Table 3.
Multivariate regression results for the global sample – COVID-19 vulnerability, social protection, and interaction.
| (1) | (2) | (3) | (4) | (5) | |
|---|---|---|---|---|---|
| BRCS-O | BRCS-COVID-19 | Domestic work | Ag/paid work | Domestic or ag/paid work | |
| 1. COVID-19 vulnerability | −0.042*** | −0.038** | 0.055*** | 0.018 | 0.057*** |
| (0.016) | (0.016) | (0.014) | (0.014) | (0.014) | |
| 2. Received social protection | 0.016 | 0.043 | −0.052 | 0.045 | −0.012 |
| (0.054) | (0.052) | (0.044) | (0.038) | (0.041) | |
| 3. Vulnerable x social protection | 0.049 | 0.029 | 0.090*** | −0.022 | 0.067** |
| (0.031) | (0.032) | (0.028) | (0.027) | (0.028) | |
| Number of observations | 4,577 | 4,581 | 5,700 | 5,734 | 5,700 |
| Unstandardised outcome mean and SD | |||||
| Mean | 10.281 | 12.742 | 3.890 | 1.050 | 4.940 |
| SD | 3.000 | 3.293 | 3.511 | 2.594 | 4.264 |
Notes. This table uses data for girls and boys across all countries (Ethiopia, Jordan, and Palestine). Each panel presents coefficient estimates from a separate specification and each cell presents coefficients from a separate estimation. Standard errors are given in parentheses below each coefficient estimate. Outcomes of interest are labelled at the top of each column, and independent variables of interest are labelled at the start of each row. All outcomes are standardised to the mean and standard deviation within country. In columns 1 and 2, the BRCS-O is the standardised sum of the adolescent's total score on the original 4-item BRCS and the BRCS-COVID-19 is the standardised sum of the adolescent's total score on the 5 COVID-19-related coping items. For Ethiopia, the BRCS measures were administered to a random subgroup of one-third of the total sample, resulting in a smaller number of respondents for these outcomes (columns 1 and 2). In column 3, the outcome is the standardised number of hours the adolescent reports spending on domestic tasks and care work. In column 4, the outcome is the standardised number of hours the adolescent reports spending on work outside the household and income-generating tasks, including time spent on agricultural work for the household, working in a family business, or engaging in paid work outside the home. The outcome in column 5 is the standardised sum of the total number of hours reported for tasks in columns 3 and 4. The mean and standard deviation of the unstandardised outcome is provided at the bottom of the table. Each regression includes controls for the adolescent's age, gender, wealth, marital status, disability status, school enrolment status (in March 2020), and a set of location and ethnicity indicators for each country (see Table 1, Panel B). Standard errors are clustered at the subkebele level in Ethiopia to account for sampling design, and at the individual level in Jordan and Palestine. *** p<.01, ** p<.05, * p<.10.