Table 7.
Full model results for negative binomial regression models predicting offending
| Father-child dyads (N = 180) | Mother-child dyads (N = 132) | |||
|---|---|---|---|---|
| Model 1 | Model 2 | Model 3 | Model 4 | |
| Coef. (SE) | Coef. (SE) | Coef. (SE) | Coef. (SE) | |
| Child arrest | 1.108** (0.425) | −1.894* (0.769) | 4.114*** (0.932) | 0.544 (0.508) |
| Parent arrest | 0.019 (0.554) | −0.016 (0.549) | −1.108*** (0.288) | −1.470*** (0.211) |
| Child arrest * parent arrest | – | 3.063** (1,032) | – | 4.465*** (0.659) |
| Prior offending | 0.001 (0.004) | 0.000 (0.004) | −0.020* (0.009) | −0.010(0.007) |
| G3 male | 1.313** (0.846) | 1.308** (0.369) | 0.154 (0.364) | 0.172 (0.320) |
| Black | −1.301 (0.846) | −1.306 (0.853) | 1.578*** (0.432) | 1.677*** (0.339) |
| Hispanic | −0.619 (0.896) | −0.576 (0.893) | 0.257 (0.547) | 0.903* (0.401) |
| Both parents | −0.346 (0.601) | −0.354 (0.599) | −0.233 (0.535) | −0.130 (0.538) |
| Financial assistance | 0.709 (0.514) | 0.738 (0.514) | −0.902* (0.243) | −0.714 (0.405) |
| Conventional values | −3.555*** (0,602) | −3.622*** (0,649) | −0.910*** (0.243) | −0.725* (0.286) |
| Gang member | −0.406 (0.466) | −0.395 (0.476) | 2.328 (1.281) | 2.072 (1.205) |
| Delinquent peers | −1.580*** (0.430) | −1.529*** (0.428) | 1.098 (0.678) | 1.247 (0.695) |
| Unstructured socializing | −0.303 (0.265) | −0.312 (0.272) | 0.080 (0.367) | −0.161 (0.316) |
| School commitment | 1,062* (0.528) | 1.064* (0.522) | −2.403*** (0.651) | −2.327*** (0.657) |
| Supervision | 0.091 (0.643) | 0.134 (0.662) | 1.464* (0.704) | 0.700 (0.641) |
| Attachment to mother | 0.907 (0.636) | 0.886 (0.634) | 0.343 (0.445) | 0.478 (0,463) |
| G2 infrequent with G3 | 1.591*** (0.406) | 1.645*** (0.394) | – | – |
| G2 offending history | 0.017(0.099) | 0.018 (0.100) | 0.328* (0.130) | 0.378** (0.112) |
| Community arrest rate | −0.248* (0.113) | −0.257* (0.110) | −0.065 (0.113) | −0.080 (0.100) |
| Missing covariate | 0.541 (0.361) | 0.532 (0.358) | −20.405*** (0.875) | −21.940 (0.853) |
Note. To account for the survey design, all standard errors are clustered by parent age at birth and robust standard errors are presented
Coef. = coefficient, SE = standard error, IRR = incidence rate ratio
*
p < 0.05,
**
p <0.01,
***
p < 0.001 (two-tailed test)