Analysis | Description | Rationale |
1 | Multivariate Normality Test | To test for the multivariate normality assumption with Mardia’s multivariate kurtosis and skewness, Henze–Zirkler’s consistent test, Doornik–Hansen omnibus test, and Energy test. |
2 | Detecting outliers | To detect outliers with Mahalanobis distance. |
3 | CFA on each study measures | To confirm the factor structure of MBI-HSS and CD-RISC 10 in this special Greek population (OTs), ensuring no misspecifications in the SEM model. |
4 | Cronbach’s alpha & | To calculate the internal consistency reliability and the model-based reliability of MBI-HSS and CD-RISC 10, before the SEM measurement model with ωt coefficient |
5 | Spearman rho Correlations, and means | Descriptive statistics and burnout scores. |
6 | Test the SEM measurement model and model-based reliability | To evaluate the measurement model fit with a CFA and to evaluate reliability of the measurement variables and the reliability latent variables with ωt coefficient |
7 | Test the full SEM model fit | To evaluate if the structural model fit is adequate. |
8 | A priori & post hoc power analysis of the full SEM model | To calculate the required sample for achieving a power of 80% to reject a wrong model. An alpha level of 0.05 was assumed with an RMSEA misspecification of 0.05 |
9 | Primary Hypotheses testing (Primary hypotheses H1 - H5) | To test the hypothesized relationships between 4 latent variables with 4 direct associations, and 1 mediation. No covariates were used. |
10 | Latent Profile Analysis (LPA) (Secondary Hypotheses H1 - H2) | Use the scores of the latent variables of the SEM model to profile the sample and check if the profiles that emerged confirm the hypotheses tested with the SEM model. |