Next Article in Journal
Is Sport Really Religion? Critical Assessments
Previous Article in Journal
A Midrashic and Patristic Journey: Towards an Ethic of Peace Beyond Just War
Previous Article in Special Issue
The Relevance of Frankl’s Logotherapy for Today and the Future: Religion and “Man’s Search for Meaning”
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Religions
Volume 17
Issue 2
10.3390/rel17020147
religions-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

“Caravans of Care”: Religiosity, Meaning in Life, and Family Functioning in Parental Burnout

by
Wioletta Tuszyńska-Bogucka
1,* and
Roman Ryszard Szałachowski
2
1
Faculty of Psychology and Pedagogy, University of Economics and Innovation in Lublin, 20-209 Lublin, Poland
2
Faculty of Social Sciences, University of Szczecin, 71-017 Szczecin, Poland
*
Author to whom correspondence should be addressed.
Religions 2026, 17(2), 147; https://doi.org/10.3390/rel17020147
Submission received: 27 December 2025 / Revised: 21 January 2026 / Accepted: 22 January 2026 / Published: 28 January 2026
(This article belongs to the Special Issue Viktor Frankl and the Future of Religion)
Browse Figures
Versions Notes

Abstract

Religiosity has been proposed as a meta-resource that may help parents endure chronic caregiving demands by anchoring parenting in meaning, values, and a sense of purpose—an idea compatible with Frankl’s view that suffering becomes more bearable when it is intelligible and oriented toward a “why.” In a cross-sectional sample of Polish parents (N = 339), we examined whether religiosity related to parental burnout through existential resources (presence and searching for meaning in life) and family-level resources (family functioning) while accounting for basic sociodemographic factors. The zero-order religiosity–burnout association was small and not statistically significant. Yet an indirect-only pattern emerged: higher religiosity was linked to lower parental burnout primarily through higher presence of meaning in life and better family functioning, including a small but reliable serial connection between these two resources. By contrast, global perceived support and generic searching for meaning did not add explanatory value once core resources were considered. In meaning-centred terms, religious faith may operate less as a direct “shield” against burnout and more as a pathway to sustaining everyday family practices and a stable sense of meaning that helps parents carry long-term demands. In other words, religiosity appears linked to lower burnout mainly indirectly via meaning and everyday family functioning, rather than through a sizable direct association.

1. Introduction

Parental burnout (PBA) is a syndrome of chronic exhaustion, emotional distancing from one’s child, feeling fed up with the parental role, and a perceived gap between the parent one has become and the parent one would like to be. It develops under sustained caregiving strain and is linked to reduced well-being and parenting difficulties (Lin et al. 2023; Roskam et al. 2018; Roskam and Mikolajczak 2020). Cross-national work further indicates that PBA reflects an imbalance between chronic demands and parents’ psychological and social resources, with profiles varying across contexts (Le Vigouroux and Scola 2018; Matias et al. 2023; Roskam et al. 2021; Szałachowski et al. 2025).
This resource-balance view aligns with Conservation of Resources (COR) theory, which frames parental burnout as emerging when chronic caregiving strains valued resources, and highlights “resource caravans” and contextual “passageways” that shape access to protective assets (Halbesleben et al. 2014; Hobfoll 2011, 2012; Hobfoll et al. 2018). Narrative reviews of parental burnout likewise distinguish clustered risk factors—such as chronic overload, perfectionistic standards, limited emotion-focused skills, and constrained support—from protective resources in parents and family systems (Mikolajczak and Roskam 2018; Mikolajczak et al. 2023; Roskam et al. 2018).
Religiosity is a plausible meta-resource in this framework: it can organize lower-order resources—existential (meaning in life), systemic (family functioning), and interpersonal (perceived social support)—and channel them toward adaptation. Religious frameworks offer meanings, norms, and narratives that help caregivers interpret adversity and endure chronic stress (Frankl 1959; Pargament 1997; Park 2005) and are embedded in communities that provide social capital and can be associated with better health and flourishing (VanderWeele 2017a, 2017b). From a family-resilience perspective, belief systems, communication, cohesion, and flexibility are central processes through which families withstand and rebound from chronic strain (Patterson 2002; Walsh 2003, 2016a, 2016b). Frankl’s logotherapy adds a more specific lens by emphasizing the “will to meaning” and the role of meaning in making suffering intelligible and worth enduring (Frankl 1959, 1988), consistent with contemporary meaning-making models (Park 2005, 2010) and with work portraying religiosity as a durable guide in the human search for meaning (Pargament 1997; Park 2010).
Meaning in life is typically differentiated into the presence of meaning and searching for meaning (Steger et al. 2006). Empirical syntheses indicate that the presence of meaning is reliably associated with better mental health, whereas searching for meaning is more ambivalent and can show small adverse associations with well-being—particularly when global meaning is low (He et al. 2023; Li et al. 2021; Park and George 2013). A meta-analysis of purpose in life—a construct closely related to discovered meaning—similarly shows that higher purpose is linked to fewer depressive and anxiety symptoms across clinical and non-clinical samples (Boreham and Schutte 2023).
Perceived social support is an adjacent interpersonal domain often linked to stress and burnout. In stress-buffering accounts, perceived support can reduce the impact of stressors or support coping (Cohen and Wills 1985). In the broader religion-and-health literature, associations between religiosity/spirituality and mental health appear to be largely indirect, operating through psychosocial resources such as social support, meaning, and adaptive coping (VanderWeele 2017a, 2017b). Within family systems, religiosity can also shape organizational patterns and shared rituals, thereby strengthening resilience processes (Patterson 2002; Walsh 2016b).
Building on these lines of work, we frame the religiosity–parental burnout link within an integrated model (Figure 1) grounded in four theoretical anchors: COR (resource loss/gain dynamics and resource “caravans” shaped by contextual passageways; Hobfoll 2011, 2012; Hobfoll et al. 2018), logotherapy (meaning as a core existential resource under chronic strain; Frankl 1959, 1988), meaning-making models (presence vs. searching for meaning as distinct processes (Park 2005, 2010; Steger et al. 2006), and family resilience/system perspectives (family functioning as a systemic protective resource; Patterson 2002; Walsh 2003, 2016a, 2016b). These frameworks motivate treating religiosity as a distal orientation linked to parental burnout mainly through proximal existential and family-systemic resources, and comparing alternative mediation specifications as ordering-compatibility tests rather than evidence of temporal causation.
In the present study, religiosity was operationalized as the centrality of religion (Huber and Huber 2012; Zarzycka et al. 2020). We tested whether religiosity relates to parental burnout indirectly through existential resources (presence and searching for meaning in life) and family-systemic resources (family functioning), while considering perceived social support as an adjacent interpersonal domain. Analyses followed a benchmark–sensitivity logic: Model A served as the benchmark, whereas Models B/C tested whether conclusions about the two core resources (presence of meaning and family functioning) depend on their assumed ordering, interpreted as ordering-compatibility rather than temporal precedence. Beyond documenting whether religiosity relates to parental burnout, the present study clarifies how this association unfolds through meaning-centred and family-level resources.

2. Materials and Methods

2.1. Evolution of the “Religiosity–Parental Burnout” Models

Our modelling strategy combined theory-driven specification with limited, transparently reported model evolution. Guided by COR theory and Frankl’s logotherapy, we first specified a parallel mediation benchmark (Model A), in which religiosity was treated as a higher-order orientation linked simultaneously to existential, family-systemic, and interpersonal resources. After estimating this reference model, we introduced two serial extensions (Models B and C) to examine, in a theory-consistent way, whether the observed associations were more compatible with an existential-first versus a family-first ordering of the two most stable resources. Given the cross-sectional nature of the data, these serial specifications are not treated as tests of temporal precedence. Instead, they serve as competing sensitivity models around the same empirical core pattern. Accordingly, similarity between the meaning-first (B) and family-first (C) results is interpreted as evidence consistent with equifinality (i.e., that the data do not privilege a single serial ordering) rather than as support for one fixed causal sequence.
Notation note (Models B/C). To maintain consistent notation across models, we retain the parallel-model mediator labels (M1 = presence of meaning in life, M2 = searching for meaning in life, M3 = perceived social support, M4 = family functioning). The serial sensitivity specifications involve M1 and M4 only, with the mediator link set to M1 → M4 in Model B (meaning-first) and M4 → M1 in Model C (family-first), while M2 and M3 are included as covariates in the outcome equation.

2.1.1. Parallel Mediation—Benchmark Pattern (Model A)

In Model A, religiosity (X) was examined in a parallel multiple-mediation framework with four candidate resources—presence of meaning in life, searching for meaning in life, perceived social support, and family functioning (M1–M4)—predicting parental burnout (Y). The model estimates unique indirect effects through each mediator while allowing correlations among mediators, without imposing directional relations. Conceptually, Model A provides a benchmark for identifying which resources are most closely associated with lower parental burnout (Appendix A, Figure A1).

