Statistical package for social sciences for windows spss version 22

IBM Statistical Package for Social Sciences for Windows (SPSS) version 22.0 (SPSS, Chicago, Illinois) program was used to evaluate findings in this study. Categorical variables are reported as numbers and percentages. Continuous variables were checked for the normal distribution assumption using the Shapiro–Wilk test. Kruskal–Wallis test was used for comparison of nonparametric quantitative data. Mann–Whitney U test was used to compare nonparametric variables between two groups. Fisher's exact test and Continuity (Yates) Correction were used to compare qualitative data. Spearman's rho test was used to investigate the relationships between numerical variables. Significance was evaluated at p < .05.

The Statistical Package for Social Sciences for Windows (SPSS, version 22.0; Armonk, NY) and MPLUS (version 8.0; Los Angeles, CA) were jointly used to conduct the statistical analyses. Descriptive data were statistically analyzed using χ² and Mann–Whitney U tests.
Prior to establishing the physician’s behavior model, the reliability and validity of the TPB survey were determined in the psychometric testing phase by calculating Cronbach’s alpha and factor analysis. The reliability with Cronbach coefficient alpha values for each construct of SDM behavior was 0.89 (attitudes), 0.78 (subjective norms), 0.65 (perceived behavior control), and 0.83 (behavior intention) in the main study. The confirmatory factor analysis test results showed good model fit indices of the expected four-factor model based on TPB theory with TLI = 0.98, CFI = 0.99, and RMSEA = 0.078.
The structural equation model was applied to understand the associations of SDM behavior and its determinants based on the TPB theoretical framework.
Since the responses were ordinal variables, means and variance adjusted weighted least squares extraction estimation was applied to examine the direct and indirect associations among the study variables.