychological literature shows that interactions between religious / spiritual (R / S) struggles and other aspects of human functioning are complex and a ﬀ ected by “third” factors. Still, we have only a few studies that conﬁrm R / S struggles as a source of well-being and indicate the ways in which it happens. In the present study, we aimed to verify whether the relationship between R / S struggles and life satisfaction was mediated by dispositional gratitude that seems to o ﬀ er protection in times of adversity and turmoil. The sample consisted of 440 Roman Catholics (331 women) from Poland aged between 18 and 40. We applied the Religious Comfort and Strain Scale, the Satisfaction with Life Scale, and the Gratitude Questionnaire. In line with our hypotheses, it was conﬁrmed that respondents with higher life satisfaction were more likely to display a higher sense of trust in God. They also declared a lower fear / guilt and perception of God as abandoning people. Gratitude correlated positively and signiﬁcantly with religious comfort, and negatively with emotions towards God and social interactions surrounding religion. Moreover, it can be a ﬃ rmed that dispositional gratitude mediated the relationship between three of four dimensions of religious strain and life satisfaction: religious comfort, negative emotions towards God, and negative social interactions surrounding religion.


Introduction
Poland, with at least three-quarters of adults (87%) identifying themselves as Catholics (Pew Research Center; PRC Pew Research Center), is one of the European countries where Catholics are still the largest religious group. However, different studies (Büssing et al. 2014) report that the situation in Poland has lately begun to change, and religion is gradually becoming a private matter for some believers, beyond institutionalized participation. In fact, the religious profile of Poland, presented by a recent PRC (Pew Research Center) report and based on an overall combined index, shows that 23% of Polish young adults under 40 are significantly less likely to consider religion as very important in their lives, compared with 40% of older ones. Similar patterns are found in 45 other states out of 106 countries surveyed by PRC (Pew Research Center). Although the reasons for the age gap in religious commitment are complex and mostly related to different historical and socio-cultural factors, it is plausible to assume that such disparity may reflect struggles faced by the young generation on different levels of religious and spiritual life (Kwako Golo et al. 2019). A significant body of literature (Sedlar et al. 2018) has revealed that religious/spiritual (R/S) struggles are not uncommon and are strongly interconnected with greater health problems (McConnell et al. 2006). People can struggle with their faith in response to trauma or stressful events (Currier et al. 2018), a serious, life-threatening experience such as heart failure (Magyar-Russell et al. 2014;Park et al. 2011), different forms of cancer (Fitchett et al. 2004;Sherman et al. 2005), bereavement (Burke and Neimeyer 2014;Cowchock et al. 2010;Neimeyer and Burke 2011), indicators of emotional distress (Exline and Rose 2005) such as anxiety (Zarzycka and Zietek 2018) and depression (Exline et al. 2000;Murphy et al. 2016), and a lower well-being or satisfaction with life (Wilt et al. 2016;Zarzycka and Zietek 2018). R/S struggles might be related to extrinsic, intrinsic, and quest religiousness (Hill and Pargament 2008), as well. For example, in a few studies (Johnson et al. 2008), quest, which expresses religious doubt, was positively related to religious struggle, hence questioning one's religious beliefs can be perceived as one type of spiritual crisis (Fisher 2017). An interesting pattern of results was also obtained by researchers exploring distinct motivations for life satisfaction and being religious (Felere and Lavrič 2008). It was found that while well-being positively correlated with intrinsic religiosity, it correlated negatively with extrinsic religiosity and religious quest (Dowson and Miner 2015;García-Alandete and Valero 2013).
Nevertheless, little research exists on the character of the relationship between R/S struggles and life satisfaction (Zarzycka and Puchalska-Wasyl 2019). Specifically, it still remains unclear through which mediating or moderating mechanism both variables are associated. In fact, Hill (2005) observes that comprehending the varieties of "how" an individual's R/S life is experienced remains of utmost importance, especially when psychologists try to understand the process that links religion to some other phenomena. We chose a group of young adults, building upon the conclusions of some researchers (Galek et al. 2007;Krause et al. 1999) who found that the noxious effects of religious doubt were less strongly felt in older than in younger adults. According to some other outcomes (Krause 2015) obtained across different samples, as age increases, religious doubt diminishes.