2.1.2. Serial Model B: Existential-First Sequence (Conceptual and Estimated)

Drawing on Frankl’s meaning-centred account, the ex ante existential-first specification (Appendix A, Figure A2a) positioned presence of meaning (M1) upstream of family functioning (M4) and perceived support (M3) as candidate pathways linking religiosity (X) to burnout (Y), with searching for meaning (M2) treated as potentially ambivalent. The estimated Model B (Appendix A, Figure A2b) implemented a focused serial structure in which X predicted the two core resources (M1 and M4), and a directional link between them was specified (M1 → M4), yielding two specific indirect effects (X → M1 → Y; X → M4 → Y) and one sequential indirect effect (X → M1 → M4 → Y). Searching for meaning and perceived social support were entered as covariates in the outcome equation. This serial ordering is interpreted as an ordering-consistency test rather than evidence of temporal precedence.

2.1.3. Serial Model C: Family-First Sequence

In a final step, we added a complementary family-first serial model (Model C), grounded in family resilience and systemic perspectives. This specification was introduced as a sensitivity check after estimating the parallel benchmark (Model A) and the meaning-first serial specification (Model B), to examine whether the substantive conclusions depend on the assumed ordering of the two core mediators.
Conceptually, Model C proposes a reversed ordering of the two core mediators relative to Model B. Religiosity (X) is entered as a predictor of both family functioning (M4) and presence of meaning in life (M1), while the family-first assumption is implemented by specifying a directional link between the mediators (M4 → M1). Statistically, this corresponds to a serial mediation structure with the two core mediators reversed relative to Model B, with perceived social support and searching for meaning in life entered as covariates in the outcome equation. As with Model B, this specification is interpreted as a competing ordering tested for compatibility with the observed associations rather than as a temporal claim.

2.2. Hypotheses and Analytic Logic

Drawing on Frankl’s meaning-centred account and a resource-based perspective, we treated religiosity (CRS centrality) as a distal orientation expected to relate to parental burnout mainly through downstream existential and family-systemic resources rather than via a strong zero-order association. Analyses followed a benchmark–sensitivity logic: Model A (parallel mediation) served as the confirmatory benchmark, whereas Models B and C probed whether conclusions about the two focal resources depend on their ordering in a serial chain (meaning-first vs. family-first). Given the cross-sectional design, serial orderings are interpreted as tests of ordering compatibility (equifinality), not temporal precedence.
Primary (Model A; confirmatory) hypotheses:
H1. 
Religiosity will relate to lower parental burnout primarily through a negative total indirect effect via the candidate resources entered in parallel.
H2a–H2b (core resources).
Indirect effects through presence of meaning in life (H2a) and family functioning (H2b) will be negative.
H2c–H2d (secondary).
Indirect pathways via perceived social support and searching for meaning are treated as secondary and may be weak or absent once core resources are included.
H3 (exploratory).
The residual direct association (c′) after accounting for mediators is exploratory and, if nonzero, will be interpreted as model-specific residual variance (compatible with inconsistent mediation).
Sensitivity (Models B/C; ordering tests):
RQ1. 
Do conclusions about the core resources depend on whether meaning is upstream of family functioning (Model B) or vice versa (Model C)?
RQ2. 
Is there evidence for a nonzero sequential indirect component in either ordering, and does it materially change the total indirect effect relative to Model A?

2.3. Measures

Meaning in life. Meaning in life was assessed with the Meaning in Life Questionnaire (MLQ; Steger et al. 2006; Polish adaptation: Kossakowska et al. 2013), which distinguishes Presence of meaning (MLQ_P) and Searching for meaning in life (MLQ_S). Consistent with Frankl’s meaning-centred perspective, Presence reflects an established, life-orienting sense of purpose, whereas Searching may indicate ongoing or unsettled meaning-making and, under chronic strain, can function as an index of existential tension rather than a ready resource (Boreham and Schutte 2023; George and Park 2016; Frankl 1959; Park 2010; Park and George 2013; Steger 2012). In analyses, MLQ_P score was treated as an existential resource, whereas MLQ_S was treated as a potentially ambivalent indicator.
Family functioning. Family functioning was assessed with the Family APGAR Scale (Smilkstein 1978; Polish adaptation: Szałachowski and Tuszyńska-Bogucka 2025). The scale includes five items (Adaptability, Partnership, Growth, Affection, Resolve) rated on a Likert-type scale, with higher scores indicating better perceived family functioning.
Perceived social support. Perceived social support was measured with the Multidimensional Scale of Perceived Social Support (MSPSS; Zimet et al. 1988; Polish adaptation: Buszman and Przybyła-Basista 2017), capturing support from family, friends, and significant others. We focused on perceived (rather than received) support because stress and coping models emphasize that appraisals of available resources shape responses to demands, and because support is most protective when matched to relevant stressors (Cohen and Wills 1985; Kaniasty 2020; Rui and Guo 2022). This focus is also compatible with evidence that religiosity relates to health largely through relational and meaning-related pathways (VanderWeele 2017a, 2017b).
Religiosity. Religiosity was operationalized with the Centrality of Religiosity Scale (CRS; Huber and Huber 2012; Polish adaptation: Zarzycka 2007), which assesses the salience of religious contents across five domains (ideological beliefs, religious experience, private practice, public practice, intellectual interest). We interpret CRS results as indexing centrality (a stable frame of reference) rather than doctrinal content or situation-specific coping.
Parental burnout. Parental burnout was measured with the Parental Burnout Assessment (PBA; Roskam et al. 2018; Polish adaptation: Szczygieł et al. 2020), capturing emotional exhaustion, emotional distancing, reduced parental fulfilment, and contrast with one’s previous parental self.
Total scores. For parsimony, we computed total scores for all instruments except the MLQ to reduce model complexity, limit multicollinearity, and facilitate interpretation of indirect effects (Hayes 2022). MLQ Presence and MLQ Searching were analysed separately, consistent with the scale’s structure and guidance (Li et al. 2021; Steger et al. 2006; Steger 2012). The psychometric properties of the tools we used are presented in Appendix A, Table A1.
Covariates. All analyses adjusted for parent sex, age, number of children, relationship/marital status, and education level (Le Vigouroux and Scola 2018; Mikolajczak and Roskam 2018; Mikolajczak et al. 2023; Ren et al. 2024; Roskam et al. 2018; Roskam et al. 2021; Saadati et al. 2024). Details are presented in Appendix A, Table A2.

2.4. Study Design

The study used a cross-sectional, correlational design to examine indirect mechanisms linking religiosity to parental burnout. Drawing on meaning-centred and resource-based accounts, we treated religiosity (CRS centrality) as a distal orientation expected to relate to burnout through lower-order resources spanning existential (presence and searching for meaning) and family-systemic/interpersonal domains (family functioning; perceived social support). All constructs were assessed via self-report questionnaires at the individual level, with a fixed set of sociodemographic covariates. Analyses followed a benchmark–sensitivity logic: Model A served as the benchmark, whereas Models B/C tested whether conclusions about the two core resources depend on their ordering (meaning-first vs. family-first), with searching and perceived support retained as covariates in the outcome equation.

2.5. Analytic Strategy

Estimation. Unless stated otherwise, main-text models used z-standardized continuous predictors winsorized at the 1st–99th percentiles, baseline (non-winsorized) estimates are reported in Appendix A, Table A3. Model formulations follow Hayes’ PROCESS templates: the benchmark parallel mediation (Model A) is equivalent to PROCESS Model 4, whereas the serial specifications (Models B and C) are equivalent to PROCESS Model 6 (Hayes 2022). Mediation models were estimated with OLS and HC3 standard errors (Long and Ervin 2000). Indirect effects were evaluated using two-sided 95% percentile bootstrap CIs (B = 1000; seed = 42), with evidence for an indirect effect inferred when the CI excluded zero (Preacher and Hayes 2008). Unless noted otherwise, all tested models were adjusted for sex, age, number of children, marital/relationship status, and education (dummy-coded where applicable). Robustness checks addressed influential observations (Cook’s D > 4/n) and multicollinearity (VIF) and are reported in Appendix A, Table A4. Given the cross-sectional design, serial specifications are interpreted as ordering-consistency tests rather than temporal causation (Maxwell and Cole 2007). Analyses were conducted in Python (version 3.14) using the statsmodels package (v. 0.14.3), together with standard scientific computing libraries including NumPy, pandas, SciPy, and scikit-learn. To avoid ambiguity about notation, note that continuous variables were z-standardized; therefore, OLS coefficients for continuous paths can be interpreted as standardized effects. In the text, we denote indirect effects as β_indirect, but these are the standard product-of-coefficients estimates reported as “Indirect effect (β)” in Panels B of tables in the main text and Appendix A.
Robustness checks. We repeated the bootstrap for indirect effects with 5000 resamples (B = 5000) and obtained largely similar confidence intervals and conclusions. We also tested the Presence × Search interaction in the outcome equation to probe a common conditional interpretation of searching for meaning (Appendix A, Table A5, Table A6, Table A8 and Table A9). The only substantive exception was the Model B indirect pathway via family functioning, which shifted from significant to non-significant when the number of bootstrap resamples increased to B = 5000 (Appendix A, Table A8); we therefore treat this particular channel as fragile.

2.6. Assumption Checks and Preliminary Analysis

Descriptives and correlations (Table 1) indicated that religiosity correlated positively with presence of meaning, perceived social support, and family functioning, and negatively with searching for meaning. Presence of meaning and family functioning showed the strongest negative associations with parental burnout; perceived support was more modestly related, whereas searching was weakly positively related. Intercorrelations among meaning, support, and family functioning were moderate, consistent with clustering of resources.

