Table 3.
Multilevel binary logistic regression model of shadow education duration (using appropriate education as the reference category).
| M5 | M6 | M7 | M8 | |
|---|---|---|---|---|
| Variables | Insufficient Education | Excessive Education | Insufficient Education | Excessive Education |
| Control variable | YES | YES | YES | YES |
| Family income | 0.496 *** | 0.774 *** | 0.502 *** | 0.785 *** |
| (0.024) | (0.063) | (0.025) | (0.066) | |
| Before 2018 (base group) | ||||
| 2020 | 0.648 *** | 1.935 *** | ||
| (0.057) | (0.335) | |||
| 2022 | 0.766 *** | 0.628 ** | ||
| (0.067) | (0.126) | |||
| Constant | 19,344.301 *** | 0.476 | 12,299.165 *** | 0.464 |
| (13,851.352) | (0.669) | (8946.199) | (0.671) | |
| ICC of family level | 0.564 | 0.492 | 0.573 | 0.526 |
| ICC of community level | 0.101 | 0.466 | 0.120 | 0.526 |
| Observations | 7439 | 2424 | 7439 | 2424 |
Notes: 1. Coefficients represent odds ratios; standard errors in parentheses 2. *** p < 0.01, ** p < 0.05.