Religious and Spiritual Struggles
R/S struggles denote hypothetically challenging facets of R/S life (Sedlar et al. 2018) and "occur when some aspects of R/S belief, practice, or experience become a focus of negative thoughts or emotions, concern, or conflicts" (Exline et al. 2014b, p. 208). This phenomenon is also defined as the experience of pressure or confusion about the sacred, spiritual, or religious apprehensions that influence relationships with the Divine, others, and the self (Gutierrez et al. 2017;Pargament et al. 2005).
In the divine domain, some people perceive God more as controlling and just than caring, loving, merciful, and comforting (Edmondson et al. 2008). For these individuals, God is distant, revengeful, and punitive (Ellison and Lee 2010) and they tend to harbor feelings of disappointment, abandonment by God, and anger at a higher power (Ano and Pargament 2013;Ellison and Lee 2010;King et al. 2017). In some cases, troubled relationships with God lead to questioning His existence (Ellison and Lee 2010). Interpersonal R/S struggles implicate religious conflicts between an individual and religious institutions, community members, clergy, followers of other confessions, and/or unbelievers (Exline et al. 2014a;Sedlar et al. 2018;Zarzycka and Zietek 2018). According to Ellison and Lee (2010), not all religious groups are perceived as supportive networks. Indeed, negative interpersonal religious experience may stem from disagreements over ordinary problems linked to jealousy, selfishness, or hypocrisy within the community, and more considerable issues related to theology, politics, morality, or economics. In the intrapersonal or intrapsychic domain (Ellison and Lee 2010), R/S struggles may take the form of both long-lasting religious doubting ) and impressions of unworthiness, fear, and guilt (Ano and Pargament 2013;King et al. 2017). Such persistent and unresolved doubts can threaten important characteristics of personal identity, are severely confusing, and are inversely associated with different indicators of well-being . Distressing worries can emerge from challenges presented by scientific advances, doubts concerning religious doctrines, injustice, loss of meaning in life, one's own lack of coherence, and transgressions (Ellison and Lee 2010;Zarzycka and Puchalska-Wasyl 2019;Zarzycka and Zietek 2018).
Investigations point to a normative character of R/S struggles that constitute an important and expected part of human life (Exline et al. 2014a(Exline et al. , 2014bExline and Rose 2005;Magyar-Russell et al. 2014;Sedlar et al. 2018;Tobin and Slatcher 2016). In fact, some theologians maintain that religious doubts are inevitable and may play a beneficial role in personal spiritual growth (Ellison et al. 2013). Likewise, Pargament et al. (2005, p. 246) emphasize that in both Jewish and Christian traditions, R/S struggles denote essential "forks in the road" that, depending on how they are coped with, can lead to development and renewal or hopelessness and desperation. Resolved, adaptive, or positive R/S struggles usually reflect a secure bond with God, constructive relatedness with others, and lifelong confidence, based on the conviction of being loved by the Divinity. In contrast, unresolved, maladaptive, or negative R/S strains involve a more vulnerable kind of relationship with God, a feeling of spiritual alienation from other believers, and reduced meaning of personal life.
While a significant body of evidence reveals that a positive pattern of religious coping results in mental health (Ellison et al. 2013) and higher life satisfaction through the capacity to make meaning of religious doubt (Zarzycka and Zietek 2018), a negative pattern shows reverse associations with quality of life in its different dimensions and adverse mental health effects among various samples (Ellison et al. 2013).

