Table 2.
Simple bootstrapped (n=1000) mediation models of direct and indirect (via sleep disruption) effects of Group on pain responses.
| Effect | Estimate | Bootstrap SE |
t | 95% Bias Corrected CI |
|---|---|---|---|---|
| DV: Pain Threshold Composite | ||||
| RLS-to-PSQI (Sleep Disturbance) | 3.5 | .70 | 5.0* | |
| PSQI-to-DV | −.06 | .03 | −2.1* | |
| RLS-to-DV (Total Effect) | −.29 | .14 | −2.0* | |
| RLS-to-DV (Direct Effect) | .08 | .17 | −.45 | |
| RLS-to-DV (Indirect Effect) | −.21 | .10 | a | (LL = −.44; UL = −.04) |
| DV: Temporal Summation (49°C) | ||||
| RLS-to-PSQI (Sleep Disturbance) | 3.5 | .70 | 5.0* | |
| PSQI-to-DV | .03 | .03 | 1.0 | |
| RLS-to-DV (Total Effect) | .57 | .15 | 3.8* | |
| RLS-to-DV (Direct Effect) | .45 | .19 | 2.4* | |
| RLS-to-DV (Indirect Effect) | .11 | .10 | a | (LL = −.07; UL = .34) |
| DV: Temporal Summation (51°C) | ||||
| RLS-to-PSQI (Sleep Disturbance) | 3.5 | .70 | 5.0* | |
| PSQI-to-DV | .05 | .03 | 1.4 | |
| RLS-to-DV (Total Effect) | .57 | .15 | 3.8* | |
| RLS-to-DV (Direct Effect) | .41 | .18 | 2.2* | |
| RLS-to-DV (Indirect Effect) | .16 | .10 | a | (LL = −.01; UL = .38) |
Note. Table shows unstandardized coefficients for the indirect effect of RLS (IV) on pain outcomes (DVs specified within Table) through self-reported global sleep disturbance (PSQI; mediator). PSQI = Pittsburgh Sleep Quality Index; CI = Confidence Interval; LL = lower limit; UL = upper limit.
a
A p-value for the indirect effect is not provided because this value is depends upon a normal distribution of the indirect effect. Given that indirect effects are positively skewed, interpretation of this p-value is misleading and should thus not be used as a determinant of statistical mediation.
*
p < .05