SOME STEPS IN PARTIAL LEAST SQUARE METHOD
Criterion |
Description |
|
Overall model |
Test of model fit (estimated model) |
Standardized Root Mean Square Residual (SRMR) < 0.08 |
Measurement model |
Confirmatory composite factor analysis (saturated model) |
SRMR < 0.08 |
Composite reliability (CR) |
CR is a measurement of internal consistency and must not be lower than 0.6 |
|
Indicator reliability |
Absolute standardized outer loadings should be higher than 0.7 |
|
The average variance extracted (AVE) |
AVE is the measurement of the amount of variance that is captured by construct about the amount of variance due to measurement error. The AVE should be higher than 0.5 |
|
Fornell – Larcker criterion |
To ensure the discriminant validity, the AVE of each latent variable should be higher than the squared correlations with all other latent variables. |
|
Cross-loadings (HTMT) criterion |
If an indicator has a higher correlation with another latent variable than with its respective latent variable, the appropriateness of the model should be reconsidered. |
|
Structural model |
R2 of endogenous latent variables |
R2 values of 0.67, 0.33, or 0.19 are considered as substantial, moderate, or weak (Chin, 2010) |
Estimates for path coefficients |
Estimates for path coefficients should be assessed in terms of the sign, magnitude, and significance |
|
Effect size (f2) |
The effect size (f2) is the measurement of the impact of each predictor construct on the dependent construct. Values of 0.02, 0.15, and 0.35 can be viewed as a weak, medium, or large effect at the structural level |
|
Prediction relevance (Q2) |
This measurement is calculated by the blindfolding procedure. Q2 above zero gives evidence that the observed values are well reconstructed and that model has predictive relevance. |
|
Multi-group Analysis |
|
A procedure to verify the significant differences in the estimation of parameters for defined groups (Matthews, 2017). There is a significant difference if p-value < 0.05 |
» Tin mới nhất:
» Các tin khác: