Skip to main content
. Author manuscript; available in PMC: 2008 Nov 25.
Published in final edited form as: Sociol Q. 2007 Apr 24;48(2):229–251. doi: 10.1111/j.1533-8525.2007.00077.x

TABLE 1.

Means and Standard Deviations for Dependent and Independent Variables, and Correlation Matrix (n = 78)

1 2 3 4 5 6 7 8 9 10
1. Δ Homicide 1.00
2. SE change .44 1.00
3. Ln inequality -.02 .74 1.00
4. Singles .47 .27 .14 1.00
5. Alcohol .46 .05 -.25 .22 1.00
6. Ln education -.26 -.13 .39 -.14 -.30 1.00
7. Ln polity -.34 -.10 -.05 -.45 -.26 .04 1.00
8. Urban -.15 -.25 .34 .14 -.13 .65 -.13 1.00
9. Ln males .10 .24 -.01 .29 .03 -.20 -.34 -.02 1.00
10. N. Caucasus -.28 .01 .09 -.30 -.41 .14 .26 -.15 -.37 1.00
Mean 13.6 1.4 6.0 15.6 28.7 27.0 69.3 39.0 15.3
SD 8.6 1.1 2.8 2.1 17.5 13.8 4.6 16.5 1.2

Notes: Means and standard deviations are for initial values, before any log transformations. The mean and standard deviation for homicide change are based on raw change scores. In the correlation matrix and model estimation, the change score is the difference between the observed rate for 2000 minus the predicted rate for 2000 based upon the 1991 rate (see text).