Skip to main content

Table 2 Confirmatory Factor Analyses with the MAFS data (n = 397)

From: Validation of the Dutch version of the Multidimensional Adolescent Functioning Scale (MAFS)

  MLR estimation (FIML) WLSMV estimationb (pairwise)
Number of free parameters Log likelihood AIC BIC Chi-square (df) CFI RMSEA
1-factor model 69 − 7135.6 14,409.1 14,684.0 440.1 (93)* 0.71 0.097
1-factor model (M)1 79 430.2 (91)* 0.72 0.097
3-factor model 72 − 7004.7 14,153.4 14,440.3 262.3 (95)* 0.86 0.067
3-factor model (M)1 82 243.3 (93)* 0.88 0.064
Bifactor model 92 − 6970.3 14,124.6 14,491.1 215.7 (89)* 0.90 0.060
Bifactor model (M)a 102 197.4 (88)* 0.91 0.056
  1. aIn these models, the residuals of the reverse-scored items are set to be freely estimated to account for a possible method effect
  2. bWLSMV bases model estimation on a pairwise estimated polychoric correlation matrix, which is less optimal than
  3. *p < 0.001
  4. FIML in case of missing values. It is merely included to obtain the CFI and RMSEA
  5. MLR Robust maximum likelihood, FIML Full information maximum likelihood, WLSMV Mean and variance adjusted weighted least squares, AIC Akaike Information Criterion, BIC Bayesian Information Criterion, CFI Comparative Fit Index, RMSEA Root Mean Square Error of Approximation