Table 1. Model assessment criterions... | Emerald Publishing

Table 1.

Model assessment criterions

Assessment of measurement model (Outer model)
Criterion	Details	Reference
Internal consistency (composite reliability)	The reliability must be >0.7	(Fornell, 1994; Hair et al., 2021)
Indicator reliability	The outer loading for the indicator must be >0.7
	According to Hair et al. (2021), outer loading which holds a value between 0.4 and 0.7 should be considered for removal if the deletion process must lead to an increase in the AVE and composite reliability
	Indicator with outer loading below 0.4 should be eliminated
Convergent validity	Individual item reliability (>0.70)
Convergent validity	Average variance extracted (>0.50)
Discriminate validity	Cross loading: the outer loading to which the factor belongs to a construct is higher than if it is placed in another construct
Discriminate validity	Fornell-Larcker criterion stated that the square rote of AVE must be higher for the construct to which the factor belongs to comparing with other constructs
Assessment of structural model (Inner model)
The coefficient of determination (R²)	(Chin, 1998) recommended the R² value of 0.67, 0.33 and 0.19 and measured as substantial, moderate and weak, respectively	(Hair et al., 2021; Henseler et al., 2016)
Effect size (f²)	Less than 0.02: no effect	(Cohen, 2013)
	0.02–0.15: small effect
	0.15–0.35: medium effect
	More than 0.35: large effects
Predictive relevance (Q²)	Fornell and Cha suggested the value of cv-red value is more than 0 then the model shows predictive relevance while if the value is less than 0 then the indication of the model lack predictive relevance	(Fornell, 1994)
The goodness of fit of the model (GoF)	(Chin, 1998) suggested 0.10 as small GoF, 0.25 as medium GoF and 0.36 as large GoF	(Cohen, 2013)

Assessment of measurement model (Outer model)
Criterion	Details	Reference
Internal consistency (composite reliability)	The reliability must be >0.7	(Fornell, 1994; Hair et al., 2021)
Indicator reliability	The outer loading for the indicator must be >0.7
	According to Hair et al. (2021), outer loading which holds a value between 0.4 and 0.7 should be considered for removal if the deletion process must lead to an increase in the AVE and composite reliability
	Indicator with outer loading below 0.4 should be eliminated
Convergent validity	Individual item reliability (>0.70)
Convergent validity	Average variance extracted (>0.50)
Discriminate validity	Cross loading: the outer loading to which the factor belongs to a construct is higher than if it is placed in another construct
Discriminate validity	Fornell-Larcker criterion stated that the square rote of AVE must be higher for the construct to which the factor belongs to comparing with other constructs
Assessment of structural model (Inner model)
The coefficient of determination (R²)	(Chin, 1998) recommended the R² value of 0.67, 0.33 and 0.19 and measured as substantial, moderate and weak, respectively	(Hair et al., 2021; Henseler et al., 2016)
Effect size (f²)	Less than 0.02: no effect	(Cohen, 2013)
	0.02–0.15: small effect
	0.15–0.35: medium effect
	More than 0.35: large effects
Predictive relevance (Q²)	Fornell and Cha suggested the value of cv-red value is more than 0 then the model shows predictive relevance while if the value is less than 0 then the indication of the model lack predictive relevance	(Fornell, 1994)
The goodness of fit of the model (GoF)	(Chin, 1998) suggested 0.10 as small GoF, 0.25 as medium GoF and 0.36 as large GoF	(Cohen, 2013)