TABLE 3.
Multilevel estimates for models predicting emotional exhaustion.
| Step 1 | Step 2 | Step 3 | |
| Intercepts | 4.11** | 4.11** | 4.12** |
| Level-1 | |||
| Sleep quality | −0.02 | −0.02 | |
| PA (t1) | 0.04 | 0.03 | |
| NA (t1) | −0.09 | −0.07 | |
| PA (t2) | 0.07 | 0.08 | |
| NA (t2) | 0.16† | 0.18† | |
| δpa(residual) × δna(residual) | 0.17* | ||
| Level-2 | |||
| Age | 0.01 | 0.01 | |
| Gender | 0.03 | −0.05 | |
| Education | −0.01 | 0.01 | |
| Positive affectivity | −0.38† | −0.44* | |
| Negative affectivity | 0.28 | 0.24 | |
| Variance | |||
| σ2 | 0.69 | 0.63 | 0.62 |
| T 00 | 1.21** | 1.21** | 1.18** |
| T 11 | 0.18† | 0.21† | |
| R2Level–1 | 0.09 | ||
| R2level–1 interaction | 0.02 | ||
†p < 0.1, *p < 0.05, **p < 0.01. R2Level–1 = (σ2 of Step 1–σ2 of Step 2)/σ2 of Step 1. R2level–1 interaction = (σ2 of Step 2–σ2 of Step 3)/σ2 of Step 2.