2.7. Sample Characteristics

Participants were recruited via convenience sampling through online parenting groups and social media. Data were collected between September 2022 and January 2023 using an online questionnaire. Eligibility criteria were: (a) age ≥ 18 years, (b) being a parent/legal guardian of at least one child, and (c) sufficient Polish proficiency. No incentives were offered. Missing data were low and primarily item-level. Scale scores were computed according to scoring manuals; when occasional item nonresponse occurred, a score was computed if at least 80% of items were completed (otherwise set to missing). One case was excluded due to missingness on key variables, yielding a final analytic sample of N = 339. Because recruitment relied on posts shared in open online parenting groups and social media, the total number of individuals exposed to the invitation is unknown; therefore, a formal response rate could not be calculated. This recruitment strategy entails potential self-selection, as parents who are more active in online communities or more interested in the study topic may have been more likely to participate.
Analyses used complete-case data for variables included in a given model; effective Ns are reported in the corresponding tables when they differ. The sample comprised 232 women (68.2%) and 108 men (31.8%). Mean age was 44.98 years (SD = 7.72). Participants reported an average relationship duration of 16.40 years (SD = 8.45; range = 2–42) and a mean of 1.78 children (SD = 0.75; range = 1–4). Most were in formal/stable partnerships (~56%), followed by informal partnerships (~24%) and single parents (~16%); other arrangements accounted for ~4% (Appendix A, Table A2). The majority reported higher education and rated their health and financial situation as good or very good. The sample, therefore, reflects relatively high life stability, which may attenuate burnout levels and constrain variability in stress-related processes. Findings should be interpreted with caution regarding generalizability to more socioeconomically strained populations. In addition, because recruitment relied on convenience sampling via online groups, the sample may overrepresent more engaged, higher-functioning parents, further limiting generalizability.

3. Results

3.1. Benchmark Pattern—Parallel Mediation (Model A)

Given the cross-sectional design, mediation paths are interpreted as ordering-compatible association patterns rather than temporal causation. The outcome model explained R2 ≈ 0.29 (95% bootstrap CI [0.24, 0.42]), and the residual direct effect was small and not statistically distinguishable from zero (c′ = 0.076, SE = 0.055, p = 0.166). Two indirect pathways were supported (percentile bootstrap, B = 1000): via presence of meaning in life (β_indirect = −0.093, 95% CI [−0.164, −0.040]) and via family functioning (β_indirect = −0.068, 95% CI [−0.123, −0.027]). Indirect paths via searching for meaning (β_indirect = −0.005, 95% CI [−0.023, 0.013]) and perceived social support (β_indirect = −0.014, 95% CI [−0.040, 0.006]) were negligible. The total indirect effect was negative (β_indirect = −0.179, 95% CI [−0.265, −0.111]) (Table 2, Figure 2).
Among covariates, omnibus tests indicated that education and marital/relationship status were associated with parental burnout, whereas age, sex, and number of children were not (Appendix A, Table A5). For a compact comparison of the covariate-adjusted total association (c) and c′, see Appendix A, Table A7. The sign mismatch (c slightly negative; c′ slightly positive) indicates an inconsistent pattern and is interpreted cautiously in the Discussion. Models B and C test whether conclusions about the two core resources depend on their ordering (meaning-first vs. family-first), rather than implying temporal precedence. Panel A is identical across Table 2, Table 3 and Table A6A because Models A–C retain the same outcome equation; the models differ in the mediator equations and in the decomposition of indirect effects.

3.2. Existential Sequence—Meaning-First Serial Model (Model B)

Model B examined whether the benchmark pattern is compatible with a meaning-first ordering, focusing on the two core resources that carried robust indirect effects in Model A (presence of meaning in life and family functioning). Searching for meaning and perceived social support were retained as covariates in the outcome equation to account for adjacent domains. Coefficients displayed in Figure 3 are direct estimates from the mediator equations (Appendix A, Table A6B).
In the serial specification (Figure 3; Table 3), the predictors explained about 29% of the variance in parental burnout (R2 ≈ 0.29, 95% bootstrap CI [0.24, 0.42]). The residual direct effect of religiosity was small and not statistically distinguishable from zero (β = 0.076, SE = 0.055, p = 0.166). Presence of meaning in life (β = −0.245, SE = 0.076, p = 0.001) and family functioning (β = −0.292, SE = 0.077, p < 0.001) were robustly associated with lower burnout, whereas perceived social support (β = −0.084, SE = 0.063, p = 0.182) and searching for meaning (β = 0.028, SE = 0.053, p = 0.604) were not uniquely associated with burnout once all variables were entered simultaneously. Because the serial specifications (Models B/C) retain the same outcome equation as the benchmark, Panel A is intentionally identical across Table 2, Table 3 and Table A6A; the models differ in the mediator equations and in how the total indirect effect is decomposed into specific paths.
Indirect-effect estimates replicated the core pattern. Religiosity was indirectly related to lower burnout via higher presence of meaning (β_indirect = −0.093, 95% CI [−0.164, −0.040]) and, in the main-text bootstrap (B = 1000), via family functioning (β_indirect = −0.035, 95% CI [−0.083, −0.003]); however, this family-functioning-only indirect was borderline/fragile and became non-significant under B = 5000 (Appendix A, Table A8). An exploratory Presence × Search interaction predicting parental burnout was non-significant (Appendix A, Table A9). The sequential indirect component in the meaning-first ordering was small but reliable (X → Meaning → Family → Y; β_indirect = −0.032, 95% CI [−0.063, −0.013]), and the total indirect effect remained negative (β_indirect = −0.161, 95% CI [−0.250, −0.095]) (Table 3). This pattern is statistically compatible with a meaning-first arrangement as an ordering-consistency test, not as evidence of temporal precedence in the population. In sensitivity checks, the family-functioning-only indirect effect in Model B was borderline/fragile: its 95% CI included zero in the B = 5000 replication (Appendix A, Table A8); accordingly, we interpret this specific channel cautiously. Indirect effects reported in the text (β_indirect) correspond to the product-of-coefficients estimates shown in Table 3 (Panel B).

3.3. Systemic Sequence—Family-First Serial Model (Model C)

The family-first specification (Model C) reversed the order of the two core resources relative to Model B by placing family functioning (M4) upstream of presence of meaning in life (M1) in the serial link (M4 → M1). Religiosity (X) was entered as a predictor of both mediators, while perceived social support and searching for meaning in life were retained as covariates in the outcome equation (Appendix A, Table A6A; Figure A3). Coefficients displayed inside Figure A3 are direct estimates from the mediator equations (Appendix A, Table A6C). Model C, therefore, yields two specific indirect effects (via family functioning; via presence of meaning) and one sequential indirect effect (X → family functioning → presence of meaning → Y).
In the outcome equation (Appendix A, Table A6A), religiosity showed a small positive point estimate that was not statistically distinguishable from zero (β = 0.076, SE = 0.055, p = 0.166). In contrast, better family functioning (β = −0.292, SE = 0.077, p < 0.001) and higher presence of meaning (β = −0.245, SE = 0.076, p = 0.001) were robustly associated with lower parental burnout, whereas perceived social support (β = −0.084, SE = 0.063, p = 0.182) and searching for meaning (β = 0.028, SE = 0.053, p = 0.604) were not uniquely associated with burnout once all variables were entered simultaneously. The model explained about 29% of the variance in parental burnout (R2 ≈ 0.29, 95% bootstrap CI [0.24, 0.42]).
Indirect effects mirrored the benchmark pattern. Religiosity was indirectly related to lower burnout via family functioning (β_indirect = −0.068, 95% CI [−0.123, −0.027]) and via presence of meaning (β_indirect = −0.075, 95% CI [−0.134, −0.031]), and the family-first sequential component was small but reliable (X → Family → Meaning → Y; β_indirect = −0.018, 95% CI [−0.038, −0.006]). The total indirect effect remained negative (β_indirect = −0.161, 95% CI [−0.250, −0.095]) and closely matched the meaning-first serial model. Taken together, Models B and C converge on the same substantive conclusion: religiosity relates to lower parental burnout primarily through the joint configuration of meaning in life and everyday family functioning, rather than through global perceived support or a generic search for meaning.

3.4. Verification of Hypotheses

H1 was supported: the total indirect effect of religiosity on parental burnout was negative in the benchmark model and remained negative across Models A–C. H2a–H2b were supported in the primary analyses: indirect effects via presence of meaning in life and (in the B = 1000 bootstrap) via family functioning were negative; H2c–H2d were not supported because searching for meaning and global perceived social support did not add indirect effects once the two core resources were included. H3 was not supported: the residual direct effect (c′) was small and positive but not statistically distinguishable from zero. RQ1–RQ2: ordering comparisons (Model B vs. Model C) did not change conclusions about the two core resources; the family-functioning-only indirect path in Model B was the sole borderline/fragile component, becoming non-significant under B = 5000 (Appendix A, Table A8).
Key take-home message. Religiosity related to lower parental burnout mainly through higher presence of meaning and better family functioning; global perceived support and generic searching did not add indirect effects, and conclusions about the two core resources did not depend on mediator ordering (meaning-first vs. family-first).

