Amos 18

Data were analyzed using SPSS and AMOS 18.0 (IBM Corp., Armonk, NY, USA) software. A frequency analysis was performed to examine participants’ demographic characteristics. Further, the instruments used in this study were psychometrically validated using Cronbach’s α (reliability) and CFA (convergent, discriminatory, and law validity). Lastly, descriptive statistics and a path analysis were used to verify the fit of the hypothesis model to the structural relationship of each variable. The statistical significance was set at p < 0.05.

Using the convenience sampling method between October 1st and 31st in 2019, this study conducted a field survey of a sample of hotel workers who worked in four-star hotels or above in Seoul, Incheon, and Gyeonggi Provinces. We explained to the participants the details of our study and asked for permission to collect data. The participation would be voluntary and anonymous, guaranteeing confidentiality.
To perform an empirical analysis, 400 questionnaires were distributed, 350 copies of which were reclaimed. Of the reclaimed copies, 19 copies whose responses were judged to be unreliable were excluded, resulting in the use of 331 copies for empirical analysis. To test the theoretical model designed, the study used SPSS 18.0 (IBM, Armonk, NY, USA) to perform an exploratory factor analysis. It then performed a reliability analysis for each factor and conducted a confirmatory factor analysis using AMOS 18.0 (IBM, Armonk, NY, USA) to verify the conceptual independence of this study. Further, it performed a correlation analysis to verify the direction and discriminant validity of the hypotheses, and then, structural equation modeling was employed to test the hypotheses.

This study was conducted with 844 Korean adolescents through an online questionnaire survey from January 7 to January 15, 2021. We used a questionnaire implemented in a prior study to ensure reliability and validity, and the data were analyzed using SPSS 18.0 and AMOS 18.0 software (IBM Corp., Armonk, NY, USA). The data analysis methods were as follows: (1) A frequency analysis was conducted to check the general characteristics of the subjects. (2) Cronbach’s α was used to verify the reliability of the survey tool. (3) To verify the validity of the survey tool, a CFA was conducted. (4) To verify the level of participants’ recognition of each variable, descriptive statistics were extracted. (5) Independent samples t-test and one-way analyses of variance (ANOVAs) were used to verify the differences between each variable according to demographic characteristics. (6) A path analysis was conducted to verify more accurately the relationship between sports participation, IHLC, and wellness.

Data analysis was conducted using SPSS 25.0 (IBM, Chicago, IL, USA) and Amos 18.0 (IBM, Chicago, IL, USA) software. The specific analysis method was as follows. First, a frequency analysis was conducted to determine the demographic characteristics of the participants. Second, a factor analysis was performed to verify the validity of the questions, while the reliability of the measurement questions was validated using Cronbach’s α. For the factor analysis, an exploratory factor analysis (EFA) of the Varimax mode orthogonal rotation was first performed to examine the factor structure of the questions to measure the variables. Next, a confirmatory factor analysis (CFA) was conducted to confirm whether the derived factor structure was consistent with the actual empirical data. Third, structural equation modeling was utilized to analyze the structural relationships influencing each factor. The structural equation models were analyzed using a two-step approach. First, a CFA was conducted on the individual measurement models or simultaneously on the factors and variables included in both the measurement model and the theoretical model. This process confirmed the reliability in a single dimension and the validity between concepts. Second, we linked and analyzed the factors that appeared in the research model and evaluated the structural relationships.

A structural equation model (SEM) based on a cross-sectional dataset was used to analyze the factors influencing FD. SEM is a statistical technique for testing and estimating causal relations which uses a combination of statistical data and qualitative causal assumptions. SEM integrates factor analysis and path analysis, and examines the causal relationships among a set of variables within an integrated framework [56 (link)]. SEMs are increasingly used in ecological and geographical research as a multivariate analysis that can represent theoretical variables and address complex sets of hypotheses. The database was created in SPSS (IBM Corp. Armonk, NY, USA) and then imported into AMOS 18.0 (IBM Corp. Armonk, NY, USA). Various measures exist to assess the goodness-of-fit of a SEM, including chi-square (χ²), goodness of fit index (GFI), comparative fit index (CFI), normed fit index (NFI) and Akaike information criterion (AIC).

The data were analyzed using SPSS 18.0 (SPSS Inc., Chicago, IL, USA) and AMOS 18.0 (SPSS Inc., Chicago, IL, USA) software programs. Firstly, descriptive statistics were examined. Then, the two-stage approach recommended by Anderson and Gerbing [51 (link)] was used to evaluate the proposed model. Structural equation modeling (SEM) was used to examine the proposed model empirically. The goodness-of-fit statistics and commonly recommended threshold of SEM included: χ²/df < 8 (the ratio of chi-square to the degree of freedom), GFI > 0.8 (goodness-of-fit index), AGFI > 0.8 (adjusted goodness-of-fit index), RMSEA < 0.08 (root mean square error of approximation), RMR < 0.08 (root mean square residual), NFI > 0.9 (normed fit index), IFI > 0.9 (incremental index of fit), TLI > 0.9 (Tucker-Lewis index), and CFI > 0.9 (comparative fit index).
The common method variance (CMV) test, reliability, convergent validity, and discriminant validity were conducted to verify the adequacy of the measurement model for each construct. Harman’s single-factor test was used to identify the CMV. The procedure includes exploratory factor analysis (EFA) and confirmatory factor analysis (CFA). Secondly, the structural model of path relationships linking the six constructs was validated. Finally, the hypothesized model was tested and the results regarding standardized indirect and total effects were reported.