The research used a convenience sample recruited at İzmir Kâtip Çelebi University. Participants were 17 to 55 years old, with a mean age of 23.22 years, and 65.5% were women. The sample was strongly concentrated in theology, with 223 students, followed by social and human sciences, health sciences and engineering; medicine included only one participant. Nine people did not answer the future-anxiety scale, leaving 391 complete cases for analyses involving that outcome. Because the design measured all variables at one point, it can identify statistical associations but cannot establish temporal order, causation or the direction of influence among wellbeing, moral orientation and anxiety.

Mental wellbeing was assessed with the 14-item Warwick–Edinburgh Mental WellBeing Scale, which ranges from 14 to 70, with higher scores representing greater wellbeing. Future anxiety was measured with the five-item Dark Future Scale, ranging from 0 to 30, with higher scores indicating greater anxiety. The scales showed strong internal consistency: Cronbach’s alpha values ranged from 0.88 to 0.92 and McDonald’s omega values from 0.92 to 0.94. Across the complete cases, mental wellbeing had a moderate inverse correlation with future anxiety, r = −0.30, p < 0.001. Religious moral orientation was essentially unrelated to future anxiety, r = 0.001, p = 0.977, whereas secular moral orientation had a small positive correlation with it, r = 0.16, p < 0.01.

Moral orientation was measured with a 14-item instrument containing two separately scored dimensions rather than one combined total. Religious moral orientation, comprising eight items, captures reliance on religious beliefs and transcendent sources when making moral judgments. Secular moral orientation, comprising six items, captures reliance on reason, human experience and ethical principles not necessarily grounded in religious authority. Importantly, the two scores are not opposite ends of a single scale: a person can draw on both religious and secular reference systems. The measures describe which systems people tend to use in moral judgments, not the quality of their moral conduct, and the study did not treat secular orientation as a direct measure of religiosity.

In the primary religious-orientation model, the coefficient for mental wellbeing was negative, b = −0.250, 95% confidence interval [−0.321, −0.178], p < 0.001, indicating that higher wellbeing scores accompanied lower future-anxiety scores. The interaction between mental wellbeing and religious moral orientation was also negative under standard normal-theory inference, b = −0.007, 95% confidence interval [−0.012, −0.002], p = 0.003. Adding this interaction explained an additional 2.0% of outcome variance, ΔR² = 0.020, corresponding to a small effect size of f² = 0.023. Descriptive slope estimates showed an inverse relationship at every examined level of religious orientation, but it appeared stronger at high orientation than at low orientation.

Those apparent differences did not survive every check. When the researchers used 10,000 percentile-bootstrap resamples, the 95% confidence interval for the mental wellbeing by religious-orientation interaction widened to [−0.0136, 0.0002], narrowly including zero. An analysis using HC3 heteroscedasticity-consistent standard errors produced a similar borderline result: b = −0.0072, p = 0.0503, with a confidence interval of [−0.0145, 0.0000]. The authors say the reason for the divergence between standard and robust inference could not be determined from the interval comparison alone. They consequently characterize the religious interaction as estimation-sensitive and provisional, not robustly confirmed.

The exploratory secular-orientation model also showed the central inverse association between mental wellbeing and future anxiety, b = −0.225, p < 0.001. However, its interaction term did not reach the conventional threshold, b = −0.006, 95% confidence interval [−0.012, 0.001], p = 0.078, and accounted for only 0.7% additional variance, ΔR² = 0.007, f² = 0.008. The secular interaction was also non-significant with bootstrap inference and with HC3 standard errors, p = 0.224. A joint model containing both interactions produced similar results: the religious term was significant with conventional and HC3 inference, while the secular term was not. Crucially, neither the contrast between the unstandardized coefficients, p = 0.877, nor the contrast between fully standardized coefficients, p = 0.553, showed that religious and secular orientations moderated the wellbeing–anxiety relationship differently.

Post hoc analyses adjusted the separate models for gender, age, perceived economic status and faculty. With conventional standard errors, both interaction terms crossed the conventional significance threshold, but neither remained significant with HC3 inference: religious orientation had p = 0.084 and secular orientation p = 0.155. These models were supplementary rather than preregistered, and sparse faculty and grade categories reduced their stability. A post hoc power analysis indicated that the complete-case sample provided approximately 80% power to detect an interaction of f² = 0.02 at α = 0.05, but this was not an a priori sample-size calculation and does not rule out smaller effects. The cross-sectional design, single-university convenience sample, self-reports and lack of preregistration further limit generalizability and causal interpretation.

The results do not establish that religious moral frameworks reduce anxiety or make wellbeing more effective against it. One possible explanation is that a stable religious framework supplies meaning, values and interpretive guidance when people face uncertainty, but the study did not measure meaning in life, worldview coherence, identity clarity, coping or social support as mechanisms. Such explanations therefore remain hypotheses for longitudinal or experimental research. The findings also do not show that secular frameworks are irrelevant: the analysis lacked evidence for a distinct secular moderating effect, not proof that secular meaning systems lack psychological relevance. Replication with preregistered hypotheses, larger and more diverse samples, multiple institutions and cultures, robust intervals and longitudinal data is needed to determine whether the small interaction is reliable and what, if any, mechanism explains it.

Subject of Research: Whether religious and secular moral orientations moderate the association between mental wellbeing and future anxiety in university students.

Article Title: Religious and secular moral orientations as potential moderators of the association between mental wellbeing and future anxiety

Article References: Akbolat, A. (2026). Religious and secular moral orientations as potential moderators of the association between mental wellbeing and future anxiety. Frontiers in Psychology, 17, Article 1952527. https://doi.org/10.3389/fpsyg.2026.1952527

Image Credits: AI Generated

DOI: 10.3389/fpsyg.2026.1952527

Keywords: mental wellbeing, future anxiety, religious moral orientation, secular moral orientation, moderation analysis, university students, Türkiye, psychological uncertainty, cross-sectional study, Religious, secular, moral