4. Discussion

Religiosity as a Meaning-Centred Resource under Chronic Caregiving Strain. From a meaning-centred perspective, the relevance of religiosity for parental burnout lies not in stress reduction per se, but in its capacity to sustain orientation, purpose, and coherence under conditions of chronic caregiving strain. In line with Frankl’s logotherapy, suffering becomes psychologically bearable when it is intelligible and anchored in a personally endorsed “why.” The present findings align with this proposition by showing that religiosity operates as a meta-resource: it supports a consolidated presence of meaning and facilitates the translation of this meaning into everyday family practices—such as communication, routines, and mutual responsiveness—that are directly implicated in burnout resilience. Notably, once these meaning-centred and family-system resources were accounted for, neither global perceived social support nor an unsettled search for meaning contributed additional explanatory value. This pattern is theoretically informative. It suggests that while searching for meaning may reflect existential tension under strain, it is the presence of meaning—already stabilized and experientially accessible—that confers protection. Clinically, this interpretation cautions against assuming that religiosity or generic support alone buffers parental stress. Instead, it points to the importance of interventions that strengthen meaning in parenting roles and enhance concrete family-system functioning. In this way, the findings refine Frankl’s central proposition by illustrating how meaning-oriented resources are instantiated in everyday family life under sustained caregiving demands.
Religiosity as a meta-resource and a resource caravan: small, context-sensitive effects. Religiosity can be interpreted as a meta-resource insofar as it provides interpretive frames, moral commitments, and identity anchors that can shape appraisals and support routinized practices in daily life. At the same time, the broader mental-health literature suggests that religion/spirituality–mental health associations are typically small and heterogeneous. Meta-analytic and umbrella-review evidence indicate that effects vary by the facet of religiosity assessed, the outcome domain, and the study design (Brečka et al. 2024; Halbesleben et al. 2014). Longitudinal evidence further shows that the direction and size of religion–mental health links can change when within-person processes are modeled, suggesting that religious engagement may sometimes respond to distress rather than causally reducing it (Prati 2024). These findings motivate a cautious interpretation: religiosity may function less as a direct “buffer” and more as a source of meaning-making and behavioral scripts that become protective when translated into concrete family functioning (VanderWeele 2017a, 2017b).
Why presence of meaning matters—and why searching can look different. Presence of meaning in life emerged as a robust mediator. Parenthood-focused research suggests that a stronger presence of meaning in parenthood is associated with more positive parenting-related functioning and experiences (Dong et al. 2025). This aligns with meta-analytic evidence that meaning “presence” is reliably associated with lower psychological distress, whereas meaning “search” is more often associated with higher distress, especially when it reflects unresolved questions or unmet needs (He et al. 2023). In the present context, religiosity may support parental functioning by supplying a coherent narrative for caregiving, thereby strengthening the “presence” component, while not necessarily reducing ongoing searching (e.g., during periods of instability or mismatch between ideals and lived experience) (Park 2010; Steger et al. 2006).
Family functioning as a core resource in the burnout ecology. Family functioning was the second robust mediator and remained consequential regardless of model specification. Recent reviews converge on the idea that parental burnout is tightly coupled with couple and co-parenting processes, including day-to-day coordination, fairness in the division of labor, conflict management, and the predictability of family routines (Ren et al. 2024). Interdependent and dyadic work similarly links non-supportive co-parenting dynamics to higher burnout (Favez and Bader 2024). In related research on parenting stress and burnout, relationship quality and marital satisfaction have been identified as proximal pathways through which stress translates into exhaustion and disengagement (Blanchard et al. 2025; Xu et al. 2024). Longitudinal evidence also suggests that fluctuations in stress and social support covary with parental burnout trajectories, highlighting the importance of sustained relational resources rather than one-time “support availability” (Holopainen et al. 2025). In line with the robustness checks, the Model B family-functioning-only indirect effect should be treated as borderline/fragile (Appendix A, Table A8), even though the association between family functioning and burnout in the outcome equation was stable across models.
Configuration rather than temporal sequence: interpreting the ordering tests. The two serial specifications were treated as ordering-consistency tests rather than evidence of temporal causation. In cross-sectional data, a statistically nonzero sequential component indicates that a proposed ordering is compatible with the observed pattern of associations, not that resources activate in a fixed temporal sequence. The present results were robust to whether meaning in life or family functioning was placed upstream, which supports a configuration interpretation: parents report less burnout when both existential and family-system resources are relatively intact. This reading fits a “resource caravan” view in which resources cluster and reinforce one another, and it underscores why ordering comparisons are best used to check the plausibility of competing conceptual models rather than to infer temporal dynamics (Mikolajczak et al. 2023).
Why global perceived support did not add an indirect pathway. Global perceived social support did not add a detectable indirect effect once meaning and family functioning were included. This null finding is consistent with a growing view that support is most protective against parental burnout when it is parenting-specific and behaviorally enacted—particularly via co-parenting support, shared caregiving load, and reductions in conflict—whereas broad global perceived support may be too distal to register as a distinct mechanism once family functioning is modeled (Favez and Bader 2024; Kaniasty 2020; Ren et al. 2024). Moreover, psychological-process work suggests that support can matter conditionally: in one recent model, emotion-regulation strategies and perceived support jointly related to parental burnout, implying that support may function as an enabling context rather than a stand-alone mediator (Lin et al. 2022).
Potential costs: searching, spiritual struggles, and the small positive residual. Although the residual direct effect (c′) was not statistically distinguishable from zero, its small positive sign is theoretically compatible with potential costs of certain forms of religiosity in parenting contexts. Contemporary research differentiates negative religious coping from broader religious/spiritual struggles and shows that both prospectively predict psychological distress (Cowden et al. 2022). Such struggles can involve guilt, perceived inadequacy, or perceived divine punishment, and they may be amplified when parenting ideals are experienced as moral imperatives. Evidence from coping research during COVID-19 also suggests that specific coping styles—rather than religiosity per se—are associated with symptom burden (Holopainen et al. 2025; Pankowski and Wytrychiewicz-Pankowska 2023; Pereira et al. 2024). Complementing this, protective-factor research on parental burnout during the COVID-19 pandemic indicates that resilience, optimism, and humor are linked to reduced parental burnout and better parental well-being under prolonged stress (Enav et al. 2024). Against this backdrop, the non-mediation by generic searching for meaning is interpretable as well: searching is not uniformly beneficial and can signal unresolved tension, which may coincide with spiritual struggles and contribute to residual risk (He et al. 2023; Mancini et al. 2023; Pargament et al. 2011). Given that c′ is small and not statistically distinguishable from zero, we treat these potential-cost interpretations as hypotheses about heterogeneity and competing pathways (e.g., negative religious coping or spiritual struggles) rather than as evidence that religiosity increases burnout (Exline et al. 2014).
Limitations. Several limitations qualify the present findings. First, the design was cross-sectional and correlational. Accordingly, indirect effects should be interpreted as a statistical decomposition of associations rather than evidence of temporal mediation; the serial specifications (Models B/C) are best read as ordering-consistency tests rather than demonstrations of causal sequence (Maxwell and Cole 2007). Second, all constructs were assessed via self-report in a single survey wave, which may inflate associations through shared method variance and does not capture within-family dynamics as they unfold in daily routines (Podsakoff et al. 2003; Shiffman et al. 2008). Relatedly, the study focused on perceived (global) social support, which may be less proximal to parenting strain than parenting-specific or received support (Haber et al. 2007).
Third, generalization beyond the studied context should be made with caution. The sample reflects one cultural setting and a digitally engaged parent population, which may limit applicability to contexts differing in religious climate, family policy, or parenting norms (Bethlehem 2010). Consistent with this contextual sensitivity, covariate diagnostics indicated associations of education and marital/relationship status with burnout; future research should examine whether the proposed pathways vary across these groups.
Fourth, religiosity was operationalized as centrality of religiosity—a broad orientation—rather than specific practices or coping strategies. While this choice fits the meta-resource framing, it may mask heterogeneity, including negative religious coping and spiritual struggles that could contribute to the small positive residual direct effect (c′) (Huber and Huber 2012; Pargament et al. 2011). Future research should employ longitudinal or intensive repeated-measures designs, incorporate practice-based indicators and differentiated religious coping, and include multi-informant or behavioral indicators of family functioning (De Los Reyes and Kazdin 2005).
Implications and directions for future work. The applied implication follows directly from the indirect-effect pattern: prevention and intervention efforts may be more effective when they strengthen stable meaning in parenting and improve everyday family functioning (communication, routines, co-parenting coordination), rather than simply intensifying religiosity or offering diffuse reassurance of support. Intervention-oriented research provides early evidence that structured, skills-based programs—such as cognitive-behavioral stress-management approaches tailored for parents—can reduce parental burnout (Urbanowicz et al. 2023). Future work would benefit from longitudinal designs that separately model within-person change and between-person differences, include parenting-specific support measures (instrumental co-parenting, division of labor, conflict processes), and assess negative religious coping and spiritual struggles directly to test whether they account for small positive residuals in the religiosity–burnout association (Prati 2024; Ren et al. 2024). Education and marital/relationship status showed omnibus associations with parental burnout (Appendix A, Table A5) and may therefore be candidates for moderation tests in future work (e.g., moderated mediation), rather than being treated only as covariates.
Conclusion. Religiosity was associated with lower parental burnout mainly through a configuration of higher presence of meaning in life and better family functioning. Global perceived support and searching for meaning did not provide unique mediation once core resources were included. A small positive c′ point estimate co-occurred with protective indirect effects and should, therefore, be interpreted cautiously until potential cost-related mechanisms are measured directly. Put metaphorically, religiosity may be a lighthouse rather than a breakwater: it does not remove the storm, yet it can illuminate a sense-making route and support coordinated family routines that help parents endure ongoing demands with less burnout.

