Skip to main content
. Author manuscript; available in PMC: 2010 Jul 1.
Published in final edited form as: AIDS Behav. 2008 Nov 22;13(4):700–715. doi: 10.1007/s10461-008-9488-8

Table 4.

Logistic set regression: UAI at the most recent encounter (N = 413)

Model −2 log L Overall model Change in −2 log L
Set 1: demographic and cultural set 400.28 X2(7) = 8.57
Set 2: psychological set 390.60 X2(9) = 18.26* 9.68*
Set 3: situational set 359.24 X2(14) = 49.62** 31.36**
Final model Coefficient Wald X2 Odds ratio 95% CI

Set 1: demographic and cultural characteristics
Intercept −1.13 1.28
Education level −0.15 .26 0.86 0.48–1.54
Income 0.14 2.04 1.15 0.95–1.39
Age −0.00 0.01 1.00 0.97–1.03
HIV positive status −0.01 0.00 0.99 0.48–2.05
Brazilian 0.01 0.00 1.01 0.50–2.03
Dominican 0.47 1.98 1.60 0.83–3.08
U.S. acculturation −0.15 0.47 0.86 0.56–1.32
Set 2: psychological characteristics
Self-efficacy for condom use −0.57 14.05** 0.57 0.42–0.76
Depression 0.03 0.02 1.03 0.67–1.59
Set 3: situational characteristics
Closeness with partner 0.31 6.17* 1.37 1.07–1.75
Concern about STIs −0.62 4.30* 0.54 0.30–0.97
Drug use in encounter 0.55 2.77 1.74 0.91–3.33
Sexual desire in encounter 0.21 1.57 1.23 0.89–1.70
Seroconcordance 0.70 5.58* 2.02 1.13–3.63
*

P<.05;

**

P<.01