Skip to main content
. Author manuscript; available in PMC: 2017 Dec 1.
Published in final edited form as: Am Sociol Rev. 2016 Oct 27;81(6):1276–1304. doi: 10.1177/0003122416673029

Table 2.

Growth Curve Models of Neighborhood Average Family Income: Panel Study of Income Dynamics, 1971-2011

Neighborhood Average Family Income
Model 1 Model 2 Model 3 Model 4
Fixed Effects
For Intercept (initial status)
Intercept 37.96***
(0.48)
41.94***
(0.51)
29.07***
(1.17)
21.71***
(1.42)
Black −8.42***
(0.63)
−4.76***
(0.70)
−3.24***
(0.71)
Neighborhood average family income in adolescence 0.25***
(0.03)
0.24***
(0.02)
For Linear Growth Rate
Intercept 0.97***
(0.04)
1.29***
(0.05)
0.82***
(0.15)
0.86***
(0.15)
Black −0.77***
(0.08)
−0.57***
(0.10)
−0.56***
(0.09)
Neighborhood average family income in adolescence 0.01**
(0.00)
0.01**
(0.00)
For Quadratic Growth Rate
Intercept −11.59***
(0.90)
−14.64***
(1.14)
−9.31***
(3.24)
−11.21***
(3.27)
Black 9.12***
(1.86)
7.05***
(2.14)
6.20**
(2.14)
Neighborhood average family income in adolescence −0.09*
(0.04)
−0.07
(0.05)
Control Variables
Married/cohabiting at t 1.06***
(0.24)
Number of children under 18 at t −0.42***
(0.09)
Years of education at t 0.40***
(0.07)
Family income at t-1 0.03***
(0.00)
Home ownership at t 0.38
(0.24)
Family income in adolescence 0.01***
(0.00)
Moved to different tract in same county between t-1 and t 0.63***
(0.20)
Moved to different county between t-1 and t 2.40***
(0.32)

Random Effects
Level-1: within-person 138.29*** 138.57*** 138.56*** 137.57***
Level-2: in initial status 186.23*** 137.08*** 121.68*** 121.13***
Level-2: in rate of change 0.86*** 0.82*** 0.83*** 0.80***

Goodness-of-fit
AIC 218548.7 218090.6 217906.9 217584.0
BIC 218630.8 218197.3 218038.3 217781.1

N of person-periods 27,237 27,237 27,237 27,237
N of persons 1,214 1,214 1,214 1,214

Note: Sample consists of black and white Panel Study of Income Dynamics respondents who established their own household between 1970 and 1979. Standard errors are in parentheses. Coefficients and standard errors for the quadratic growth rate are multiplied by 1000.

*

p < .05;

**

p < .01;

***

p < .001 (two-tailed tests).