Skip to main content
. 2021 Jul 12;11:14351. doi: 10.1038/s41598-021-93523-y

Table 2.

Descriptive statistics results for confounding variables and baseline measures.

2700 K 6500 K Statistics
100 lx 1000 lx 100 lx 1000 lx CCT Illuminance CCT*illuminance R2
EMM (SE) EMM (SE) EMM (SE) EMM (SE) F df p F df p F df p
Sleep duration 7.67 (0.19) 7.62 (0.19) 7.98 (0.19) 8.37 (0.19) 5.81 (1,48) 0.02 1.12 (1,48) 0.30 1.86 (1,48) 0.18 0.05
Sleep quality 1.54 (0.54) 2.79 (0.54) 2.50 (0.54) 2.96 (0.54) 0.87 (1,48) 0.36 3.26 (1,48) 0.08 0.70 (1,48) 0.41 0.10
Light sensitivity 7.67 (0.40) 7.13 (0.40) 7.75 (0.40) 7.38 (0.40) 0.10 (1,48) 0.76 6.08 (1,48) 0.02 0.20 (1,48) 0.66 − 0.10
Sleepiness (KSS) 2.96 (0.25) 2.88 (0.25) 3.17 (0.25) 2.83 (0.25) 0.08 (1,48) 0.78 1.45 (1,48) 0.24 0.52 (1,48) 0.47 0.05
Happy 3.67 (0.14) 3.63 (0.14) 3.63 (0.14) 3.71 (0.14) 0.02 (1,48) 0.90 0.30 (1,48) 0.90 0.27 (1,48) 0.60 0.01
Sad 2.00 (0.17) 1.71 (0.17) 1.83 (0.17) 1.67 (0.17) 0.26 (1,48) 0.61 2.73 (1,48) 0.11 0.20 (1,48) 0.65 0.05

EMM estimated marginal means, SE standard errors.

A significant difference is indicated by bold labels.

Effect sizes (pseudo R2-values (R2pseudo)) indicates the proportion (percentage) of the reduction in the variance of residuals from the null model to the final (full) model at level two; it is worth mentioning that this measure can also have negative values71.