The relationship between personality traits and happiness: the mediating role of emotional regulation

Most participants in this study were from Saudi Arabia, comprising 55.9% of the total sample. In terms of sex distribution, males constituted 47.0%, while individuals aged 20 years or older constituted 51.2%. Regarding marital status, the majority of the participants (95.8%) were single. Regarding education, participants were distributed across various academic years, with 44.4% in their first year.

Construct reliability was assessed using Cronbach’s alpha and composite reliability (CR), with all CR values exceeding the recommended threshold of 0.70 [23]. Cronbach’s alpha for each construct surpassed the threshold of 0.63 for each construct. Convergent validity, evaluated using Average Variance Extracted (AVE), ranged from 0.50 to 0.69. According to Fornell and Larcker (1981), these values indicate a satisfactory convergent validity [24]. The factor loadings, as presented below, were consistent with the results of Hair et al. (2006), where the minimum interpretability level for the structure ranges from 0.30 to 0.40 [25]. All factors and their respective subscales exhibited a high level of significance. Table 1 provides a comprehensive overview of the reliability and validity results, as well as the factor loadings of the items. Additionally, the variance inflation factor (VIF) ranged between 1.13 and 2.06 (Table 1).

The Fornell-Larcker criterion evaluates discriminant validity. The diagonal values (bold) indicate the square roots of AVE for each construct. Discriminant validity is deemed satisfactory when these values exceed the off-diagonal correlations. In Table 2, discriminant validity seems to be supported because the bold diagonal values are higher than their corresponding off-diagonal correlations.

The Heterotrait-Monotrait Ratio of Correlations (HTMT) was used to evaluate discriminant validity. Values exceeding one indicate potential issues with discriminant validity. In the table, all HTMT values are below 1, which indicates satisfactory discriminant validity. These constructs exhibited an appropriate level of differentiation, thereby supporting the distinctiveness of the measured traits (Table 3).

The model fit statistics in Table 4 provide insight into the adequacy of the proposed model. A Standardized Root Mean Square Residual (SRMR) of 0.070 suggests a good fit. The Unweighted Least Squares discrepancy (d_ULS) and Bentler’s Comparative Fit Index (d_G) values at 3.030 and 0.590, respectively, indicate an acceptable fit. However, the high chi-square value of 2507.470 raises concerns and suggests a potential lack-of-fit. It is important to interpret this cautiously, in conjunction with other fit indices. Notably, the Normed Fit Index (NFI) of 0.700 falls below the commonly accepted threshold of 0.90, suggesting a possible need for model refinement. Further examination and potential adjustments of the model may enhance its overall fit and validity.

Table 5 presents the values of R-square (R²) and Q Square (Q²). In terms of happiness, 41% of the variance was explained (R² = 0.41) and it had a predictive relevance of 20% (Q² = 0.20). Reappraisal had an explained variance of 27% (R² = 0.27) and a predictive relevance of 14% (Q² = 0.14). For Suppression, the explained variance was 17% (R² = 0.17) and the predictive relevance was 10% (Q² = 0.10). These values demonstrate the model’s ability to explain and predict variance in the dependent variables.

Table 6 presents the essential path coefficients and results of the hypothesis testing in the structural equation model. Notably, Agreeableness significantly enhanced happiness (β = 0.10, p < 0.001), demonstrating substantial positive effects on reappraisal (β = 0.24, p < 0.001) and suppression (β = 0.18, p < 0.001). Extraversion strongly influenced happiness (β = 0.39, p < 0.001), with a positive impact on reappraisal (β = 0.14, p < 0.001) and a negative effect on suppression (β = -0.21, p < 0.001). Neuroticism significantly impeded happiness (β = -0.25, p < 0.001) and reappraisal (β = -0.26, p < 0.001), while having a detrimental effect on suppression (β = -0.33, p < 0.001). Openness had a noteworthy positive effect on suppression (β = 0.11, p = 0.01). Consciousness demonstrated a modest, yet significant, positive impact on happiness (β = 0.08, p = 0.04). However, its influence on reappraisal (β = 0.05, p = 0.23) and suppression (β = 0.06, p = 0.22) was not statistically significant. Notably, there was a negative correlation between extraversion and suppression (β = -0.21, p < 0.001), indicating that nurses who were more extraverted tended to engage in less emotional suppression. Likewise, neuroticism negatively affected both reappraisal (β = -0.26, p < 0.001) and suppression (β = -0.33, p < 0.001), suggesting that neurotic nurses are less likely to utilize these coping strategies. This overview outlines the relationships between personality traits - such as agreeableness, extraversion, neuroticism, consciousness and openness- and their contributions to happiness and emotional regulation in the model.

Table 7 presents a summary of the total effects observed in the model, offering valuable insights into the strength and significance of these relationships. These findings highlighted several noteworthy results. First, agreeableness exhibited significant positive effects on reappraisal (β = 0.240, p < 0.001), happiness (β = 0.110, p < 0.001), and suppression (β = 0.180, p < 0.001). Second, extraversion demonstrated substantial positive effects on reappraisal (β = 0.140, p < 0.001) and happiness (β = 0.410, p < 0.001) while exhibiting a negative impact on suppression (β = -0.210, p < 0.001). Third, neuroticism had negative effects on reappraisal (β = -0.260, p < 0.001), happiness (β = -0.250, p < 0.001), and suppression (β = -0.330, p < 0.001). Finally, openness had a significant positive effect on suppression (β = 0.110, p = 0.010). These outcomes enhance our understanding of the direction and strength of the relationships within the model.

Table 8 presents the results of the mediation analysis, providing details of the direct, indirect, and total effects. In terms of agreeableness, there was a significant positive direct effect on happiness (β = 0.100, p < 0.001), but the indirect effect was not significant (β = 0.010, p = 0.590), resulting in a significant total effect of 0.110 (p < 0.001). Consciousness demonstrated a positive direct effect on happiness (β = 0.080, p = 0.040) with no significant indirect effect, leading to a significant total effect of 0.080 (p = 0.040). Extraversion displayed a substantial positive direct effect on happiness (β = 0.390, p < 0.001) with a small but significant indirect effect (β = 0.020, p = 0.020), resulting in a significant total effect of 0.410 (p < 0.001). Neuroticism exhibited a significant negative direct effect on happiness (β = -0.250, p < 0.001), with no significant indirect effect, leading to a significant total effect of -0.250 (p < 0.001). Openness demonstrated a nonsignificant direct effect on happiness (β = 0.010, p = 0.750) and no significant indirect effect, resulting in a nonsignificant total effect of 0.010 (p = 0.820). These findings offer valuable insights into the complex relationship between personality traits and happiness, presenting various patterns of direct and indirect effects.

Multiple-group analysis revealed no significant differences between Jordan and Saudi Arabia in the relationship between personality traits and happiness, reappraisal, and suppression. No difference in the path coefficients was statistically significant (p > 0.05), indicating similar patterns in these associations across the two countries (Table 9).