TABLE 5.
Within‐person fluctuations in sleep as predictor of negative mood the next day
| Dependent variable: negative mood | ||||
|---|---|---|---|---|
| Self‐reported sleep minutes predicting negative mood | Self‐reported sleep/wake problems predicting negative mood | Objective sleep minutes Predicting negative mood | Objective sleep efficiency predicting negative mood | |
| Fixed effects | ||||
| b (SE), p | b (SE), p | b (SE), p | b (SE), p | |
| Intercept | 1.90 (0.04), p < .001 | 1.88 (0.03), p < .001 | 1.90 (0.09), p < .001 | 1.91 (0.04), p < .001 |
| Time | −0.05 (0.01), p < .001 | −0.05 (0.01), p < .001 | −0.05 (0.01), p < .001 | −0.05 (0.01), p < .001 |
| Weekend vs. Weekday | −0.07 (0.03), p = .024 | −0.08 (0.03), p = .007 | −0.09 (0.03), p = .009 | −0.09 (0.03), p = .005 |
| Daily pain | 0.09 (0.05), p = .075 | 0.09 (0.05), p = .083 | 0.08 (0.05), p = .14 | 0.08 (0.05), p = .14 |
| Previous day negative mood | 0.20 (0.03), p < .001 | 0.15 (0.03), p < .001 | 0.24 (0.03), p < .001 | 0.24 (0.03), p < .001 |
| SleepPC (within‐person) | −0.0003 (0.0002), p = .049 | 0.001 (0.01), p = .79 | −0.0003 (0.0002), p = .19 | −0.001 (0.002), p = .71 |
| Averaged Sleep (between‐person) | −0.0002 (0.0003), p = .60 | 0.07 (0.01), p < .001 | −0.0001 (0.0004), p = .86 | 0.002 (0.004), p = .56 |
| Sex | −0.20 (0.05), p < .001 | −0.17 (0.05), p < .001 | −0.22 (0.05), p < .001 | −0.22 (0.05), p < .001 |
| Race/ethnicity | −0.22 (0.06), p < .001 | −0.14 (0.05), p = .013 | −0.19 (0.06), p = .02 | −0.19 (0.06), p = .002 |
| Income‐to‐needs ratio | −0.03 (0.01), p = .018 | −0.03 (0.01), p = .044 | −0.03 (0.01), p = .014 | −0.04 (0.01), p = .013 |
| Random effects (Variance estimates) | ||||
| Level 2 | ||||
| Intercept | 0.18 | 0.14 | 0.17** | 0.17** |
| Time | 0.003 | 0.002 | 0.003** | 0.003 |
| Weekend | 0.02 | 0.03* | 0.04* | 0.03* |
| Pain | 0.08 | 0.09 | 0.06 | 0.06 |
| Previous day negative mood | 0.08 | 0.07 | 0.08*** | 0.09** |
| Sleep | 0.000** | 0.001* | 0.000** | 0.0001* |
*
p < .05
**
p < .01
***
p < .001; The significant level 2 random effects that are reported as 0.000 are not in fact zero, but are so small that they round to this value.
Note: Results from models with robust standard errors reported. Bold indicates a significant within‐person or between‐person association between negative mood and sleep. PC = person‐centered. Weekend coded 0 = weekday, 1 = weekend; Pain coded 0 = no, 1 = yes; Sex coded 0 = female, 1 = male; Race/ethnicity coded 0 = White/European American, 1 = racial/ethnic minority.