Skip to main content
. Author manuscript; available in PMC: 2011 Oct 21.
Published in final edited form as: AJS. 2006 Sep;112(2):567–612. doi: 10.1086/506418?origin=JSTOR-pdf

TABLE 3.

Regression of Change in % black from One Decade to the Next on Test and Control Variables, 1980–90

Model 1 (Reaggregated) Model 2 (Merged)
Intercept −26.5145* 36.5098*
Average exposure −2.9304* −1.8254
Highest %black in any adjacent tract .1854*** .1259***
%black1980 −.2134*** −.1183***
Median property value −.0001 −.0001
Median gross rent −.0016 −.0140
% owner occupied −.0996* .0531a
% families with children .1194** −.1167* a
% housing built before 1940 .0690 .0687
% housing built 1940–49 .1457* .0582
% housing built 1950–59 .1045 .0586
% housing built 1960–69 .0837 .0880
% same house as five years ago .0302** −.3001*** a
% high school graduates .0752 .0966
Median family income .0006 −.0002
Median family income squared .0001 .0001
ρ .1311* .1392*
Log likelihood −3,415 −2,681
N 607 504
a

This pair of variable coefficients (model 1 and model 2) are significantly different from one another. Spatial trend variables were utilized in the models but not reported here.

*

P < .05.

**

P < .01.

***

P < .001.