Life Satisfaction and R/S Struggles
Interest in life satisfaction, understood as a conscious judgment of one's quality of life on the basis of one's own unique set of principles (Pavot and Diener 1993), has a long history in the field of psychology (Lim and Putnam 2010). Various academic researchers distinguish a wide range of factors that affect a subjective assessment of human well-being (Lim and Putnam 2010) and consider religion to be one of them. In fact, religion seems to have beneficial effects on physical and mental health (George et al. 2002;Koenig et al. 2001;Zarzycka and Puchalska-Wasyl 2019;Zarzycka and Zietek 2018). Although a positive relationship between religiosity and life satisfaction is well-established (Krok 2014; ten Kate et al. 2017), less is known about R/S struggles and their potential influence on people's quality of life. The majority of studies report that religious struggles worsen health (Ano and Vasconcelles 2005;Smith et al. 2003), weaken personal adjustment and harm well-being (Ano and Vasconcelles 2005), make recovery and rehabilitation more difficult (Fitchett et al. 1999), and predict higher mortality rates (Pargament et al. 2001). In a cross-sectional survey, Büssing and colleagues (Büssing et al. 2013) found that a lack of positive spiritual feelings ("spiritual dryness") among 3.824 Catholic priests correlated moderately negatively with life satisfaction. Correspondingly, results derived from an ongoing nationwide survey in the United States among participants at least 66 years old (Krause 2015) showed that greater religious doubt was associated with less life satisfaction. In the Alcoholics Anonymous group, Zarzycka et al. (2017) observed that religious comfort correlated positively with the general assessment of quality of life (r = 0.52, p < 0.001) and life satisfaction (r = 0.41, p < 0.001). Fear/guilt and negative emotions towards God correlated negatively with quality of life (r = −0.29, p < 0.01; r = −0.36, p < 0.001). Negative social interactions surrounding religion did not correlate with the assessment of quality of life. In a nationally representative sample of American adults (N = 2.208), the magnitudes of the correlations between the R/S struggle subscales (divine, demonic, interpersonal, intrapersonal, and ultimate-meaning) and life satisfaction were slightly higher than with happiness (between −0.14 and −0.32 vs. −0.08 and −0.30, respectively) (Abu- Raiya et al. 2015). In similar studies (Zarzycka and Puchalska-Wasyl 2019), only divine, demonic, and ultimate-meaning struggle correlated negatively with well-being. In most cases, the relations between different types of religious conflicts mentioned in the majority of the foregoing studies were rather weak.
However, negative forms of religious coping can also relate to positive psychological adjustment (Ano and Vasconcelles 2005) and act as potential "turning points" (Exline and Rose 2005, p. 312). For example, some researchers claim (Desai and Pargament 2015) that people can benefit from their struggles in the religious or spiritual sphere. Other psychologists (Exline and Rose 2005) remark that R/S struggles may, paradoxically, contribute to enriching individuals' lives. Pargament and colleagues (Pargament et al. 2005) observe that religious crises are perceived in many religious traditions as opportunities for spiritual growth. Likewise, this view finds its confirmation in the philosophical wisdom (Proffitt et al. 2007) and psychological theories (e.g., Erikson and his development through well-solved crises; Tedeschi and Calhoun and their concept of post-traumatic growth, to mention only a few) that consider struggles and major crises an occasion for growth experience.

R/S Struggles, Life Satisfaction, and Gratitude
An extensive review of the psychological literature shows that interactions between R/S struggles and other aspects of human functioning are quite complex and often affected by "third" factors (Pargament et al. 2005). Still, we have only a few studies that confirm R/S struggles as a source of well-being and indicate the ways in which they happen (Zarzycka and Zietek 2018). For example, the relationship between R/S struggles and well-being is found to be mediated or moderated by internal dialogical activity (Zarzycka and Puchalska-Wasyl 2019), the support received in times of crisis (Holcomb and Nonneman 2004), age, living status-alone or in a marriage, illness or soundness, and identification with a particular religious community or spiritual tradition (Zarzycka and Zietek 2018). The mediatory or moderator role of the above-mentioned variables may lead to an increase or decrease in well-being during R/S struggles (Zarzycka and Puchalska-Wasyl 2019).
In the present study, we chose dispositional gratitude as a potential mediator between R/S struggles and life satisfaction. Although the word "gratitude" has a number of various connotations, depending on the context in which it occurs (Sansone and Sansone 2010), for our purpose, we assumed the definition according to which dispositional gratitude refers to a life orientation towards perceiving and appreciating the positive (Wood et al. 2008a(Wood et al. , 2008b. This conceptualization is in line with Bertocci and Millard (1963, p. 389) definition, according to which gratitude is "the willingness to recognize the unearned increments of value in one's experience". At its core lies the deep conviction that appreciation arises in the framework of meaning. In other words, being thankful relates to relevant events that do not necessarily have immediately positive character to which people are typically accustomed. Likewise, Emmons (2013) highlights that a grateful attitude to life seems to offer protection in times of turmoil and builds a more extensive resiliency in the face of adversity and crisis. Moreover, it has the power to heal, energize, and transform hardship into opportunity. Some convincing evidence of the positive influence of gratitude in different difficult circumstances comes from several empirical studies. For example, Fredrickson and colleagues (Fredrickson et al. 2003) confirmed that even in the context of the 11 September attacks, gratitude was one of the three most frequently experienced emotions. The tragedy of this event made many people more aware and thankful for their own safety and the safety of their families. Therefore, gratitude seems to be a powerful resilience factor that may help individuals to deal with various forms of crises. A recent report by Wong and colleagues (Wong et al. 2017) confirmed that participants in the Gratitude Group Program showed a significant and clinically meaningful decrease in psychological distress. Hence, gratitude helps people to take up a perspective by which they can view life in its wholeness and are not overwhelmed by difficult conditions. Still, in another study, researchers (Poelker et al. 2017) observed that gratitude played an important role in the lives of Guatemalan adolescents who were economically disadvantaged. In fact, the grateful-feeling youth had a higher desire to repay the kindness shown to them, and experienced greater satisfaction with life. These data confirm that besides its connection with R/S struggles, gratitude positively relates to life satisfaction (Büssing et al. 2018;Fagley 2012;Froh et al. 2009aFroh et al. , 2009b. Well-documented research shows that grateful people are more prompt to perceive positive experiences in their daily life (Büssing et al. 2018) and consider negative events as less influential on the present (Lambert et al. 2009). The latter competence helps to reframe adverse and harmful memories from the past and assists in mood repair resulting from a stressful event (Watkins 2004).