Author Contributions

Conceptualization, R.R.S. and W.T.-B.; methodology, R.R.S. and W.T.-B.; software, W.T.-B.; validation, R.R.S. and W.T.-B.; formal analysis, R.R.S. and W.T.-B.; investigation, R.R.S. and W.T.-B.; resources, R.R.S. and W.T.-B.; data curation, R.R.S. and W.T.-B.; writing—original draft preparation, R.R.S. and W.T.-B.; writing—review and editing, R.R.S. and W.T.-B.; visualization, W.T.-B.; supervision, W.T.-B.; project administration, W.T.-B.; funding acquisition, R.R.S. and W.T.-B. All authors have read and agreed to the published version of the manuscript.

Funding

This research received no external funding.

Institutional Review Board Statement

The study was conducted in accordance with the Declaration of Helsinki and approved by the Institutional Ethics Committee of the Department of Psychology, Faculty of Social Sciences, University of Szczecin, 71-017 Szczecin, Poland (KB15/2022) for studies involving humans.

Informed Consent Statement

Informed consent was obtained electronically prior to participation in the anonymous online survey through active confirmation after reading the study information sheet. Participation was voluntary, withdrawal was possible at any time without penalty, no identifying data were collected, and all data were processed and reported in aggregated form in accordance with GDPR regulations.

Data Availability Statement

The datasets generated and analyzed during the current study are available from the corresponding author on reasonable request.

Acknowledgments

During the preparation of this manuscript/study, the authors used ChatGPT (OpenAI; GPT-4-series, accessed in 2025) for the purposes of language aid to improve clarity and style, and to support the preparation of analytical scripts in Python. The authors have reviewed and edited the output and take full responsibility for the content of this publication.

Conflicts of Interest

The authors declare no conflicts of interest.

Abbreviations

The following abbreviations are used in this manuscript:
PBAParental burnout
CORConservation of Resources Theory
MLQMeaning in Life Questionnaire
MLQ_PPresence of meaning in life (MLQ scale)
MLQ_SSearching for meaning in life (MLQ scale)
CRSCentrality of Religiosity Scale
MSPSSMultidimensional Scale of Perceived Social Support

Appendix A

Appendix A.1. Theoretical Models

Religions 17 00147 g0a1
Figure A1. Parallel mediation—benchmark pattern (Model A).
Figure A1. Parallel mediation—benchmark pattern (Model A).
Religions 17 00147 g0a1
Religions 17 00147 g0a2
Figure A2. Competing specifications of Model B: (a) conceptual serial ‘existential-first’ model (ex ante specification), (b) estimated meaning-first serial mediation model (Model B; PROCESS Model 6), specified after the Model A benchmark (core resources), with searching for meaning and perceived social support included as covariates.
Figure A2. Competing specifications of Model B: (a) conceptual serial ‘existential-first’ model (ex ante specification), (b) estimated meaning-first serial mediation model (Model B; PROCESS Model 6), specified after the Model A benchmark (core resources), with searching for meaning and perceived social support included as covariates.
Religions 17 00147 g0a2

Appendix A.2. Measures and Metrics

Table A1. Psychometric properties of the measures.
Table A1. Psychometric properties of the measures.
ConstructScale (Polish Version)ItemsResponse FormatObserved Range (This Study)Higher Scores Indicate…Typical Internal Consistency (Prior PL Data)
ReligiosityCRS (Centrality of Religiosity)155-point Likert15–75Stronger centrality/salience of religionhigh (α ≥ 0.90)
Presence of meaning in lifeMLQ–Presence57-point Likert5–35Greater presence of meaning in lifehigh (α ≈ 0.80–0.90)
Searching for meaning in lifeMLQ–Search57-point Likert5–35More active search for meaninghigh (α ≈ 0.80–0.90)
Perceived social supportMSPSS127-point Likert12–84Higher perceived support from family, friends and close othersvery high (α ≈ 0.90–0.95)
Family functioningFamily APGAR53–5-point Likert0–10Better perceived family functioninggood (α ≈ 0.70–0.85)
Parental burnoutParental Burnout Assessment (PBA)237-point Likert0–92Higher parental burnoutvery high (α ≈ 0.90–0.95)

Appendix A.3. Participants

Table A2. Sample socio-demographic characteristics at baseline (N = 340; one case was later excluded due to missing data, yielding N = 339 for analyses).
Table A2. Sample socio-demographic characteristics at baseline (N = 340; one case was later excluded due to missing data, yielding N = 339 for analyses).
VariableCategory/StatisticValue
Year of birthMin/Max1960/2001
Year of birthIQR (25th–75th percentile)1975–1985
Sex (1)Women232 (68.2%)
Sex (2)Men108 (31.8%)
Education (1)Primary or lower1 (0.3%)
Education (2)Basic vocational28 (8.2%)
Education (3)Secondary55 (16.2%)
Education (4)Bachelor/engineering degree155 (45.6%)
Education (5)Master’s degree or higher101 (29.7%)
Number of childrenM (SD)1.78 (0.75)
Number of childrenRange1–4
Marital status (1)Single2 (0.6%)
Marital status (2)In a relationship, not cohabiting12 (3.5%)
Marital status (3)Cohabiting (unmarried)83 (24.4%)
Marital status (4)Married190 (55.9%)
Marital status (5)Divorced/widowed/other53 (15.6%)

