Table 2.
Mixed model analyses of the direct effects between air lead and assault rates for all six suburbs with time lags between 15 and 24 years
| Time Lag (number of cases with complete information) | F | df | p | Fixed effects (SE) | ω2 (%) |
|---|---|---|---|---|---|
| 15 years (n = 87) | .857 | 86.940 | .357 | 30.50 (32.951) | 0.46 |
| 16 years (n = 90) | .045 | 89.434 | .832 | 6.20 (29.191) | −0.09 |
| 17 years (n = 93) | 6.534 | 92.673 | .012 | 72.23 (28.256) | 5.54 |
| 18 years (n = 96) | 14.021 | 95.874 | .000 | 104.87 (28.007) | 11.66 |
| 19 years (n = 97) | 29.922 | 96.784 | .000 | 145.61 (26.619) | 22.88 |
| 20 years (n = 98) | 34.989 | 97.895 | .000 | 159.66 (26.993) | 25.60 |
| 21 years (n = 98) | 61.285 | 97.761 | .000 | 196.05 (25.044) | 38.38 |
| 22 years (n = 98) | 41.507 | 97.864 | .000 | 180.09 (27.954) | 28.70 |
| 23 years (n = 94) | 7.064 | 93.865 | .009 | 85.75 (32.264) | 5.13 |
| 24 years (n = 89) | 9.613 | 88.995 | .003 | 99.22 (32.003) | 7.72 |
F F-test, df degrees of freedom, p p-value, Fixed effects the estimated change in assaults per 100,000 population for a 1 μg/m3 increase in lead in air, SE standard error, ω 2 an estimate of the amount of variance accounted for by lead in air