Global Journal of Management and Business Research, E: Marketing, Volume 23 Issue 1

complex cause-effect relationships in marketing and management research (Bascle, 2008; Gudergan et al., 2008). Sample size and indication dispersion standards are also lax (Hair et al., 2017; Sarstedt et al., 2020). Anderson and Gerbing's two-stage analytic technique was used to evaluate this study's hypotheses (Anderson & Gerbing, 1988). The initial measurement model test verified the instrument's reliability and validity. The second stage estimated the study's hypothesized linkages using the structural model. a) Measurement Model The reliability of the elements was tested by Cronbach’s alpha ( α ) and Composite Reliability (CR). The Alpha and CR value of the construct should be at least 0.70 or higher(Hair Jr et al., 2006). At first, the whole sample was evaluated, and items with factor loadings under 0.70 were removed (Chin, 1998). Hair et al. (2019) state that factor loadings must be at least 0.50. Table 2 shows that all items have loadings of more than 0.70, ranging from 0.892 to 0.800. This study's alpha values vary from 0.903 to 0.825, and CRs values from0.925 to 0.895. Moreover, the average variance extracted (AVE) value is higher than0.50. It means that the constructs are clarified more than half of the variance in its measures on average. Generally, an AVE value higher than 0.50 is acceptable (Hair et al., 2010). This study's results are reliable since the AVE value varies from 0.672 to 0.776. Discriminant validity examines the uniqueness of constructs (Henseler et al., 2015). The current study examined the Fornel-Larcker criterion and the Heterotrait and Monotrait (HTMT) ratio to investigate the discriminant validity. The cross-loadings of items are provided in Table 3. Discriminant validity is shown by all factor loadings greater than cross-loadings. According to Fornell and Larcker (1981), the discriminant validity of a construct may be determined by comparing the square roots of all AVE values to the relationships of the latent variable. Table 4 shows the strong discriminant validity by demonstrating that the square root of AVE for each item is greater than the sum of its correlations with the other factors. Table 5 shows that none of the HTMT scores were higher than 0.90. Although discriminant validity was adequate, multicollinearity might be an issue. Thus, a variance inflation factor (VIF) test was performed to assess multicollinearity. All constructs had VIF values between 1.906 and 3.525, much lower than the criterion of 5 (Diamantopoulos & Winklhofer, 2001). Multicollinearity had not been an issue. A standardized root mean square residual (SRMR), normed-fit index (NFI), and Chi-Square were also used to assess the overall goodness-of-fit of the structural model. Hu and Bentler (1999) state that lowers SRMR values indicates increased goodness-of-fit. The SRMR value of the current study is 0.054, which is below the recommended cutoff limit (0.08) (Hu & Bentler, 1999).According to Bentler (1990), NFI should be between 0 and 1, with higher values suggesting a good model fit. The NFI score is 0.872, which is near 1 and acceptable (Bentler, 1990). The chi-Square value is 446.045, which exceeds the cutoff point (Sarstedt et al., 2020). Because the samples were self-reported and obtained from a single source, Harman’s single-factor test was used to exclude the possibility of common method bias (CMB) (Podsakoff et al., 2012). A single factor explained 27.19% of the variance, just below the 50% criterion. Thus, the results of this investigation indicate that CMB is not a major issue. Table 2: Reliability and Validity Results Items FL α CR AVE VIF Customer expectation (EXP) 0.903 0.925 0.672 EXP1 0.843 2.54 EXP2 0.818 2.139 EXP3 0.822 2.245 EXP4 0.815 2.157 EXP5 0.800 2.013 EXP6 0.822 2.166 Perceived service quality (PSQ) 0.843 0.895 0.680 PSQ1 0.850 2.03 PSQ2 0.827 1.917 PSQ3 0.805 1.841 PSQ4 0.816 1.772 Firm Marketing Communication (FMC) 0.825 0.895 0.740 FLC1 0.839 1.824 FLC 2 0.887 2.101 FLC 3 0.854 1.775 Customer satisfaction (CS) 0.886 0.916 0.686 CS1 0.857 2.528 Customer Loyalty in the Fitness Club Industry: The Role of Club Communication, Customer Expectation and Perceived Service Quality 50 Global Journal of Management and Business Research Volume XXIII Issue I Version I Year 2023 ( )E © 2023 Global Journals