Appendix A.4. Robustness Tests

Table A3. Robustness of the serial-model outcome equation (core specification used in Models B/C): baseline (no winsorization) versus 1–99% winsorized.
Table A3. Robustness of the serial-model outcome equation (core specification used in Models B/C): baseline (no winsorization) versus 1–99% winsorized.
PredictorBaseline Outcome Model
(Listwise, No Winsorization)		Winsorized Outcome Model
(1–99%; Base Specification)
β (Std.)SE (HC3)pβ (Std.)SE (HC3)p
Religiosity (X)0.0710.0540.1880.0760.0550.166
Presence of meaning in life (M1)−0.2520.0800.002−0.2450.0760.001
Searching for meaning in life (M2)0.0320.0530.5530.0280.0530.604
Perceived social support (M3)−0.0820.0640.197−0.0840.0630.182
Family functioning (M4)−0.2810.078<0.001−0.2920.077<0.001
0.2870.291
Table A4. Variance inflation factors (VIF)—Collinearity diagnostics (winsorized base model).
Table A4. Variance inflation factors (VIF)—Collinearity diagnostics (winsorized base model).
PredictorVIF
Religiosity 1.35
Presence of meaning in life 1.51
Family functioning 1.43
Covariate
Searching for meaning in life 1.13
Perceived social support 1.44
Age 1.23
Number of children 1.07
Table A5. Covariates diagnostics.
Table A5. Covariates diagnostics.
CovariateFull Winsorised Base
Specification		After Excluding High-Leverage
Observations (Cook’s D > 4/n)
β (Std.)SE (HC3)pβ (Std.)SE (HC3)p
Age−0.0450.0530.395−0.0330.0440.460
Number of children0.0070.0490.890−0.0150.0470.753
Sex (higher vs. reference)−0.0640.0470.174−0.0320.0370.396
Education level (dummy-coded set;
ref = Bachelor/engineering degree)		——0.013——0.007
Marital/relationship status (dummy-coded set; ref = Married)——0.008——0.028
Note. p-values for the dummy-coded sets (education level; marital/relationship status) are omnibus Wald tests based on HC3 robust covariance.
Religions 17 00147 g0a3
Figure A3. Systemic sequence—family-first serial model (Model C).
Figure A3. Systemic sequence—family-first serial model (Model C).
Religions 17 00147 g0a3
Table A6. (A). Model C outcome (Panel A) and indirect effects (Panel B). (B). Mediator equations underpinning Model B (Figure 3). (C). Mediator equations underpinning Model C (Figure A3).
Table A6. (A). Model C outcome (Panel A) and indirect effects (Panel B). (B). Mediator equations underpinning Model B (Figure 3). (C). Mediator equations underpinning Model C (Figure A3).
A
Panel A. Paths to the outcome (OLS + HC3; standardized β)
VariableB (β stand.)SE (HC3)tp
Religiosity0.0760.0551.3860.166
Family functioning−0.2920.077−3.774<0.001
Presence of meaning in life−0.2450.076−3.2120.001
Perceived social support−0.0840.063−1.3350.182
Searching for meaning in life0.0280.0530.5190.604
R2 ≈ 0.29, 95% bootstrap CI [0.24, 0.42]
Panel B. Indirect effects (percentile bootstrap, two-sided 95% CI, B = 1000)
Indirect pathIndirect effect (β)95% CI (lower)95% CI (upper)
X → Family functioning → Y−0.068−0.123−0.027
X → Presence of meaning in life → Y−0.075−0.134−0.031
X → Family functioning → Presence of meaning in life → Y−0.018−0.038−0.006
Total indirect effect−0.161−0.250−0.095
B
Outcome equation
(dependent variable)		PredictorStandardized βHow obtained
Panel A. Predicting presence of meaning in life (M1)
M1 = Presence of meaning in lifeReligiosity (X)0.382Direct estimate from mediator equation (OLS + HC3)
Panel B. Predicting family functioning (M4)
M4 = Family functioningReligiosity (X)0.121Direct estimate from mediator equation (OLS + HC3)
M4 = Family functioningPresence of meaning in life (M1)0.292Direct estimate from mediator equation (OLS + HC3)
C
Outcome equation (dependent variable)PredictorStandar-dized βHow obtained
Panel A. Predicting family functioning (M4)
M4 = Family functioningReligiosity (X)0.232Direct estimate from mediator equation (OLS + HC3)
Panel B. Predicting presence of meaning in life (M1)
M1 = Presence of meaning in lifeReligiosity (X)0.305Direct estimate from mediator equation (OLS + HC3)
M1 = Presence of meaning in lifeFamily functioning (M4)0.322Direct estimate from mediator equation (OLS + HC3)
Table A7. Compact comparison of the religiosity–burnout association: zero-order r, total effect (c), and direct effects (c′) across model specifications.
Table A7. Compact comparison of the religiosity–burnout association: zero-order r, total effect (c), and direct effects (c′) across model specifications.
Estimate TypeSpecification/Where ReportedEffect Size (Standardized Unless Noted)SE (HC3)tp
Zero-order associationPearson’s r (raw summed scores; Table 1)r = −0.10——0.115
Total effect (c)OLS total effect (winsorized z; covariates): Burnout ~ Religiosity + covariates−0.1040.064−1.6130.107
Direct effect (c′)Model A (parallel benchmark; Table 2)0.0760.0551.3860.166
Direct effect (c′)Model B (serial, meaning-first; Table 3)0.0760.0551.3860.166
Direct effect (c′)Model C (serial, family-first; Table A6A)0.0760.0551.3860.166
Note. This identity is shown as a descriptive consistency check for the reported effects.
Descriptive identity check. The reported effects satisfy the mediation identity c = c ′ + ∑ a b . For Model A, c ′ = 0.076   (Table 2, Panel A) and the total indirect effect is − 0.179   (Table 2, Panel B), yielding c ′ + ∑ a b = − 0.103   as a descriptive decomposition. The corresponding estimated total effect is c = − 0.104   (Appendix A, Table A7); the small difference reflects rounding. For Models B and C, the corresponding components are taken from Table 3 and Appendix A Table A6A. Panel A in Table 2 and Table 3 is intentionally identical because it reports the same outcome equation; Models B/C differ in the mediator equations and in how the indirect effect is decomposed, not in the set of predictors entered for Y.
Table A8. Sensitivity check: percentile-bootstrap indirect effects with B = 5000 (Models B and C).
Table A8. Sensitivity check: percentile-bootstrap indirect effects with B = 5000 (Models B and C).
EffectModel B (Estimate, 95% CI)Model C (Estimate, 95% CI)
Total indirect effect−0.151 [−0.234, −0.083]−0.151 [−0.234, −0.083]
Indirect via presence of meaning−0.091 [−0.160, −0.036]−0.075 [−0.135, −0.029]
Indirect via family functioning †−0.024 [−0.066, 0.010]−0.060 [−0.112, −0.020]
Serial indirect effect−0.036 [−0.063, −0.015]−0.016 [−0.035, −0.004]
Note. In the main text, specific indirect-effect point estimates are reported as products of the OLS path coefficients from the fitted models (see Table 3 and Appendix A, Table A6A). Table A8 reports percentile-bootstrap 95% confidence intervals based on B = 5000 resamples, because the family-only indirect effect in Model B is small, its CI is sensitive to bootstrap precision; we therefore describe this channel as borderline when the CI includes zero. In Table A8, the Model B indirect effect via family functioning is the only channel whose confidence interval crosses zero, so conclusions about this specific pathway should be stated as borderline rather than fully robust. The dagger (†) marks this row. All other indirect components remain negative and consistent with the main-text conclusions.
Table A9. Exploratory test of the Presence × Search interaction predicting parental burnout (HC3).
Table A9. Exploratory test of the Presence × Search interaction predicting parental burnout (HC3).
TermβSE (HC3)tp95% CI
Presence of meaning (M1)−0.2550.080−3.200.001[−0.411, −0.099]
Searching for meaning (M2)0.0480.0540.890.375[−0.058, 0.153]
Presence × Search0.0400.0630.630.529[−0.084, 0.164]