Research Problem and Hypotheses
In this study, we aimed to verify whether the relationship between R/S struggles and life satisfaction was mediated by dispositional gratitude. Whereas there is enough evidence that R/S struggles may exacerbate quality of life (Exline and Rose 2005), the ways in which R/S strains can increase psychological resources and lead to constructive outcomes are still understudied and less established (Zarzycka and Zietek 2018). Therefore, based on the literature review, we hypothesized that: Hypothesis 1 (H1). Religious comfort is positively related to life satisfaction, and other dimensions of religious struggle (fear/guilt, negative emotions towards God, negative social interactions related to religion) correlate negatively with life satisfaction.
Hypothesis 2 (H2). Religious comfort is positively related to a grateful disposition, and other dimensions of religious struggle (fear/guilt, negative emotions towards God, negative social interactions related to religion) correlate negatively with a grateful disposition.

Hypothesis 3 (H3).
A grateful disposition mediates the effect of four dimensions of religious struggle on life satisfaction.
Hypothesis H3 can induce some questions, as the appropriateness of a cross-sectional design for studying mediation is still being discussed (Aguinis et al. 2017;Maxwell and Cole 2007). Nevertheless, Kline (2015) and Fairchild and McDaniel (2017) recognize that an indirect effect can be established when there is a substantial rationale for temporal ordering specifications of the examined variables. In this respect, the literature pertinent to the assessment of the relationship between religion and satisfaction mostly shows that religiosity affects life satisfaction (Yeniaras and Akarsu 2017). Silberman (2005, p. 645) describes religiousness as "a lens through which reality is perceived and interpreted". This is in accordance with theories that treat religion as an antecedent which has its psychological effects on cognitions, emotions, and behaviors (Heelas 1985). Although there are no studies on R/S struggles and life satisfaction with a grateful disposition as the intervening variable, there is some empirical research on the psychological antecedents of satisfaction and well-being that provides an initial base that such an indirect effect can be confirmed.

Ethics Approval
The study and the protocol were approved by the Bioethics Committee of the Institute of Psychology at the University of Szczecin and conducted according to the Declaration of Helsinki.

Participants
The sample consisted of 440 Roman Catholics (331 women) aged between 18 and 40. The average age was approx. 23 (M = 23.29; SD = 5.06). In terms of subjective evaluations of their own levels of religiosity, 8% of participants declared a lack of religious involvement, 33%-low, 25%-medium, 26%-rather high, and 8%-very high.

