Table 1.
Fit Indices for Compassion Models Tested in Both Validation Samples (Stages 3 and 4).
| Scale | Sample | Model | CFI | RMSEA [90% CI] | NNFI | SRMR | χ2 (df) | AIC |
|---|---|---|---|---|---|---|---|---|
| Compassion for Others | 1,242 Health care staff (Stage 3) | One-factor | .718 | .122 [.119, .126] | .685 | .089 | 3338.294 (170) | 42176.726 |
| Five-factor | .973 | .039 [.035, .043] | .968 | .028 | 466.435 (160) | 38170.026 | ||
| Five-factor hierarchicala | .972 | .039 [.035, .043] | .968 | .029 | 475.491 (165) | 38174.744 | ||
| 371 Students (Stage 4) | One-factor | .632 | .126 [.119, .132] | .589 | .107 | 1163.712 (170) | 14200.646 | |
| Five-factor | .966 | .040 [.030, .049] | .959 | .045 | 252.665 (160) | 13104.222 | ||
| Five-factor hierarchicala | .964 | .040 [.030, .049] | .959 | .047 | 261.210 (165) | 13103.945 | ||
| Compassion for the Self | 1,216 Health care staff (Stage 3) | One-factor | .638 | .142 [.139, .146] | .596 | .132 | 4360.676 (170) | 51699.527 |
| Five-factor | .947 | .056 [.052, .060] | .937 | .050 | 775.599 (160) | 46658.552 | ||
| Five-factor hierarchicala | .939 | .059 [.056, .063] | .930 | .068 | 871.920 (165) | 46772.251 | ||
| 371 Students (Stage 4) | One-factor | .580 | .156 [.149, .163] | .530 | .155 | 1703.097 (170) | 16986.098 | |
| Five-factor | .930 | .065 [.058, .073] | .917 | .069 | 413.800 (160) | 15362.973 | ||
| Five-factor hierarchicala | .925 | .067 [.059, .074] | .914 | .084 | 437.055 (165) | 15380.924 |
Note. AIC = Akaike information criterion; CFI = comparative fit index; CI = confidence interval; NNFI = nonnormed fit index; RMSEA = root mean square error of approximation; SRMR = standardized root mean square residual. Bold indices (CFI, RMSEA, NNFI, and SRMR) indicate acceptable fit according to liberal cutoff criteria when rounded up or down to two decimal places.
a
Five-factor hierarchical refers to a model in which all five factors load on an overarching compassion factor.