References

  1. Bethlehem, Jelke. 2010. Selection bias in web surveys. International Statistical Review 78: 161–88. [Google Scholar] [CrossRef] [Scilit]
  2. Blanchard, M. Annelise, Margaret L. Kerr, Heather Kirkorian, Rachel Barr, Yorgo Hoebeke, and Alexandre Heeren. 2025. The affective dynamics of parenting: Inertia of emotional distance characterizes severe parental burnout. Journal of Family Psychology 39: 465–77. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  3. Boreham, Ian D., and Nicola S. Schutte. 2023. The relationship between purpose in life and depression and anxiety: A meta-analysis. Journal of Clinical Psychology 79: 2736–67. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  4. Brečka, Tibor A., Radek Ptáček, Ivan Sebalo, Martin Anders, and Martina Sebalo Vňuková. 2024. Impact of religion and spirituality on the incidence of depression and mental health among young adults in the Czech Republic. Frontiers in Psychology 15: 1423730. [Google Scholar] [CrossRef] [Scilit]
  5. Buszman, Krystyna A., and Hanna Przybyła-Basista. 2017. Polska adaptacja Wielowymiarowej Skali Postrzeganego Wsparcia Społecznego MSPSS. Psychologia Społeczna 12: 123–45. [Google Scholar]
  6. Cohen, Sheldon, and Thomas Ashby Wills. 1985. Stress, social support, and the buffering hypothesis. Psychological Bulletin 98: 310–57. [Google Scholar] [CrossRef]
  7. Cowden, Richard G., Kenneth I. Pargament, Zhuo J. Chen, Edward B. Davis, Austin W. Lemke, Kevin J. Glowiak, Sandra Y. Rueger, and Everett L. Worthington, Jr. 2022. Religious/spiritual struggles and psychological distress: A test of three models in a longitudinal study of adults with chronic health conditions. Journal of Clinical Psychology 78: 544–58. [Google Scholar] [CrossRef] [Scilit]
  8. De Los Reyes, Andres, and Alan E. Kazdin. 2005. Informant discrepancies in the assessment of psychopathology: A critical review, theoretical framework, and recommendations for further study. Psychological Bulletin 131: 483–509. [Google Scholar] [CrossRef] [Scilit]
  9. Dong, Lei, Lei Yang, Siu-ming To, Ming-wai Yan, and Li Shen. 2025. The presence of meaning in parenthood and parental self-efficacy: A serial mediation model from social support to parent-child relationships. BMC Psychology 13: 108. [Google Scholar] [CrossRef] [Scilit]
  10. Enav, Yael, Yaara Iron, Inbal Kivenson Bar-On, and Miri Scharf. 2024. Protective factors against parental burnout during the COVID-19 pandemic. Scientific Reports 14: 30415. [Google Scholar] [CrossRef] [Scilit]
  11. Exline, Julie J., Kenneth I. Pargament, Joshua B. Grubbs, and Ann Marie Yali. 2014. The Religious and Spiritual Struggles Scale: Development and initial validation. Psychology of Religion and Spirituality 6: 208–22. [Google Scholar] [CrossRef] [Scilit]
  12. Favez, Nicolas, and Michel Bader. 2024. Non-Supportive Coparenting and a High Interdependent Self-Construal as Risk Factors for Parental Burnout in Mothers and Fathers in Switzerland. Marriage & Family Review 60: 295–315. [Google Scholar] [CrossRef] [Scilit]
  13. Frankl, Viktor E. 1959. Man’s Search for Meaning. Boston: Beacon Press. [Google Scholar]
  14. Frankl, Viktor E. 1988. The Will to Meaning: Foundations and Applications of Logotherapy. New York: Meridian. [Google Scholar]
  15. George, Login S., and Crystal L. Park. 2016. Meaning in life as comprehension, purpose, and mattering: Toward integration and new research questions. Review of General Psychology 20: 205–20. [Google Scholar] [CrossRef] [Scilit]
  16. Haber, Mason G., Jay L. Cohen, Todd Lucas, and Boris B. Baltes. 2007. The relationship between self-reported received and perceived social support: A meta-analytic review. American Journal of Community Psychology 39: 133–44. [Google Scholar] [CrossRef] [Scilit]
  17. Halbesleben, Jonathon R. B., Jean-Pierre Neveu, Samantha C. Paustian-Underdahl, and Mina Westman. 2014. Getting to the “COR”: Understanding the role of resources in conservation of resources theory. Journal of Management 40: 1334–64. [Google Scholar] [CrossRef] [Scilit]
  18. Hayes, Andrew F. 2022. Introduction to Mediation, Moderation, and Conditional Process Analysis, 3rd ed. New York: Guilford Press. [Google Scholar]
  19. He, Xiao-Xin, Xin-Qiang Wang, Michael F. Steger, Li-Jun Ji, Kai Jing, Ming-Fan Liu, and Bao-Juan Ye. 2023. Meaning in life and psychological distress: A meta-analysis. Journal of Research in Personality 104: 104381. [Google Scholar] [CrossRef] [Scilit]
  20. Hobfoll, Stevan E. 2011. Conservation of resource caravans and engaged settings. Journal of Occupational and Organizational Psychology 84: 116–22. [Google Scholar] [CrossRef] [Scilit]
  21. Hobfoll, Stevan E. 2012. Conservation of resources and disaster in cultural context: The caravans and passageways for resources. Psychiatry: Interpersonal and Biological Processes 75: 227–32. [Google Scholar] [CrossRef] [Scilit]
  22. Hobfoll, Stevan E., Jonathon Halbesleben, Jean-Pierre Neveu, and Mina Westman. 2018. Conservation of resources in the organizational context: The reality of resources and their consequences. Annual Review of Organizational Psychology and Organizational Behaviour 5: 103–28. [Google Scholar] [CrossRef] [Scilit]
  23. Holopainen, Annaleena, Marije L. Verhage, Carlo Schuengel, Anne Tharner, and Mirjam Oosterman. 2025. Longitudinal Maternal Stress, Social Support, and Their Associations with Parental Burnout During the Pandemic. Parenting 25: 282–303. [Google Scholar] [CrossRef] [Scilit]
  24. Huber, Stefan, and Odilo W. Huber. 2012. The Centrality of Religiosity Scale (CRS). Religions 3: 710–24. [Google Scholar] [CrossRef] [Scilit]
  25. Kaniasty, Krzysztof. 2020. Social support, interpersonal, and community dynamics following disasters caused by natural hazards. Current Opinion in Psychology 32: 105–9. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  26. Kossakowska, Marlena, Piotr Kwiatek, and Tomasz Stefaniak. 2013. Sens w życiu. Polska wersja kwestionariusza MLQ (Meaning in Life Questionnaire). Psychologia Jakości Życia 12: 111–31. [Google Scholar] [CrossRef] [Scilit]
  27. Le Vigouroux, Sarah, and Céline Scola. 2018. Differences in parental burnout: Influence of demographic factors and personality of parents and children. Frontiers in Psychology 9: 887. [Google Scholar] [CrossRef] [Scilit]
  28. Li, Jian-Bin, Kai Dou, and Yue Liang. 2021. The relationship between presence of meaning, search for meaning, and subjective well-being: A three-level meta-analysis based on the Meaning in Life Questionnaire. Journal of Happiness Studies 22: 467–89. [Google Scholar] [CrossRef] [Scilit]
  29. Lin, Gao-Xian, Amit Goldenberg, Gizem Arikan, Anna Brytek-Matera, Kamila Czepczor-Bernat, Denisse Manrique-Millones, Moïra Mikolajczak, Hannah Overbye, Isabelle Roskam, Dorota Szczygieł, and et al. 2022. Reappraisal, social support, and parental burnout. The British Journal of Clinical Psychology 61: 1089–102. [Google Scholar] [CrossRef] [Scilit]
  30. Lin, Gao-Xian, Amit Goldenberg, Gizem Arikan, Anna Brytek-Matera, Kamila Czepczor-Bernat, Denisse Manrique-Millones, Moïra Mikolajczak, Hannah Overbye, Isabelle Roskam, Dorota Szczygieł, and et al. 2023. Aiming to be perfect parents increases the risk of parental burnout, but emotional competence mitigates it. Current Psychology 42: 1362–70. [Google Scholar] [CrossRef] [Scilit]
  31. Long, J. Scott, and Laurie H. Ervin. 2000. Using Heteroscedasticity Consistent Standard Errors in the Linear Regression Model. The American Statistician 54: 217–24. [Google Scholar] [CrossRef] [Scilit]
  32. Mancini, Christopher J., Vanessa Quilliam, Christina Camilleri, and Stephen Sammut. 2023. Spirituality and negative religious coping, but not positive religious coping, differentially mediate the relationship between scrupulosity and mental health: A cross-sectional study. Journal of Affective Disorders Reports 14: 100680. [Google Scholar] [CrossRef] [Scilit]
  33. Matias, Marisa, Joyce Aguiar, Anne Marie Fontaine, Ege Akgun, Gizem Arikan, Kaisa Aunola, Elizabeth Barham, Wim Beyers, Emilie Boujut, Anna Brytek-Matera, and et al. 2023. Profiles of Parental Burnout around the Globe: Similarities and Differences across 36 Countries. Cross-Cultural Research 57: 499–538. [Google Scholar] [CrossRef] [Scilit]
  34. Maxwell, Scott E., and David A. Cole. 2007. Bias in cross-sectional analyses of longitudinal mediation. Psychological Methods 12: 23–44. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  35. Mikolajczak, Moïra, and Isabelle Roskam. 2018. A theoretical and clinical framework for parental burnout: The balance between risks and resources (BR2). Frontiers of Psychology 9: 886. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  36. Mikolajczak, Moïra, Kaisa Aunola, Matilda Sorkkila, and Isabelle Roskam. 2023. 15 years of parental burnout research: Systematic review and agenda. Current Directions in Psychological Science 32: 276–83. [Google Scholar] [CrossRef] [Scilit]
  37. Pankowski, Daniel, and Kinga Wytrychiewicz-Pankowska. 2023. Turning to religion during COVID-19 (Part II): A systematic review and meta-analysis. Journal of Religion and Health 62: 544–84. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  38. Pargament, Kenneth I. 1997. The Psychology of Religion and Coping: Theory, Research, Practice. New York: Guilford Press. [Google Scholar]
  39. Pargament, Kenneth I., Margaret Feuille, and Donna Burdzy. 2011. The Brief RCOPE: Current psychometric status of a short measure of religious coping. Religions 2: 51–76. [Google Scholar] [CrossRef] [Scilit]
  40. Park, Crystal L. 2005. Religion as a meaning-making framework in coping with life stress. Journal of Social Issues 61: 707–29. [Google Scholar] [CrossRef] [Scilit]
  41. Park, Crystal L. 2010. Making sense of the meaning literature: An integrative review of meaning making and its effects on adjustment to stressful life events. Psychological Bulletin 136: 257–301. [Google Scholar] [CrossRef] [Scilit]
  42. Park, Crystal L., and Login S. George. 2013. Assessing meaning and meaning making in the context of stressful life events: Measurement tools and approaches. The Journal of Positive Psychology 8: 483–504. [Google Scholar] [CrossRef] [Scilit]
  43. Patterson, Joän M. 2002. Integrating family resilience and family stress theory. Journal of Marriage and Family 64: 349–60. [Google Scholar] [CrossRef] [Scilit]