Data Collection
The participants were recruited through selected online Internet communities and groups for Roman Catholics. All of the respondents who agreed to take part in the survey were presented with general information about the research goals and were prompted with a web-based informed consent. Only after providing their agreement, the participants were invited to fill in the questionnaires. Respondents under the age of 18 and above 40 were excluded from participation as a target sample group was selected to meet the "young adults" criterion.

Assessment of the Religious Comfort and Strain Scale (RCSS)
RCSS, designed by Exline and colleagues (Exline et al. 2000) and adapted into Polish by Zarzycka (2014), is a questionnaire of 28 items that measures religiosity as a source of comfort (one subscale) and struggle (three subscales). Participants rate each item on an 11-point Likert scale (from 0 = not at all to 10 = extremely) (Zarzycka 2014). Religious comfort estimates the sense of trust in God, and feeling God as omnipotent and caring. It assesses perceiving faith as a source of strength, meaning, and purpose in life (Zarzycka et al. 2017). The subscale of fear and guilt concerns a preoccupation with one's own misdeeds and feelings of being unforgiven by God. The subscale of negative emotions toward God refers to negative feelings toward God, perceiving God as abandoning people, unjust, punishing, and untrustworthy. The subscale of negative social interactions surrounding religion evaluates the intensity of adverse emotions and relations with fellow congregants and believers. In the current study, the α of religious comfort was 0.98, of fear/guilt was 0.80, of negative emotions toward God was 0.84, and negative interactions surrounding religion was 0.78.

Assessment of the Satisfaction with Life Scale (SWLS)
SWLS, originated by Diener et al. (1985) and adapted into Polish by Juczyński (2001), is a short five-item scale that measures one's life satisfaction as a whole, not in specific aspects. The respondents evaluate each of five statements by using multiple-choice answers on a 7-point Likert scale that ranges from 1 = strongly disagree to 7 = strongly agree. The possible range of scores is between 5 and 35. The higher the final score, the more intense is the general satisfaction with life. Different studies report a good coefficient alpha of 0.82 (Diener et al. 1985). In the present study, Cronbach's alpha was 0.85.

Assessment of the Gratitude Questionnaire (GQ-6)
GQ-6, developed by McCullough et al. (2002) and adapted into Polish by Kossakowska and Kwiatek (2014), is a six-item self-report instrument assessing gratitude as a dispositional trait (example item: "I am grateful to a wide variety of people"). Respondents endorse each item on a 7-point Likert-type scale, ranging from 1 = strongly disagree to 7 = strongly agree. Two items are negatively worded and require reverse coding (n • 3 and n • 6). The higher the sum of points, the greater the intensity of dispositional gratitude. The reliability coefficients Cronbach's α of the original English version of the questionnaire was 0.82. In the adapted version, α was 0.71, and in the present study, α was 0.67. Such reliability recurs in many studies (Kossakowska and Kwiatek 2014).

