Table 4.
Fixed effects estimates (top) and variance-covariance estimates (bottom) for predictive multilevel models of depressive symptoms (PHQ-9) from T1 to T2
| Model 1 | Model 2 | Model 3 | Model 4 | ||
|---|---|---|---|---|---|
| Est. | Est. | Est. | Est. | t (df) | |
| Fixed effects | |||||
| Intercept | 7.95* (0.06) | 7.84* (0.05) | 7.83* (0.05) | 7.89* (0.05) | 168.46 (12 488.75) |
| Time | −0.88* (0.06) | −0.90* (0.06) | −0.99* (0.07) | −0.78* (0.06) | −12.01 (6501.68) |
| Age | −0.11* (0.01) | −0.04* (0.01) | −0.04* (0.01) | −0.03* (0.01) | −9.79 (12 488.75) |
| Gender | −1.57* (0.13) | −1.07* (0.11) | −1.07* (0.11) | −1.20* (0.10) | −11.74 (12 488.75) |
| Education | −0.60* (0.05) | −0.45* (0.05) | −0.45* (0.05) | −0.44* (0.05) | −9.54 (12 488.75) |
| Time × Age | 0.02* (0.01) | 0.01* (0.01) | 0.01* (0.01) | 0.02* (0.01) | 5.24 (6140.00) |
| Time × Gender | 0.72* (0.14) | 0.61* (0.13) | 0.60* (0.14) | 0.65* (0.13) | 4.91 (6273.98) |
| Time × Education | 0.04 (0.08) | 0.06 (0.06) | 0.05 (0.06) | 0.08 (0.06) | 1.36 (6317.98) |
| DERS T1 | 0.68* (0.01) | 0.68* (0.01) | 0.52* (0.02) | 26.14 (12 488.75) | |
| IIP T1 | 0.10* (0.01) | 0.11* (0.01) | 0.10* (0.01) | 9.67 (12 488.75) | |
| Time × DERS T1 | −0.13* (0.02) | −0.15* (0.02) | −0.37* (0.02) | −19.00 (5267.96) | |
| Time × IIP T1 | 0.03* (0.01) | 0.03* (0.01) | −0.10* (0.01) | −9.84 (5148.50) | |
| Time × DERS T1 × IIP T1 | 0.01* (0.01) | 0.01 (0.01) | 1.25 (8210.69) | ||
| DERS T2 | 0.19* (0.02) | 9.54 (12 488.75) | |||
| IIP T2 | −0.01 (0.01) | −0.57 (12 488.75) | |||
| Time × DERS T2 | 0.42* (0.02) | 20.96 (5267.58) | |||
| Time × IIP T2 | 0.11* (0.01) | 10.82 (5102.91) | |||
| Est. | Est. | Est. | Est. | Wald Z | |
| Variance intercept | 20.54* (0.39) | 9.99* (0.25) | 9.97* (0.25) | 9.62* (0.24) | 40.80 |
| Residual | 8.40* (0.17) | 8.27* (0.16) | 8.27* (0.16) | 7.34* (0.15) | 49.75 |
| AIC | 89 344.50 | 84 158.27 | 84 149.58 | 82 914.32 | – |
DERS, difficulties in emotion regulation; IIP, overall interpersonal problems; AIC, Akaike's information criterion.
Note. Standard errors and degrees of freedom are given in parenthesis. Estimations were performed by the method of maximum likelihood (ML) and with a homoscedastic error covariance structure. *p < 0.01. T1 = a period of 1 week (31st March to 7th April 2020) starting nearly 3 weeks after the implementation of strict social distancing protocols in Norway (12th March 2020). T2 = a period of three weeks (22nd June to 13th July 2020) starting 1 week after the strict social distancing protocols had been discontinued (15th June 2020). Degrees of freedom (df), t values, and Wald Z are given only for the final model.