10 Spirituality Studies 11-2 Fall 2025 ommended values above 0.90 (Hair et al. 2010). The overall model fit of the proposed model was found to be satisfactory with good fit indices for χ2/df = 1.84, RMSEA = 0.06, SRMR = 0.07, CFI = 0.91, IFI = 0.91 and TLI = 0. 90. Hence, the cut-off values defined in the literature were respected for the different adjustment indices considered. The results revealed that factor loadings of all the items in the model had values above the minimum acceptable value of 0.50 (Hair et al. 2010). Although factor loading over 0.70 is desirable, it is common to obtain weaker loadings (<0.70) in social science studies (Vinzi et al. 2009). Generally, items with loadings from 0.40 to 0.70 shall be considered for removal only if deletion results increase composite reliability (CR) or average variance extracted (AVE) over the recommended value (Hair et al. 2016). The effects of the item’s removal on CR and AVE were investigated rather than just eliminating items based on the factor loadings. In the current study, removing items with factor loadings below 0.70 didn’t significantly increase the CR and AVE values. Hence, no items were removed for further analysis. Reliability was assessed using Cronbach’s alpha, rho_a, and CR. The Cronbach’s alpha ranged from 0.75 to 0.93, whereas CR statistics ranged from 0.82 to 0.94, and rho_a ranged from 0.79 to 0.94 (Table 1). All the indicators have reliability statistics over the recommended threshold of 0.70 (Hair et al. 2010). Hence, construct validity is established. Additionally, following the recommendations of Bagozzi and Yi (1988), to assess the measurement model, the convergent validity and discriminant validity were evaluated. The Average Variance Extracted (AVE) values were used to examine convergent validity. When AVE values are greater than or equal to the recommended value of 0.50, items converge to measure the underlying construct, and hence, convergent validity is established (Fornell and Larcker 1981). As shown in Table 1, the AVE values for this study are acceptable. Although the AVE values for mindfulness and tourism well-being were slightly below the reference value, according to Fornell and Larcker (1981), the validity could be considered adequate if AVE is less than 0.5, but CR is higher than 0.7. Hence, the CR values for these constructs were greater than 0.7, indicating that convergent validity is not an issue. The criterion of Fornell-Larcker and the Heterotrait-Monotrait ratio values (HTMT) (Henseler et al. 2015) were used to test the model’s discriminant validity. In line with Fornell and Larcker (1981), discriminant validity is established when the square root of AVE for a construct is greater than its correlation with all other constructs. In this study, the square root of AVE for all the constructs was greater than its correlation with the other constructs (Table 2). Additionally, the HTMT was also tested. For instance, Kline (2011) suggests a threshold of 0.85 or less. The results of this study revealed no issues relating to discriminant validity; according to HTMT, all the values were well less than 0.85 (Table 2). Hence, this provides strong support for the establishment of discriminant validity. After the adequacy of the measures was confirmed, the structural model was examined.
RkJQdWJsaXNoZXIy MTUwMDU5Ng==