Data Analysis

Statistical software packages were used for data analysis (i.e., IBM AMOS 23 and SPSS 22). The factor structure of each measure was determined using a combination of exploratory and confirmatory factor analyses (EFA and CFA). This method ensured that all variables were accurately measuring their assigned construct. For some scales (e.g., PORPUS Quality of Life measure and the FMI), the use of CFA was not appropriate as the dimensionality of the measure had not been determined previously.

Structural equation modeling (SEM) was then used to test the proposed model. SEM is a statistical means of examining proposed relationships among hypothetical latent constructs which are indicated by observed variables, allowing for the separation of the measurement and structural components of the model. Missing data levels were less than 5% and were treated using Expectation–Maximization (EM). The normality of the data was assessed, with skewness and kurtosis at acceptable levels (i.e., skewness < 3; kurtosis < 10; skew; Chou and Bentler, 1995; Weston and Gore, 2006; Kline, 2011).

Mediation analysis tests the effect of a variable (i.e., PPTG) that accounts for the relation between a predictor variable (e.g., resilience) and an outcome variable (e.g., quality of life; Baron and Kenny, 1986). Thus, in the current study, to test for mediation, Hayes’ (2013) method of testing mediation using AMOS 23 was conducted to assess whether PTG and PPTG mediated the relationship between resilience and each outcome (i.e., anxiety, depression, and quality of life).