. Author manuscript; available in PMC: 2010 Dec 1.

Published in final edited form as: Subst Use Misuse. 2008;43(2):203–221. doi: 10.1080/10826080701690607

Table 2.

Global clustering, OLS regression, and maximum likelihood for spatial error on overall HIV rates and risk-specific rates for New Orleans 1994–1999

	General HIV rate			MSM HIV rate			IDU HIV rate			HRH HIV rate
Variables	Base	OLS	ML	Base	OLS	ML	Base	OLS	ML	Base	OLS	ML
β(Constant)		−12.888	15.793		−2.214	14.438^*		−2.914^*	−2.946^*		−0.894	−0.483
On-site outlet density		3.765^***	2.52^***		1.884^***	1.097^***		0.322^***	0.354^***		−0.092	−0.160^*
Off-site outlet density		−1.066	−4.563^**		−1.153	−2.732^**		1.254^**	1.195^**		0.371	0.316
% HH poverty		0.024	0.165		−0.218	−0.141		0.112^**	0.119^***		0.074^*	0.067^*
Population density		0.000	0.000		0.000	0.000		0.000	0.000		0.000	0.000
% Vacant		2.566^***	1.833^***		1.200^***	0.713^***		0.041	0.019		0.164^*	0.178^**
λ(Lambda)		−	0.739^***		−	0.769^***		−	0.113		−	0.218^*
R-squared		0.582^***	0.729		0.457^***	0.667		0.393^***	0.397		0.202^***	0.224
Log likelihood		−814.771	−791.421		−728.78	−702.334		−528.948	−528.607		−508.452	−506.973
Akaike info. criterion		1641.54	1594.84		1469.56	1416.67		1069.9	1069.21		1028.9	1025.95
Moran’s I	0.577^***	0.287^***	−	0.521^***	0.301^***	−	0.254^***	0.036	−	0.21^***	0.08^*	−

p ≤ 0.05,

^**

p ≤ 0.01,

^***

p ≤ 0.001.