Statistical Analysis
Data analyses were performed using the Statistical Package for the Social Sciences (SPSS software version 20, IBM, Armonk, NY, USA) with significance accepted if p < 0.05. Missing data was averted as the statements of each online questionnaire needed to be 100% complete before continuing to the next section of scales. The data distribution was checked using the statistics of skewness and kurtosis to determine how much the variables deviated from the normal curve (Morgan and Griego 1998). Descriptive statistics were calculated and Pearson's correlation coefficients ®for parametric data were assessed to explore the association between variables including religious struggle, life satisfaction, and dispositional gratitude.
The statistical power for the current study was conducted using G*Power 3.1.9.4 (Faul et al. 2007) to determine the appropriate sample size (Anderson et al. 2017). We applied the recommended higher power criteria of 0.95 (Anderson et al. 2017;Lakens 2013) and a significance criterion α of 0.05 to detect a small and realistic effect (f 2 = 0.025 is considered a benchmark threshold for a small effect). The reason for using the value of f 2 was because there was no previous research on the relationship between religious struggle and life satisfaction among nonclinical groups of participants. In a case of novel studies, Cohen (1988) suggests such a choice. Moreover, different scholars (Aguinis et al. 2005;Martell et al. 1996) claim that as long as a small effect has a significant impact for science or practice, it should be considered in research designs. The G*Power analysis showed that a power of 0.95 would require a minimum sample size of 435 respondents.
Furthermore, we used a linear regression model to control for various possible confounding variables and to verify if they were not a threat to the validity of mediation analyses. First, considering that four dimensions of religious comfort and strain were associated with each other and used as predictors while accounting for covariates, we examined whether there would be a high redundancy among predictors and multicollinearity problem by using an index of tolerance statistics and variance inflation factors (VIFs). Second, we checked the data for outliers, using Mahalanobis' distance and Cook's distance. Third, the participants' sex and age were included to control for their potential influence on the relationship between the independent variable of interest (religiosity struggle) and with the outcome variable (life satisfaction). In fact, both theoretical and empirical examinations of the impact of religious struggle in relation to demographic variables found differences between women and men, and younger and older participants (Cokley et al. 2013). The potential confounders were entered at Step 1. All variables hypothesized as predictors of life satisfaction were entered at Step 2.
The PROCESS macro (version 3.2) (Hayes 2013) was run to examine whether dispositional gratitude mediated the association between the four dimensions of religious struggle separately and life satisfaction. Figure 1 illustrates the pattern of relationships that characterize the mediation dynamic. Religious comfort, fear/guilt, negative emotions towards God, and negative social interactions surrounding religion were the independent variables, and life satisfaction was the dependent variable. Dispositional gratitude acted as a mediating variable. Thus, there were four single-level mediation models (according to Hayes, we applied Model no. 4), involving three-variable systems. For the current analysis, bootstrapping procedures were performed. As recommended by Preacher and Hayes (2008), 5000 bootstrap samples and 95% confidence intervals were used to estimate the indirect effects that are considered significant if they do not include zero. This method is considered to be more powerful than traditional mediation analyses as it does not assume the normality of the sampling distribution (Hayes 2013).
Religions 2019, 10, x FOR PEER REVIEW 7 of 16 inflation factors (VIFs). Second, we checked the data for outliers, using Mahalanobis' distance and Cook's distance. Third, the participants' sex and age were included to control for their potential influence on the relationship between the independent variable of interest (religiosity struggle) and with the outcome variable (life satisfaction). In fact, both theoretical and empirical examinations of the impact of religious struggle in relation to demographic variables found differences between women and men, and younger and older participants (Cokley et al. 2013). The potential confounders were entered at Step 1. All variables hypothesized as predictors of life satisfaction were entered at Step 2. The PROCESS macro (version 3.2) (Hayes 2013) was run to examine whether dispositional gratitude mediated the association between the four dimensions of religious struggle separately and life satisfaction. Figure 1 illustrates the pattern of relationships that characterize the mediation dynamic. Religious comfort, fear/guilt, negative emotions towards God, and negative social interactions surrounding religion were the independent variables, and life satisfaction was the dependent variable. Dispositional gratitude acted as a mediating variable. Thus, there were four single-level mediation models (according to Hayes, we applied Model no. 4), involving three-variable systems. For the current analysis, bootstrapping procedures were performed. As recommended by Preacher and Hayes (2008), 5000 bootstrap samples and 95% confidence intervals were used to estimate the indirect effects that are considered significant if they do not include zero. This method is considered to be more powerful than traditional mediation analyses as it does not assume the normality of the sampling distribution (Hayes 2013).

Preliminary Analyses
Religious comfort, fear and guilt, negative emotions towards God, negative social interactions surrounding religion, satisfaction with life, and gratitude disposition were screened for skewness and kurtosis to evaluate the normality of the scale's distribution. We assumed values less than ±2 commonly considered acceptable as a normal distribution (George and Mallery 2016). In fact, no variables exceeded the cutoffs of ±2 (Table 1).

Preliminary Analyses
Religious comfort, fear and guilt, negative emotions towards God, negative social interactions surrounding religion, satisfaction with life, and gratitude disposition were screened for skewness and kurtosis to evaluate the normality of the scale's distribution. We assumed values less than ±2 commonly considered acceptable as a normal distribution (George and Mallery 2016). In fact, no variables exceeded the cutoffs of ±2 (Table 1).

