Table 5.

Summary of paths coefficients, predictive accuracy, and validity measure.

Panel A: Paths coefficients of the research hypotheses
Relationship	Hypo.	β	Std. err.	T-value	P-value	Decision
CL -> ESG	${\mathrm{H}}_1$	0.059	0.024	2.407	0.016	Supported**
CL -> FD	${\mathrm{H}}_2$	−0.284	0.022	−12.706	0.000	Supported***
ESG -> FD	${\mathrm{H}}_3$	−0.052	0.017	−2.985	0.003	Supported***
CL -> ESG -> FD	${\mathrm{H}}_4$	−0.003	0.002	−1.784	0.074	Supported*
CL -> ROA -> FD	${\mathrm{H}}_5$	−0.058	0.008	−7.008	0.000	Supported***
ESG -> ROA -> FD	${\mathrm{H}}_6$	−0.008	0.005	−1.675	0.094	Supported*

Panel B: Analyses of predictive accuracy and validity measures of the study model
R-Square	R Square adjusted	Predictive accuracy (Geisser’s Q2 value)	Validity power (GoF value)
0.418	0.416	0.413	0.461

Note: *, **, and *** denote significance at the 10%, 5%, and 1% levels, respectively. This table summarizes the path coefficients, mediation effects, predictive accuracy, and validity measures of the research model using the Smart PLS 3 software. The variables include financial distress (FD), environmental, social, and governance (ESG), cost leadership (CL), firm size (SIZE), firm leverage (LEV), firm age (AGE), return on assets (ROA), coronavirus crisis (COV), and sector classification (SEC). Coefficients of determination values denote moderate determination; blindfolding (Geisser’s Q²) values denote the predictive accuracy of all endogenous latent variables; the GoF value denotes the validity power (high fitness) of the study model as long as it is greater than 0.36.