Spirituality Studies 11-2 Fall 2025 9 Vivien Iacob et al. consisted of two sections. The first section included the sociodemographic questionnaire. The second section included previously validated measurement items adapted from existing instruments. Mindfulness was measured by adapting 13 items from the Freiburg Mindfulness Inventory short form (FMI; Walach et al. 2006). This scale evaluates essential aspects of mindfulness and is considered one-dimensional, and the Cronbach alpha value for this study was 0.89. Spirituality was assessed using five items from the Spirituality Scale (SS; Pinto and Pais-Ribeiro 2010). This scale aims to evaluate spirituality with items focused on the meaning of life/beliefs and positive life perspective and hope. The Cronbach alpha value for this study was 0.79. Tourism well-being was examined through eight items from the Tourism Well-being Scale (TWS; Garcês et al. 2020). This scale comprises the following theoretical factors of the HOPE model: Positive Emotions, Engagement, Relationships, Meaning, Accomplishment, Creativity, Spirituality, and Optimism, and the Cronbach alpha value for this study was 0.85. Happiness was measured by using three items from the Subjective Happiness Scale (SHS; Lyubomirsky and Lepper 1999). This scale measures global subjective happiness through statements with which participants either self-rate themselves or compare themselves to others. The Cronbach alpha value for this study was 0.73. Satisfaction was assessed by adapting five items from the Satisfaction with Life Scale (SWLS; Diener et al. 1985). It assesses the subjective judgment that each individual makes about the quality of their own life. In this study, Cronbach’s alpha value was 0.92. 4.3 Data Analysis This study employed SPSS 28.0 and AMOS 24. Descriptive analyses were performed to describe the sociodemographic information and study variables (mindfulness, tourism well-being, happiness, satisfaction, and spirituality). Then, common method bias (CMB) and normality tests were performed. Then, a two-step structural equation modeling (SEM) process was employed, as Anderson and Gerbing (1988) suggested. First, the measurement model was analyzed by assessing the reliability and validity of the variables. Second, the structural model was evaluated to investigate the relationship between constructs. Finally, bootstrapping approaches were used to examine the mediating effect of spirituality. 5 Results 5.1 Test of Common Method Bias In this study, common method bias was ensured before, during, and after administering the questionnaire. Before data collection, the choice of instruments was based on scales that had already been validated. Only voluntary participants filled out the survey during data collection and were previously informed that their responses would remain anonymous. After data collection, Harman’s single factor test was used to test for any common method bias (CMB) statistically. Harman’s one-factor test was employed to address CMB, and all the constructs were subjected to Exploratory Factor Analysis (EFA). The results reflected a 29% variance, less than the recommended threshold value of 50%. Therefore, CMB was not a significant issue in this study. 5.2 Measurement Model Assessment First, the assumption of multivariate normal distribution was tested. Curran et al. (1996) proposed that skewness values smaller or equal to two (|sk|≤2) and kurtosis equal to or smaller than seven (|ku|≤7) points to the respect of the normal distribution assumption. The results of this study fulfilled the normality assumption by presenting skewness values between −0.06 and 1.09 and kurtosis values between −0.01 and −1.51. Second, the model was examined to evaluate the quality of the variables in the theoretical model (Anderson and Gerbing 1988). In the first step, confirmatory factor analysis (CFA) was conducted to assess the psychometric properties of the constructs. Measures of goodness of fit were examined considering the following adjustment indexes: (a) χ2/df considering that an acceptable fit occurs for values equal to or less than five and a very good fit emerges for values equal to or less than two (Marôco 2021); (b) Root Mean Square Error of Approximation (RMSEA), with proper adjustment values ranging between 0.05 and 0.10 (Marôco 2021); (d) Standardised Root Mean Square Residual (SRMR), acceptable adjustment values ranging between 0.05 and 0.08 (Hu and Bentler 1999); (e) Comparative Fit Index (CFI) and Tucker-Lewis Index (TLI), with excellent fit values between 0.90 and 0.95 (Marôco, 2021); and (f) Incremental Fit Index (IFI) considering the rec-
RkJQdWJsaXNoZXIy MTUwMDU5Ng==