5.1. Factor AnalysisTo create the q

5.1. Factor AnalysisTo create the questionnaire, 25 items were analyzed by exploratory factor analysis and the Varimax rotation method. The factor analysis (shown in Appendix A) revealed that commonality exceeded0.683, and factor loadings revealed eight factors, with no multiple loading items for only one factor of0.6 or more. The results of the exploratory factor analysis also revealed that the Kaiser–Meyer–Olkin (KMO) value for the sample was 0.905, which confirms that the dataset is valid for the purposes of factor analysis. In addition, the sphere formation test value for the sample was x2 = 8022.512 (df = 300, p < 0.001), and the cumulative total variance of the factors was 90.39%, which also indicates suitability for factor analysis. Reliability of the eight identified factors was confirmed by Cronbach’s α coefficient, which was higher than 0.832, thus indicating high credibility [123].5.2. ValidationValidity and feasibility of the measurement modelThe validity and appropriateness of the measurement model were determined before testing the hypotheses of this study. First, as shown in Table 2, the average variance extracted (AVE) exceeded 0.888, which indicates convergent validity [124]. Composite reliability (CR), which is an index that measures the feasibility of the measurement model, exceeds 0.900, indicating reliability. The Cronbach’s α value resulting from the PLS algorithm was 0.833. Except for vividness, the internal consistency of questionnaire items was indicated by a high reliability value of 0.9 or higher on all other factors [124,125]. Furthermore, the communality value, measuring the quality of the measurement model, exceeded 0.888, indicating the suitability of the measurement model. The size of goodness-of-fit in the PLS path model is also regarded as large if the value is 0.36 or larger; the value for location-based AR was 0.536, thus showing high goodness-of-fit [126].