Table 4.
Multilevel models of census tract percentage non-Hispanic black: 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 | 42.407*** (2.158) |
6.042*** (0.423) |
46.505*** (2.224) |
5.437*** (0.482) |
44.216*** (2.588) |
7.523*** (0.785) |
50.690*** (2.767) |
9.506*** (0.830) |
| Year 1991 (ref. = 1981) | −3.817*** (1.082) |
0.280 (0.366) |
−1.510 (1.121) |
0.580 (0.370) |
−4.387* (2.163) |
1.511** (0.547) |
||
| Year 2001 (ref. = 1991) | −3.611** (1.151) |
1.113** (0.376) |
−2.542** (1.159) |
1 493*** (0.378) |
−5.581** (1.953) |
1.526** (0.467) |
||
| Individual Characteristics | ||||||||
| Age 30–39 (ref. = 18–29) | −0.467 (1.190) |
0.194 (0.446) |
−0.402 (1.185) |
0.137 (0.443) |
||||
| Age 40–49 (ref. = 18–29) | 1.539 (1.410) |
0.170 (0.497) |
1.689 (1.404) |
0.159 (0.494) |
||||
| Age 50–59 (ref. = 18–29) | 6.379*** (1.676) |
0.402 (0.549) |
6.585*** (1.670) |
0.344 (0.543) |
||||
| Age >60 (ref. = 18–29) | 3.872* (1.708) |
0.735 (0.595) |
4.227* (1.699) |
0.589 (0.590) |
||||
| Female (1 = yes) | 0.339 (1.159) |
0.283 (0.428) |
0.220 (1.154) |
0.247 (0.424) |
||||
| Married (1 = yes) | −4.046** (1.273) |
−0.504 (0.426) |
−4.105** (1.267) |
−0.523 (0.422) |
||||
| Number of children | 0.482 (0.413) |
−0.136 (0.190) |
0.494 (0.411) |
−0.136 (0.188) |
||||
| Homeowner (1 = yes) | 1.600 (1.118) |
−1.323*** (0.377) |
1.261 (1.113) |
−1.339*** (0.373) |
||||
| Persons per room | 3.226** (1.216) |
1.177* (0.597) |
3.429** (1.212) |
1.083† (0.591) |
||||
| Education (in years) | −0.834*** (0.180) |
−0.111* (0.053) |
−0.882*** (0.179) |
−0.111* (0.052) |
||||
| Family income ($ 1,000s) | −0.122*** (0.020) |
−0.009** (0.003) |
−0.114*** (0.020) |
−0.009** (0.003) |
||||
| Employed (1 = yes) | −0.816 (1.100) |
−0.873† (0.473) |
−0.962 (1.095) |
−0.933* (0.470) |
||||
| Metropolitan Characteristics | ||||||||
| Population size (ln) | 7 694*** (1.736) |
−0.055 (0.359) |
||||||
| % non-Hispanic black | 1.244*** (0.176) |
0.471*** (0.039) |
||||||
| % Hispanic | −0.452** (0.159) |
−0.004 (0.033) |
||||||
| % Asian | −0.297 (0.301) |
−0.010 (0.066) |
||||||
| % in poverty | 0.257 (0.410) |
0.009 (0.098) |
||||||
| % of labor force in manufacturing | 0.022 (0.256) |
0.081 (0.054) |
||||||
| % of labor force in local, state, or federal government |
−0.674* (0.293) |
0.158* (0.070) |
||||||
| % new housing | −0.176 (0.126) |
0.073* (0.032) |
||||||
| % living in suburban area | 0.100 (0.089) |
0.023 (0.019) |
||||||
| Municipal fragmentation | 4.100 (7.940) |
−3.364* (1.610) |
||||||
| Variance Components | ||||||||
| Level 2: τ00 Random intercept (MSA) | 431.97 | 27.45 | 413.93 | 27.76 | 363.92 | 27.74 | 117.84 | 7.55 |
| Level 1: σ2 Residual variance | 788.46 | 108.24 | 780.92 | 107.91 | 747.21 | 106.02 | 744.50 | 105.39 |
| Pseudo-R2 Level 2 | .04 | .00 | .12 | .00 | .68 | .73 | ||
| Pseudo-R2 Level 1 | .01 | .00 | .04 | .02 | .00 | .01 | ||
| Likelihood Ratio Test χ2 (df) | 42.32*** (2) | 14.73*** (2) | 190.92*** (12) | 95.98*** (12) | 111.65*** (10) | 201.61*** (10) | ||
| Log-Likelihood (df) | −18,556 (3) | −18,542 (3) | −18,535 (5) | −18,535 (5) | −18,439 (17) | −18,487 (17) | −18,384 (27) | −18,386 (27) |
| AIC | 37,118 | 37,091 | 37,080 | 37,080 | 36,913 | 37,008 | 36,821 | 36,826 |
| BIC | 37,137 | 37,110 | 37,111 | 37,112 | 37,019 | 37,118 | 36,990 | 37,002 |
| 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