  44. Pereira, M. Graça, Jonas Fraga, Martim Santos, Ana Ferraz, and Margarida Vilaça. 2024. Parental burnout during COVID-19: The moderating role of anxiety and family functioning. Stress Health 40: e3411. [Google Scholar] [CrossRef] [Scilit]
  45. Podsakoff, Philip M., Scott B. MacKenzie, Jeong-Yeon Lee, and Nathan P. Podsakoff. 2003. Common method biases in behavioral research: A critical review of the literature and recommended remedies. Journal of Applied Psychology 88: 879–903. [Google Scholar] [CrossRef] [Scilit]
  46. Prati, Gabriele. 2024. Is religion beneficial for mental health? A 9-year longitudinal study. International Journal of Clinical and Health Psychology 24: 100491. [Google Scholar] [CrossRef] [Scilit]
  47. Preacher, Kristopher J., and Andrew F. Hayes. 2008. Asymptotic and resampling strategies for assessing and comparing indirect effects in multiple mediator models. Behaviour Research Methods 40: 879–91. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  48. Ren, Xiaohe, Yingying Cai, Jingyi Wang, and Ou Chen. 2024. Risk and protective factors of parental burnout: A systematic review. BMC Public Health 24: 17829. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  49. Roskam, Isabelle, and Moïra Mikolajczak. 2020. Parental burnout: Moving the focus from children to parents. New Directions for Child and Adolescent Development 174: 7–18. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  50. Roskam, Isabelle, Joyce Aguiar, Ege Akgun, Joyce Aguiar, Ege Akgun, Gizem Arikan, Mariana Artavia, Hervé Avalosse, Kaisa Aunola, Michel Bader, and et al. 2021. Parental burnout around the globe: A 42-country study. Affective Science 2: 58–79. [Google Scholar] [CrossRef] [Scilit]
  51. Roskam, Isabelle, Maria-Elena Brianda, and Moïra Mikolajczak. 2018. A step forward in the conceptualization and measurement of parental burnout: The Parental Burnout Assessment (PBA). Frontiers in Psychology 9: 758. [Google Scholar] [CrossRef] [Scilit]
  52. Rui, Jian Raymond, and Jieqiong Guo. 2022. Differentiating the stress-buffering functions of perceived versus received social support. Current Psychology 1: 13432–42. [Google Scholar] [CrossRef] [Scilit]
  53. Saadati, Nadereh, Seyed Salar Atashpour, and Laya Samiei. 2024. Psychological indicators of parental burnout in families with three or more children. Mental Health and Lifestyle Journal 2: 36–46. [Google Scholar]
  54. Shiffman, Saul, Arthur A. Stone, and Michael R. Hufford. 2008. Ecological momentary assessment. Annual Review of Clinical Psychology 4: 1–32. [Google Scholar] [CrossRef] [Scilit]
  55. Smilkstein, Gabriel. 1978. The Family APGAR: A Proposal for a Family Function Test and Its Use by Physicians. The Journal of Family Practice 6: 1231–39. [Google Scholar] [PubMed]
  56. Steger, Michael F. 2012. Making meaning in life. Psychological Inquiry 23: 381–85. [Google Scholar] [CrossRef] [Scilit]
  57. Steger, Michael F., Patricia A. Frazier, Shigehiro Oishi, and Matthew Kaler. 2006. The Meaning in Life Questionnaire: Assessing the presence of and search for meaning in life. Journal of Counseling Psychology 53: 80–93. [Google Scholar] [CrossRef] [Scilit]
  58. Szałachowski, Roman Ryszard, and Wioletta Tuszyńska-Bogucka. 2025. Psychometric evaluation of the family APGAR Scale in a Polish population: Reliability, validity, and factor structure. Quarterly Journal Fides et Ratio 63: 113–24. [Google Scholar] [CrossRef] [Scilit]
  59. Szałachowski, Roman Ryszard, Wioletta Tuszyńska-Bogucka, and Jacek Bogucki. 2025. “Let the system do its job and families handle the rest”? Protective and risk factors contributing to pandemic crisis parental burnout. Journal of Clinical Medicine 14: 617. [Google Scholar] [CrossRef] [Scilit]
  60. Szczygieł, Dorota, Malgorzata Sekulowicz, Piotr Kwiatkowski, Isabelle Roskam, and Moïra Mikolajczak. 2020. Validation of the Polish version of the Parental Burnout Assessment (PBA). New Directions for Child and Adolescent Development 174: 137–58. [Google Scholar] [CrossRef] [Scilit]
  61. Urbanowicz, Agata Maria, Rebecca Shankland, Jaynie Rance, Paul Bennett, Christophe Leys, and Aurélie Gauchet. 2023. Cognitive behavioral stress management for parents: Prevention and reduction of parental burnout. International Journal of Clinical and Health Psychology 23: 100365. [Google Scholar] [CrossRef] [Scilit]
  62. VanderWeele, Tyler J. 2017a. Religion and health: A synthesis. In Spirituality and Religion within the Culture of Medicine: From Evidence to Practice. Edited by Michael J. Balboni and John R. Peteet. vol. 357, p. 401. Available online: https://content.sph.harvard.edu/wwwhsph/sites/603/2019/01/ReligionHealthSynthesis.pdf (accessed on 20 June 2025).
  63. VanderWeele, Tyler J. 2017b. Religious communities and human flourishing. Current Directions in Psychological Science 26: 476–81. [Google Scholar] [CrossRef] [Scilit]
  64. Walsh, Froma. 2003. Family resilience: A framework for clinical practice. Family Process 42: 1–18. [Google Scholar] [CrossRef] [Scilit]
  65. Walsh, Froma. 2016a. Family resilience: A developmental systems framework. European Journal of Developmental Psychology 13: 313–24. [Google Scholar] [CrossRef] [Scilit]
  66. Walsh, Froma. 2016b. Strengthening Family Resilience, 3rd ed. New York: Guilford Press. [Google Scholar]
  67. Xu, Xiaoyan, Zahyah Hanafi, and Shun Zhang. 2024. Maternal parenting stress and parental burnout: The mediating role of marital satisfaction and the moderating role of social support. Frontiers in Public Health 12: 1431598. [Google Scholar] [CrossRef] [Scilit]
  68. Zarzycka, Beata. 2007. Skala Centralności Religijności Stefana Hubera. Roczniki Psychologiczne 10: 133–57. [Google Scholar]
  69. Zarzycka, Beata, Rafał P. Bartczuk, and Radosław Rybarski. 2020. Centrality of Religiosity Scale in Polish research: A curvilinear mechanism that explains the categories of centrality of religiosity. Religions 11: 64. [Google Scholar] [CrossRef] [Scilit]
  70. Zimet, Gregory D., Nancy W. Dahlem, Sara G. Zimet, and Gordon K. Farley. 1988. The Multidimensional Scale of Perceived Social Support. Journal of Personality Assessment 52: 30–41. [Google Scholar] [CrossRef] [Scilit]
Religions 17 00147 g001
Figure 1. Four theoretical pillars underpinning the integrated model of parental burnout.
Figure 1. Four theoretical pillars underpinning the integrated model of parental burnout.
Religions 17 00147 g001
Religions 17 00147 g002
Figure 2. Benchmark pattern—Parallel mediation (Model A).
Figure 2. Benchmark pattern—Parallel mediation (Model A).
Religions 17 00147 g002
Religions 17 00147 g003
Figure 3. Existential sequence—meaning-first serial model (Model B).
Figure 3. Existential sequence—meaning-first serial model (Model B).
Religions 17 00147 g003
Table 1. Descriptive statistics and correlations among study variables (raw summed scores).
Table 1. Descriptive statistics and correlations among study variables (raw summed scores).
VariableMSD12345
1. Religiosity (CRS)49.6118.39
2. Presence of meaning in life (MLQ_P)26.877.350.35 ***
3. Searching for meaning in life (MLQ_S)18.456.94−0.17 **−0.27 ***
4. Perceived social support (MSPSS)63.9815.960.18 ***0.42 ***0.01
5. Family functioning (Family APGAR)7.522.440.26 ***0.39 ***−0.12 *0.44 ***
6. Parental burnout (PBA)22.3122.30−0.10−0.37 ***0.14 *−0.27 ***−0.41 ***
Note. * p < 0.05, ** p < 0.01, *** p < 0.001. Descriptive statistics are based on raw summed scale scores (non-winsorized). Correlations are Pearson’s r (two-tailed) computed using pairwise complete observations.
Table 2. Model A outcome (Panel A) and indirect effects (Panel B).
Table 2. Model A outcome (Panel A) and indirect effects (Panel B).
Panel A. Paths to the Outcome (OLS + HC3; Standardized β)
VariableB (β Stand.)SE (HC3)tp
Religiosity0.0760.0551.3860.166
Presence of meaning in life −0.2450.076−3.2120.001
Searching for meaning in life0.0280.0530.5190.604
Perceived social support−0.0840.063−1.3350.182
Family functioning−0.2920.077−3.774<0.001
R2 ≈ 0.29, 95% bootstrap CI [0.24, 0.42]
Panel B. Indirect effects (percentile bootstrap, two-sided 95% CI, B = 1000)
Indirect pathIndirect effect (β)95% CI (lower)95% CI (upper)
X → Presence of meaning in life → Y−0.093−0.164−0.040
X → Searching for meaning in life → Y−0.005−0.0230.013
X → Perceived social support → Y −0.014−0.0400.006
X → Family functioning → Y−0.068−0.123−0.027
Total indirect effect−0.179−0.265−0.111
Table 3. Model B outcome (Panel A) and indirect effects (Panel B).
Table 3. Model B outcome (Panel A) and indirect effects (Panel B).
Panel A. Paths to the Outcome (OLS + HC3; Standardized β)
VariableB (β Stand.)SE (HC3)tp
Religiosity0.0760.0551.3860.166
Presence of meaning in life−0.2450.076−3.2120.001
Family functioning−0.2920.077−3.774<0.001
Perceived social support−0.0840.063−1.3350.182
Searching for meaning in life0.0280.0530.5190.604
R2 ≈ 0.29, 95% bootstrap CI [0.24, 0.42]
Panel B. Indirect effects (percentile bootstrap, two-sided 95% CI, B = 1000)
Indirect pathIndirect effect (β)95% CI (lower)95% CI (upper)
X → Presence of meaning in life → Y−0.093−0.164−0.040
X → Family functioning → Y †−0.035−0.083−0.003
X → Presence of meaning in life → Family functioning → Y−0.032−0.063−0.013
Total indirect effect−0.161−0.250−0.095
Note. The dagger (†) marks this row. Borderline/fragile: the CI excludes zero at B = 1000 but includes zero at B = 5000 (Appendix A, Table A8).
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Tuszyńska-Bogucka, W.; Szałachowski, R.R. “Caravans of Care”: Religiosity, Meaning in Life, and Family Functioning in Parental Burnout. Religions 2026, 17, 147. https://doi.org/10.3390/rel17020147

AMA Style

Tuszyńska-Bogucka W, Szałachowski RR. “Caravans of Care”: Religiosity, Meaning in Life, and Family Functioning in Parental Burnout. Religions. 2026; 17(2):147. https://doi.org/10.3390/rel17020147

Chicago/Turabian Style

Tuszyńska-Bogucka, Wioletta, and Roman Ryszard Szałachowski. 2026. "“Caravans of Care”: Religiosity, Meaning in Life, and Family Functioning in Parental Burnout" Religions 17, no. 2: 147. https://doi.org/10.3390/rel17020147

APA Style

Tuszyńska-Bogucka, W., & Szałachowski, R. R. (2026). “Caravans of Care”: Religiosity, Meaning in Life, and Family Functioning in Parental Burnout. Religions, 17(2), 147. https://doi.org/10.3390/rel17020147

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop