Table 5.
Results of Conditional and Unconditional Ordinal Logistic Regression Models Predicting Court Dispositions From Race/Ethnicity and Overall Risk Level
| Predictor | Model 6.1 OR [95% CI] | Model 6.2 OR [95% CI] |
|---|---|---|
|
| ||
| Intercept 1 | 4.73 [3.69, 6.07] | 5.72 [4.73, 6.91] |
| Intercept 2 | 40 [31, 52] | 48 [39, 59] |
| Black | 0.60 [0.45, 0.78] | 0.49 [0.40, 0.59] |
| Hispanic | 0.65 [0.50, 0.86] | 0.53 [0.43, 0.64] |
| Overall risk | 0.30 [0.24, 0.38] | 0.23 [0.22, 0.25] |
| Black × Overall Risk | 0.76 [0.59, 0.98] | |
| Hispanic × Overall Risk | 0.75 [0.58, 0.97] | |
| Residual deviance (df) | 8,991 (11149) | 8,996 (11151) |
Note. Disposition was treated as an ordered-categorical variable with three categories: 1 = deferred adjudication, 2 = probation without placement, and 3 = probation with placement. Intercept 1 represents the log odds of being equal to 1 (deferred adjudication vs. probation without placement and probation with placement). Intercept 2 represents the log odds of being less than or equal to 2 (deferred adjudication and probation without placement vs. probation with placement). OR = odds ratio; CI = confidence interval.