Table 3.
Estimated coefficients for associations between predictor variables and internalizing and externalizing problems, and prosocial behavior
| Internalizing problems | Externalizing problems | Prosocial behavior | |||||||
|---|---|---|---|---|---|---|---|---|---|
| FE | AR-1 | AR-1 (4 Waves) | FE | AR-1 | AR-1 (4 Waves) | FE | AR-1 | AR-1 (4 Waves) | |
| Number of moves | − 0.107 (0.334) | − 0.069 (0.055) | − 0.167** (0.033) | − 0.547 (0.327) | − 0.291** (0.054) | − 0.302** (0.104) | 0.103 (0.259) | 0.015 (0.033) | 0.023 (0.067) |
| ACE count | 0.591 (0.453) | 0.152* (0.076) | 0.211** (0.075) | − 0.153 (0.288) | 0.238** (0.040) | 0.273** (0.099) | 0.120 (0.213) | 0.025 (0.027) | 0.009 (0.032) |
| Days in care | − 0.001* (0.0003) | − 0.001** (0.0001) | − 0.001** (0.0004) | − 0.001 (0.001) | 0.001** (0.0001) | 0.001** (0.0003) | |||
| Days with birth parents | 0.001 (0.0005) | 0.001 (0.0004) | 0.0003** (0.0001) | − 0.00005 (0.0001) | 0.0002 (0.0003) | 0.0002 (0.0002) | |||
| Child gender | − 0.689 (0.577) | − 0.998** (0.466) | − 0.876** (0.303) | − 0.872 (0.609) | 0.943** (0.095) | 0.940** (0.287) | |||
| Internalizing problemst−1 | 0.408* (0.196) | 0.435* (0.182) | |||||||
| Externalizing problemst−1 | 0.456** (0.156) | 0.443** (0.169) | |||||||
| Prosocial behaviort−1 | 0.389** (0.076) | 0.431** (0.148) | |||||||
| Intercept | 3.360** (1.010) | 5.100 (2.720) | 8.080** (1.340) | 6.750** (1.020) | 5.540** (1.610) | 7.790** (0.664) | 5.45** (0.781) | 2.090 (1.250) | 1.090 (1.030) |
| Observations | 318 | 222 | 177 | 318 | 222 | 177 | 318 | 222 | 177 |
| Adjusted R2 | 0.376 | 0.292 | 0.323 | 0.498 | 0.302 | 0.261 | 0.453 | 0.384 | 0.401 |
| F statistic | 2.910** | 8.020** | 7.460** | 4.140** | 8.350** | 5.790** | 3.620** | 11.600** | 10.100** |
FE = fixed-effects model, AR-1 = autoregressive model, AR-1 (4 waves) = autoregressive model for respondents who participated in all four waves. Coefficients are unstandardized to allow for direct interpretability. Standard errors are clustered by individual respondent and presented in brackets below the coefficients. Models are adjusted for respondent (adoptive parent) age at adoption, relationship status, education, income, employment, and siblings in household (see supplementary materials for all coefficients)
*p < 0.05, **p < 0.01