We evaluated the measurement model usingX
2
and the
ratio ofX
2
to degrees of freedom; Root Mean Square Error
of Approximation (RMSEA); incremental fit index (IFI)
(Bollen, 1989); Tucker–Lewis Index (TLI) (Tucker & Lewis,
1973); Comparative Fit Index (CFI) (Bentler, 1990), and
Akaike information criterion (AIC) (Akaike, 1987). The
measurement model has good fit indices, as shown inTable 3. We used the fitted residual matrix and the standardized coefficients of the construct indicators to evaluate
convergent validity. None of the standardized residuals in
the fitted residual matrix were large enough (>|2.58|) to
demonstrate potential areas of model misfit (Jöreskog &
Sörbom, 1988). In addition, as indicated in Table 3, all of
the standardized coefficients are highly significant at
p< 0.001.
We evaluated the measurement model usingX2and theratio ofX2to degrees of freedom; Root Mean Square Errorof Approximation (RMSEA); incremental fit index (IFI)(Bollen, 1989); Tucker–Lewis Index (TLI) (Tucker & Lewis,1973); Comparative Fit Index (CFI) (Bentler, 1990), andAkaike information criterion (AIC) (Akaike, 1987). Themeasurement model has good fit indices, as shown inTable 3. We used the fitted residual matrix and the standardized coefficients of the construct indicators to evaluateconvergent validity. None of the standardized residuals inthe fitted residual matrix were large enough (>|2.58|) todemonstrate potential areas of model misfit (Jöreskog &Sörbom, 1988). In addition, as indicated in Table 3, all ofthe standardized coefficients are highly significant atp< 0.001.
การแปล กรุณารอสักครู่..