Correlation Analysis
In line with our hypotheses (H1 and H2), life satisfaction negatively correlated with fear/guilt and "negative" emotions towards God, and positively with religious comfort (Table 2). These relationships were found to be statistically significant at p-values < 0.01. In other words, it was confirmed that respondents with higher life satisfaction were more likely to display a higher sense of trust in God, a lower fear/guilt, and perception of God as abandoning people. At the same time, gratitude correlated positively and significantly with religious comfort, and negatively with emotions towards God and social interactions surrounding religion.

Multicollinearity and Confounding Variables
Although no consensus occurs regarding which VIF cutoff score is most appropriate for collinearity (Thompson et al. 2017), we assumed a VIF index higher than 5.0 (Hair et al. 2017) and a tolerance value of less than 0.2 as indicative of multicollinearity (Mehmetoglu and Jakobsen 2017). Given that the analysis of multiple regression yielded a VIF of 1.02-1.76 and a tolerance rate ranging from 0.59 to 0.98, multicollinearity indexes suggested little evidence of multicollinearity for these data. Mahalanobis' distance procedure was employed, using the chi-square distribution with a very conservative probability estimate for a case being an outlier (p < 0.001) (Fidell and Tabachnick 2003). Only three cases of 440 were identified as probable multivariate outliers. However, because a trial analysis run with and without the cases identified as outliers showed that they did not have a large effect on the regression and did not change results (Stevens 2009), we decided not to drop them from the analysis. Moreover, Cook's value (between 0.000 and 0.056) was well under the point at which the researcher should be concerned, that is, less than 1 (Fidell and Tabachnick 2003), suggesting that the cases are not problematic in terms of excessive influence on the model (Lomax and Hahs-Vaughn 2012). Hierarchical regression analyses indicated that neither sex nor age made a significant unique contribution to the model, explaining only 0.6% of variance (R 2 = 0.006): sex (β = 0.003, t = 0.085, p = 0.933) and age (β = −0.048, t = −1.179, p = 0.239). Predictors explained additional 29% of the variance even after controlling for the effects of potentially confounding variables (sex and age).

Mediational Analyses
For the purpose of further investigation of the relationship between the assessment of dimensions of religious struggle and life satisfaction, dispositional gratitude was introduced as a potential mediator which could weaken, strengthen, or have no influence on the existing correlation between the independent variables (religious comfort, fear/guilt, negative emotions toward God, and negative social interactions surrounding religion) and the dependent variable (life satisfaction) (Figure 1). Table 3 shows the statistics related to each mediation path.  According to the obtained results, it can be affirmed that dispositional gratitude mediated the relationship between three of four dimensions of religious struggle and life satisfaction: religious comfort, negative emotions towards God, and negative social interactions surrounding religion. In all three cases, the original path c dropped to c' as a result of including the mediator.

Discussion
The first purpose of this study was to examine the relationship between four dimensions of R/S struggles, life satisfaction, and gratitude. The second aim was to verify whether dispositional gratitude mediated the relationships between R/S and satisfaction. To the best of our knowledge, the current study is the first to assess directly the topic of religious struggles, subjective life satisfaction, and gratitude. This research gives enough clear findings for all three hypotheses.
Firstly, the positive correlation between religious comfort and life satisfaction is consistent with the previous literature. For example, Koenig (2012) in his systematic review finds significant positive associations between R/S and well-being in a large majority of results. Krause and Hayward (2013) reveal that people who have greater confidence in God and endorse trust-based prayer beliefs are more likely to declare that they are satisfied in their lives. Moreover, religious comfort is positively correlated with satisfaction considered as a component of life quality. Those individuals who perceive faith as a source of strength and rely on God's help report greater well-being (Büssing and Mundle 2009;Zarzycka et al. 2017). Conversely, preserving negative feelings of guilt/fear towards God and community decreases life satisfaction. These outcomes are confirmed by other studies. The data suggest (Krause and Ellison 2003) that gaining forgiveness from God, and so having lower feelings of fear, alleviates self-reproach and leads to improvement of subjective well-being. Likewise, experiential representations of God or "heart knowledge" (Zahl and Gibson 2012) predict satisfaction with life. Related findings (Büssing et al. 2013) show that the perception of being distant from God negatively correlates with life satisfaction and dispositional optimism. It is understandable if we think that spiritual dryness is a marker of crisis (Büssing et al. 2016) and as such can be considered a threat to a global assessment of own-life quality. In another study, Büssing and colleagues (Büssing et al. 2017) report that engagement in the religious practices of doing good and helping those in need is moderately related to life satisfaction and weakly to well-being.
Secondly, a similar pattern of results can be observed between R/S struggles and gratitude. In this respect, Büssing and colleagues (Büssing et al. 2015) show that reliance on God's help correlates strongly with gratitude and awe. Other studies (Aghababaei et al. 2018) illustrate that religion may facilitate gratitude. In fact, participants who have higher scores on religiosity measures tend to display higher grateful disposition. More precisely, engagement in religious practice and perceiving God as the ultimate source of the benefits (intrinsic religiosity) expresses itself in gratitude. On the other hand, using religion for personal benefits and conformity (extrinsic religiosity) negatively associates with gratitude (Watkins et al. 2003).
Finally, gratitude plays a mediating role in the relationships between three dimensions of R/S struggles (religious comfort, negative emotions towards God, negative social interactions surrounding religion) and life satisfaction. It seems that a grateful disposition, which manifests itself in a readiness to recognize the unearned increments of value in one's experience, is not indifferent to life satisfaction in the context of religious struggles both in their positive and negative dimensions. This outcome is in line with the concept of the broaden-and-build theory. According to this theory (Fredrickson 2001), positive emotions generate broad thought-action repertories that develop enduring personal physical, intellectual, psychological, and social resources. In fact, some researchers (Park et al. 2009) demonstrate that R/S struggles have effects on health behaviors that are mediated by a positive effect of self-assurance. Other findings (Wood et al. 2008a(Wood et al. , 2008b provide further evidence that gratitude leads to lower levels of stress, which undoubtedly goes along with religious strains. In conclusion, compared with less grateful people, more grateful individuals might report experiencing higher life satisfaction in spite of undergoing negative thoughts or emotions with respect to their faith. It could be related to their ability to perceive and appreciate other aspects of life not involved directly in their beliefs and their relationship with God. Other studies (Kleiman et al. 2013) demonstrate that a higher level of gratitude and appreciation is a protective factor against harmful symptoms.

Limitations
Although promising, this research is not exempt from a number of limitations. Given that the participants were recruited mainly through a university website, it cannot be expected that they reflect the general Catholic population. Moreover, the limited information provided, restricted to age, sex, and subjective evaluation of their own level of religiosity, narrows down the possibility of a deeper interpretation of the outcomes. In the light of some studies reported in the introductory part that showed an important role of health assessment in the R/S struggles, it seems that this possible 'disturbing' factor should be taken into consideration. Consequently, in future studies, it would be meaningful to broaden the research samples and include a wider range of relevant potential confounding variables that might have an effect on the results, such as: causes of religious strains, intensity of struggles, engagement in religious practices, subjective assessment of health, distress, and bereavement, economic and marital status, and levels of education. An important limitation concerns the age of the participants, restricted to young Catholics, which makes the study unrepresentative. It would be enriching to extend the sample to middle and late adults to verify the possible similarities and differences between groups in experiencing R/S struggles in the context of life satisfaction and dispositional gratitude. It seems compelling, especially when we see that the available literature on the age differences in the relationship between religious doubt and well-being is ambiguous (Krause et al. 1999). In fact, Fisher (2017) suggests that more research needs to be done in the areas of religious change and faith crises. Furthermore, although we found evidence for the mediating role of grateful disposition on the relationship between R/S struggles and life satisfaction, we were unable to assess causality in these associations because of the cross-sectional design assumed in the present study. However, we examined mediation in cross-sectional data offering an appropriate rationale for the posited mediation process. In the future, it would be meaningful to use a longitudinal design to better understand how R/S struggles might impact life satisfaction through the hypothesized mediating gratitude.

Conclusions
The study increases our knowledge about R/S struggles and gratitude among young Polish Catholics and confirms previous research about the relationship between R/S strains and life satisfaction. Although the correlational and mediatory character of our findings does not allow us to draw causal conclusions, the results imply that the co-occurrence of gratitude in the context of R/S struggles may be meaningful for life satisfaction, strengthening the association between religious comfort and subjective life satisfaction, and alleviating the link between negative emotions towards God/negative social interactions surrounding religion and life satisfaction.