Table 4 PLS-SEM decision criteria
From: Longer food miles of beef under self-insufficiency and halal requirements
Evaluation criteria | Acceptance value |
|---|---|
Models of reflective measurement | |
• Reflective indicator loadings |  ≥ 0.30 is reliable if the sample size is ≥ 350 |
• Internal consistency reliability | Value of Cronbach's alpha, composite reliability (RhoA and RhoC): • Minimum: 0.70 (0.60 for exploratory research) • Maximum: 0.95 |
• Discriminant validity | Heterotrait–monotrait (HTMT) ratio: < 0.85 |
• Convergent validity | Average variance extracted (AVE) ≥ 0.50 |
Moderation analysis | |
Criteria | The same criteria as the structural model |
Structural model | |
• Collinearity | Critical issue (variance inflation factor or VIF: ≥ 5), non-critical issue (VIF: 3–5), and non-problematic issue (VIF: < 3) |
• Significance and relevance of path coefficients | Perform bootstrapping to evaluate the significance of path coefficients under the t-value. t-value is greater than 2.58 (α = 0.01), 1.96 (α = 0.05), and 1.65 (α = 0.10) under a two-tailed test |
• R2 value | Substantial (0.75), moderate (0.50), and weak (0.25) |
Mediation analysis | |
• Complementary mediation | Indirect and direct effects are significant, and the point is in the same direction.a |
• Competitive mediation | Indirect and direct effects are significant, and the point is in the opposite direction.b |
• Indirect-only mediation | The indirect effect is significant but not for the direct |
• Direct-only non-mediation | The direct effect is significant, but not for the indirect |
• No-effect non-mediation | Indirect and direct effects are insignificant |