Table 2.
Multilevel models of census tract percentage non-Hispanic white: Black and white respondents from the 1981, 1991, and 2001 waves of the Panel Study of Income Dynamics (standard errors in parentheses)
| Model 1 |
Model 2 |
Model 3 |
Model 4 |
|||||
|---|---|---|---|---|---|---|---|---|
| Independent Variables | Black | White | Black | White | Black | White | Black | White |
| Intercept | 45.548*** (2.007) |
84.773*** (0.758) |
44.291*** (2.087) |
88.223*** (0.829) |
46.534*** (2.378) |
85.580*** (1.189) |
33.238*** (2.574) |
77.811*** (1.112) |
| Year 1991 (ref. = 1981) | 2.102* (0.980) |
−2.634*** (0.511) |
−0.500 (1.009) |
−3.366*** (0.513) |
5.010* (1.995) |
−2.319** (0.733) |
||
| Year 2001 (ref. = 1991) | −0.707 (1.042) |
−4 214*** (0.527) |
−1.798† (1.044) |
−4.865*** (0.525) |
4.334* (1.799) |
−1.912** (0.631) |
||
| Individual Characteristics | ||||||||
| Age 30–39 (ref. = 18–29) | 0.620 (1.072) |
−0.302 (0.619) |
0.734 (1.069) |
−0.112 (0.609) |
||||
| Age 40–49 (ref. = 18–29) | −1.441 (1.269) |
−0.636 (0.690) |
−1.433 (1.266) |
−0.547 (0.679) |
||||
| Age 50–59 (ref. = 18–29) | −6.646*** (1.509) |
−1.669* (0.762) |
−6.512*** (1.507) |
−1.323† (0.747) |
||||
| Age >60 (ref. = 18–29) | −3.993*** (1.538) |
−1.591† (0.825) |
−4.051** (1.533) |
−1.077 (0.811) |
||||
| Female (1 = yes) | −0.399 (1.043) |
−0.603 (0.593) |
−0.338 (1.041) |
−0.590 (0.583) |
||||
| Married (1 = yes) | 2.654* (1.146) |
0.560 (0.591) |
2.864* (1.143) |
0.511 (0.580) |
||||
| Number of children | −0.762* (0.372) |
0.245 (0.264) |
−0.813* (0.371) |
0.175 (0.259) |
||||
| Homeowner (1 = yes) | 0.659 (1.007) |
2.842*** (0.523) |
0.702 (1.004) |
2.768*** (0.512) |
||||
| Persons per room | −3.053** (1.094) |
−1.889** (0.828) |
−2.938** (1.094) |
−1.458† (0.813) |
||||
| Education (in years) | 0.934*** (0.162) |
0.333*** (0.073) |
0.961*** (0.162) |
0.333*** (0.071) |
||||
| Family income ($l,000s) | 0.110*** (0.018) |
0.018*** (0.004) |
0.107*** (0.018) |
0.019*** (0.004) |
||||
| Employed (1 = yes) | 1.101 (0.990) |
0.376 (0.656) |
1.109 (0.988) |
0.401 (0.646) |
||||
| Metropolitan Characteristics | ||||||||
| Population size (ln) | −6.936*** (1.641) |
−0.439 (0.460) |
||||||
| % non-Hispanic black | −1.063*** (0.165) |
−0.401*** (0.050) |
||||||
| % Hispanic | −0.474** (0.149) |
−0.709*** (0.043) |
||||||
| % Asian | −0.783** (0.278) |
−0.722*** (0.087) |
||||||
| % in poverty | −0.118 (0.378) |
0.083 (0.129) |
||||||
| % of labor force in manufacturing | 0.009 (0.237) |
−0.060 (0.071) |
||||||
| % of labor force in local, state, or federal government |
0.564** (0.272) |
−0.199* (0.092) |
||||||
| % new housing | 0.115 (0.116) |
−0.124** (0.042) |
||||||
| % living in suburban area | −0.064 (0.084) |
−0.034 (0.025) |
||||||
| Municipal fragmentation | −4.719 (7.483) |
7.065*** (2.097) |
||||||
| Viriance Components | ||||||||
| Level 2: τ00 Random intercept (MSA) | 380.88 | 104.24 | 375.76 | 107.29 | 322.24 | 105.23 | 112.72 | 10.90 |
| Level 1: σ2 Residual variance | 639.81 | 216.87 | 639.56 | 209.33 | 605.42 | 202.45 | 605.49 | 199.78 |
| Pseudo-R2 Level 2 | .01 | .00 | .14 | .02 | .65 | .90 | ||
| Pseudo-R2 Level 1 | .00 | .03 | .05 | .03 | .00 | .01 | ||
| Likelihood Ratio Test χ2 (df) | 4.76 (2) | 163.27*** (2) | 234.00*** (12) | 172.67*** (12) | 94.41*** (10) | 379.37*** (10) | ||
| Log-Likelihood (df) | −18,155 (3) | −20,294 (3) | −18,152 (5) | −20,213 (5) | −18,035 (17) | −20,126 (17) | −17,988 (27) | −19,937 (27) |
| AIC | 36,315 | 40,595 | 36,315 | 40,436 | 36,105 | 40,287 | 36,030 | 39,927 |
| BIC | 36,334 | 40,614 | 36,346 | 40,468 | 36,211 | 40,397 | 36,199 | 40,103 |
| N | 3,879 | 4,888 | 3,879 | 4,888 | 3,879 | 4,888 | 3,879 | 4,888 |
†
p < .10
*
p < .05
**
p < .01
***
p < .001