Table 2.
Goodness-of-fit indices for models of the Bergen Facebook Addiction Scale examined in confirmatory factor analysis
| Models | Samples | χ2 | p | df | Factor loadings | CFI | TLI | RMSEA | RMSEA 95% CI | SRMR |
|---|---|---|---|---|---|---|---|---|---|---|
| Model 1 (C) | N = 123 | 20.28 | .016 | 9 | .86 to .95 | .987 | .979 | .101 | .041 to .161 | .0147 |
| n = 59 | 14.67 | .100 | 9 | .79 to .93 | .983 | .972 | .104 | .000 to .197 | .0249 | |
| n = 64 | 10.30 | .281 | 9 | .88 to .97 | .996 | .994 | .058 | .000 to .160 | .0128 | |
| Model 2 (E) | N = 123 | 8.52 | .289 | 7 | .86 to .96 | .998 | .996 | .042 | .000 to .124 | .0099 |
| n = 59 | 8.20 | .414 | 8 | .78 to .94 | .999 | .999 | .021 | .000 to .156 | .0216 |
χ2, chi-square; df, degrees of freedom; CFI, comparative fit index; TLI, Tucker–Lewis index; RMSEA, root mean square error of approximation; CI, confidence interval; SRMR, standardized root mean residual; C, crude models; E, models with correlated error terms, values in boldface indicate misfit