Table 5.
Model fit Exploratory Factor Analysis for WURSS-44 using 3 to 10 dimensions
| Dimensions | Chi-square | df | χ2/df | CFI | TLI | RMSEA | SRMR |
| 3 | 6902.7 | 738 | 9.35 | .974 | .970 | .192 | .076 |
| 4 | 5114.1 | 699 | 7.31 | .982 | .977 | .167 | .061 |
| 5 | 3946.8 | 661 | 5.97 | .986 | .982 | .148 | .050 |
| 6 | 2922.0 | 624 | 4.68 | .990 | .987 | .127 | .041 |
| 7r | 2054.4 | 489 | 4.20 | .993 | .989 | .119 | .034 |
| 8 | 1785.5 | 553 | 3.22 | .995 | .992 | .099 | .029 |
| 8r | 1625.9 | 457 | 3.55 | .995 | .992 | .106 | .028 |
| 9 | 1430.5 | 519 | 2.75 | .996 | .994 | .088 | .024 |
| 10 | 1165.6 | 486 | 2.39 | .997 | .995 | .078 | .021 |
7r and 8r = Dimensions restricted to exclude headache, loss of appetite, and sleep well,
items which did not
CFI = Comparative Fit Index
TLI = Tuker-Lewis Index
RMSEA = Root Mean Square Error of Approximation
SRMR = Standardized Root Mean Square Residual
Hu and Bentler (1999)[58] suggest the following cut off values for good fit, CFI > .95, TLI > .95, RMSEA < .06, and SRMR < .08